BR122022001849B1 - METHOD FOR ESTIMATING A FRACTION OF FETAL NUCLEIC ACID IN A TEST SAMPLE FROM A PREGNANT WOMAN - Google Patents

Abstract

Methods, processes, systems, machines and apparatus for the non-invasive evaluation of genetic variations are provided here.

Description

Related patent applications

[001] The patent application claims the benefit of US provisional patent application No. 61/838,048 filed on June 21, 2013, entitled METHODS AND PROCESSES FOR NON-INVASIVE ASSESSMENT OF GENETIC VARIATIONS, naming Sung Kim K. et al., as inventors, and designated by the attorney-in-fact registration No. SEQ-6071-PV. The entire contents of the foregoing application are hereby incorporated by reference, including all text, tables and drawings.

Field

[002]Technology provided here refers in part to methods, processes, machines and apparatus for non-invasive evaluation of genetic variations.

Fundamentals

[003] Genetic information from living organisms (eg, animals, plants, and microorganisms) and other forms of replicating genetic information (eg, viruses) is encoded in deoxyribonucleic acid (DNA) or ribonucleic acid (RNA). Genetic information is a sequence of nucleotides or modified nucleotides that represent the primary structure of chemical or hypothetical nucleic acids. In humans, the complete genome contains about 30,000 genes located on twenty-four (24) chromosomes (see The Human Genome, T. Strachan, BIOS Scientific Publishers, 1992). Each gene encodes a specific protein that, upon expression through transcription and translation, fulfills a specific biochemical function within a living cell.

[004]Many medical conditions are caused by one or more genetic variations. Certain genetic variations cause medical conditions including, for example, hemophilia, thalassemia, Duchenne muscular dystrophy (DMD), Huntington's disease (HD), Alzheimer's disease, and cystic fibrosis (CF) (Human Genome Mutations, D.N. Cooper and M. Krawczak, BIOS Publishers, 1993). Such genetic diseases can result from an addition, substitution or deletion of a single nucleotide in the DNA of a particular gene. Certain birth defects are caused by a chromosomal abnormality, also referred to as an aneuploidy, such as Trisomy 21 (Down Syndrome), Trisomy 13 (Patau Syndrome), Trisomy 18 (Edward Syndrome), Monosomy X (Turner Syndrome) and certain aneuploidies of the sex chromosome, such as klinefelter syndrome (XXY), for example. Another genetic variation is the fetal gender which can often be determined based on the X and Y sex chromosomes. Some genetic variations can predispose an individual to, or cause, any of a number of diseases, such as, for example, diabetes, arteriosclerosis, obesity, various autoimmune diseases, and cancer (eg, colorectal, breast, ovarian, lung).

[005]Identification of one or more genetic variations or variances may lead to a diagnosis of, or predisposition towards the determination of, a particular medical condition. Identification of a genetic variation may result in facilitating a medical decision and/or employing a useful medical procedure. In certain embodiments, identifying one or more genetic variations or variances involves analyzing cell-free DNA. Cell-free DNA (CF-DNA) is composed of fragments of DNA that originate from cell death and circulate in peripheral blood. High concentrations of CF-DNA can be indicative of certain clinical conditions, such as cancer, trauma, burns, myocardial infarction, stroke, septicemia, infection and other diseases. Furthermore, cell-free fetal DNA (CFF-DNA) can be detected in maternal bloodstream and used for various non-invasive prenatal diagnoses.

[006] The presence of fetal nucleic acid in maternal plasma allows non-invasive prenatal diagnosis through the analysis of a sample of maternal blood. For example, quantitative fetal DNA changes in maternal plasma may be associated with a number of disorders associated with pregnancy, including preeclampsia, preterm labor, antepartum hemorrhage, invasive placentation, fetal Down syndrome, and other aneuploidies. fetal chromosomes. Thus, analysis of fetal nucleic acid in maternal plasma may be a useful mechanism for monitoring fetal-maternal well-being.

summary

[007] Provided herein, in certain aspects, is a method for estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising (a) obtaining counts of fragments (reads) of sequences mapped to portions of a reference genome, whose sequence reads are circulating cell-free nucleic acid reads from a test sample from a pregnant woman, (b) weight, using a microprocessor, (i) fragment counts ( reads) of sequences mapped to each chunk, or (ii) another chunk-specific parameter, to a chunk-specific fraction of fetal nucleic acid according to a weighting factor independently associated with each chunk, thereby providing estimates of the fetal fraction portion-specific weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of several samples, and (ii) counts of sequence reads mapped to each portion, or other portion-specific parameter, for the various samples, and (c) estimating a fetal nucleic acid fraction for the test sample based on the portion-specific fetal fraction estimates .

[008] Also provided herein is a method for estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising (a) obtaining counts of fragments (reads) of sequences mapped to portions of a reference genome, whose sequence reads are circulating cell-free nucleic acid reads from a test sample from a pregnant woman, (b)(i) adjust, using a microprocessor, the sequence read counts mapped to each serving according to an independently assigned weighting factor for each serving, thereby providing adjusted counts for the servings, or (b)(ii) selecting, using a microprocessor, a subset of servings, thereby providing a subset of counts, where the adjustment in (b)(i) or the selection in (b)(ii) is in accordance with the portions to which an increased amount of fetal nucleic acid reads is mapped, and (c) estimating a fetal nucleic acid fraction for the test sample based on the adjusted counts or the subset of counts.

[009] A method is also provided here for increasing the accuracy of estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising obtaining counts of fragments (reads) of sequences mapped to portions of a genome of reference, whose sequence reads are circulating cell-free nucleic acid fragments (reads) of a test sample from a pregnant woman, where at least a subset of the counts obtained is derived from a region of the genome that contributes with a greater number of counts derived from fetal nucleic acid relative to the region of fetal nucleic acid counts relative to the total counts of another region of the genome.

[010] Also provided herein is a system, machine or apparatus comprising one or more microprocessors and memory, wherein the memory comprises instructions executable by one or more microprocessors and further instructions executable by one or more microprocessors are configured to (a) access fragments Nucleotide sequence reads mapped to portions of a reference genome, the sequence reads of which are circulating cell-free nucleic acid reads from a test sample from a pregnant woman, (b) weight ( i) the sequence reads counts mapped to each portion, or (ii) another portion-specific parameter, to a specific fraction of the portion's fetal nucleic acid according to a weighting factor independently associated with each portion providing, thus estimates of the portion-specific fetal fraction according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of the various samples, and (ii) counts of sequence reads mapped to each portion, or other portion-specific parameter, for the various samples, and (c) estimating a fraction of fetal nucleic acid for the test sample based on on estimates of portion-specific fetal fraction.

[011] Also provided herein is a machine comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors and which memory comprises fragments (reads) of nucleotide sequences mapped to portions of a reference genome, which sequence reads are cell-free nucleic acid reads circulating a test sample from a pregnant woman, and which instructions executable by one or more microprocessors are configured to (a) weight, using a microprocessor, (i) counts of sequence reads mapped to each chunk, or (ii) another chunk-specific parameter, to a chunk-specific fraction of fetal nucleic acid according to a weighting factor independently associated with each chunk providing , therefore, estimates of portion-specific fetal fraction according to weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of several samples, and (ii) counts of sequence reads mapped to each chunk, or other chunk-specific parameter, for the various samples, and (b) estimating a fraction of fetal nucleic acid for the test sample based on the estimates portion-specific fetal fraction.

[012] A non-transient computer-readable storage medium with an executable program stored therein is also provided here, where the program instructs a microprocessor to perform the following: (a) access the fragments (reads) of the nucleotide sequence mapped to portions of a reference genome, the sequence reads of which are circulating cell-free nucleic acid reads from a test sample from a pregnant woman, (b) weight, using a microprocessor, (i) the counts of sequence reads mapped to each chunk, or (ii) another chunk-specific parameter, to a chunk-specific fraction of fetal nucleic acid according to a weighting factor associated with each chunk independently, thereby providing estimates of the portion-specific fetal fraction according to the weighting factors, wherein each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fetal nucleic acid fraction for each of several samples, and (ii) counts of sequence reads mapped to each portion, or other portion-specific parameter, for the various samples, and (c) estimating a fraction of fetal nucleic acid for the test sample based on the fraction estimates portion-specific fetus.

[013] Certain aspects of the technology are further described in the following description, examples, claims and drawings.

Brief description of the drawings

[014] The drawings illustrate technology modalities and are not limiting. For clarity and ease of illustration, the drawings are not drawn to scale and, in some cases, various aspects may be shown exaggerated or enlarged to facilitate understanding of particular modalities.

[015] Figure 1 shows a pairwise comparison of FRS (left vertical axis, top histogram) and the number of exons per 50 kb chunk (bin) (right vertical axis, bottom histogram) for chromosome 13. The chunks are shown in the bottom, horizontal X-axis.

[016] Figure 2 shows a pairwise comparison of FRS (left vertical axis, upper histogram) and GC content per 50 kb portion (bin) (right vertical axis, lower histogram) for chromosome 13. The portions are shown in bottom, horizontal X-axis.

[017] Figure 3 shows a pairwise comparison of the number of exons per 50 kb chunk (left vertical axis, top histogram) and the GC content per 50 kb chunk (right vertical axis, bottom histogram) for chromosome 13. portions are shown at the bottom, horizontal X-axis.

[018] Figure 4 shows a paired comparison of FRS (left vertical axis, top histogram) and the number of exons per 50 kb chunk (right vertical axis, bottom histogram) for chromosome 18. The chunks are shown on the bottom, bottom axis X horizontal.

[019] Figure 5 shows a pairwise comparison of FRS (left vertical axis, top histogram) and GC content per 50 kb portion (right vertical axis, bottom histogram) for chromosome 18. Portions are shown at the bottom, horizontal X axis.

[020] Figure 6 shows a pairwise comparison of the number of exons per 50 kb chunk (left vertical axis, top histogram) and the GC content per 50 kb chunk (right vertical axis, bottom histogram) for chromosome 18. portions are shown at the bottom, horizontal X-axis.

[021] Figure 7 shows a pairwise comparison of FRS (left vertical axis, top histogram) and the number of exons per 50 kb chunk (right vertical axis, bottom histogram) for chromosome 21. The chunks are shown at the bottom, horizontal X axis.

[022] Figure 8 shows a pairwise comparison of FRS (left vertical axis, top histogram) and GC content per 50 kb portion (right vertical axis, bottom histogram) for chromosome 21. Portions are shown at the bottom, horizontal X axis.

[023] Figure 9 shows a pairwise comparison of the number of exons per 50 kb chunk (left vertical axis, top histogram) and the GC content per 50 kb chunk (right vertical axis, bottom histogram) for chromosome 21. portions are shown at the bottom, horizontal X-axis.

[024] Figure 10 shows PERUN PAD with LOESS indices (X axis) versus PERUN PAD with LOESS Z indices based on "non-enriched fetal" portions (Y axis) for chromosome 21. The four quadrants represent agreement and disagreement. The bands in the quadrant are drawn at Z = 3. The lower left and upper right quadrants are separated by a gray diagonal dashed band. The dotted-dashed band is a regression band for non-T21 samples only. The dotted band is a regression band for T21 samples based on high FRS portions.

[025] Figure 11 shows PERUN PAD with LOESS Z indices (X axis) versus PERUN PAD with LOESS Z indices based on "enriched fetal" portions (ie, portions with high FRS) (Y axis) for chromosome 21. The four quadrants represent agreement and disagreement. The bands in the quadrant are drawn at Z = 3. The lower left and upper right quadrants are separated by a gray diagonal dashed band. The dotted-dashed band is a regression band for non-T21 samples only. The dotted band is a regression band for T21 samples based on high FRS portions.

[026] Figure 12 shows a method for determining the length of the nucleic acid fragment that includes the steps of 1) probe hybridization (P; dotted line) to fragment (solid line), 2) probe cutting, and 3) Probe length measurement. Fragment size determination is shown for a fetal-derived fragment (F) and a maternally-derived fragment (M).

[027] Figure 13 shows a distribution of fragment lengths from three different library preparation methods. They include enzymatic with automated bead cleaning, enzymatic without automated bead cleaning, and TRUSEQ with automated bead cleaning. Vertical lines represent 143 bases and 166 base fragment sizes.

[028] Figure 14 shows representation of chromosome 13 without a fragment size filter.

[029] Figure 15 shows the representation of chromosome 13 with a fragment size filter of 150 bases.

[030] Figure 16 shows representation of chromosome 18 without a fragment size filter.

[031] Figure 17 shows the representation of chromosome 18 with a fragment size filter of 150 bases.

[032] Figure 18 shows representation of chromosome 21 without a fragment size filter.

[033] Figure 19 shows the representation of chromosome 21 with a fragment size filter of 150 bases.

[034] Figure 20 shows the representation of chromosome 13 (PERUN PAD with LOESS) with filters of variable fragment size.

[035] Figure 21 shows the representation of chromosome 18 (PERUN PAD with LOESS) with filters of variable fragment size.

[036] Figure 22 shows the representation of chromosome 21 (PERUN PAD with LOESS) with filters of variable fragment size.

[037]Figure 23 shows a table that presents a description of the data used for certain analyses.

[038] Figure 24 shows an illustrative embodiment of a system in which certain modalities of the technology can be implemented.

[039] Figure 25A shows a relationship between the average FRS of a subset of portions of Chr21 (X axis) to the Z indices of normalized PERUN counts (Y axis) for the same subset of portions for samples obtained from pregnant women with a fetus with trisomy 21 (indicated by an asterisk) or euploid fetus (indicated by circles). Each portion in the subset of portions selected for Figure 25A has a FRS greater than the average FRS determined for all parts of chromosome 21 from which counts were obtained. Figure 25B shows a ratio of fetal FQA fraction estimates (X-axis) vs. Z indices of PERUN normalized counts (Y-axis) for Chr21 obtained from pregnant women with a fetus with trisomy 21 (indicated by an asterisk) or euploid fetus (indicated by circles) for all portions of Chr21 from which counts were obtained.

[040] Figure 26 shows a ratio of GC content per read (X axis) to the cumulative distribution function based on read length (CDF, Y axis) for fragments (reads) of the indicated range of fragment lengths (shown in the lower right inset) for chromosome 21.

[041] Figure 27 shows a distribution of PERUN intercepts (X axis) divided into quantiles (high, medium high, medium low and low) according to FRS per portion (bin).

[042] Figure 28 shows a distribution of PERUN Max Cross validation errors (X axis) divided into quantiles (High, Medium High, Medium Low and Low) according to FRS per portion (bin).

[043] Figure 29 shows a correlation (r = 0.81, RMedSE = 1.5) of the predicted fetal fraction percentages for 19,312 test samples from a BFF model based on 6000 training samples (X axis) in comparison to determined fetal fraction percentages of Y chromosome levels (ChrFF, Y-axis).

[044] Figure 30 shows the relative prediction error (X axis) for portion (bin) (that is, portions) with high fetal fraction content (distribution shown on the left) and low fetal fraction content (distribution shown on the right) based on FRS. Portions (bins) with high fetal content have better performance and lower error. Predictive indices are based on an elastic network regression process, with bootstrapping used to obtain density profiles.

[045] Figure 31 shows four distributions of model coefficients (X axis) determined using an elastic network regression procedure on subsets of portions (bins) separated according to the fetal fraction content (for example, low, medium-low , medium-high, high). Portions (bins) with higher fetal fraction content tend to produce higher coefficients (positive or negative).

[046]Figure 32 shows two distributions for fetal fraction estimates (X axis) determined using a BFF method for male and female test samples. The two distributions substantially overlap. Male and female fetuses show no difference in fetal fraction distribution (KS test P = 0.49).

Detailed Description

[047] Methods are provided here for analyzing polynucleotides in a nucleic acid mixture that include, for example, methods for determining the presence or absence of a genetic variation. Evaluation of a genetic variation, such as, for example, a fetal aneuploidy, from a maternal sample typically involves sequencing the nucleic acid present in the sample, mapping the fragments (reads) of sequences in certain regions of the genome, quantification of the fragments (reads) of the sequence for the sample, and quantitation analysis. Such methods often directly analyze the nucleic acid in the sample and obtain nucleotide sequence reads for all or substantially all of the nucleic acid in the sample, which can be expensive and may generate superfluous and/or irrelevant data. Certain length-based and/or sequence-based separation approaches combined with certain length-based and/or sequence-based analysis, however, can generate specific information about specific genomic regions, such as, for example, a specific chromosome, and in some cases, they can differentiate nucleic acid fragment origins, such as maternal versus fetal origin. Certain methods may include the use of sequencing methods, enrichment techniques, and length-based analysis. Certain methods described herein, in some embodiments, can be performed without determining nucleotide sequences of nucleic acid fragments. Provided herein are methods for analyzing polynucleotides in a nucleic acid mixture (eg, determining the presence or absence of a fetal aneuploidy) using a combination of length-based and/or sequence-based analysis and separation approaches.

[048] Methods, processes and machines useful for identifying a genetic variation are also provided. Identifying a genetic variation sometimes comprises detecting a copy number variation and/or sometimes comprises adjusting a level comprising a copy number variation. In some embodiments, a level is set providing an identification of one or more genetic variations or variances with a reduced probability of a false negative or false positive diagnosis. In some embodiments, identification of a genetic variation by a method described herein can lead to a diagnosis of, or predisposition towards the determination of, a particular medical condition. Identifying a genetic variation can result in facilitating a medical decision and/or employing a useful medical procedure.

[049] Systems, machines and modules are also provided here that, in some ways, perform the methods described here.

Samples

[050] Methods and compositions for nucleic acid analysis are provided here. In some embodiments, nucleic acid fragments are analyzed in a mixture of nucleic acid fragments. A nucleic acid mixture can comprise two or more species of nucleic acid fragments having different nucleotide sequences, different fragment lengths, different origins (e.g., genomic origins, fetal vs. maternal origin, cell or tissue origin, origins of sample, subject origins, and the like), or combinations thereof.

[051] Nucleic acid or a nucleic acid mixture used in the methods and apparatus described herein is often isolated from a sample obtained from a subject. A subject can be any living organism or non-living organism, including but not limited to a human, a non-human animal, a plant, a bacterium, a fungus or a protist. Any human or non-human animal may be selected, including but not limited to mammals, reptiles, birds, amphibians, fish, ungulates, ruminants, bovine (e.g. cattle), equine (e.g. horse), goat and sheep ( eg sheep, goat), pigs (e.g. pig), camelids (e.g. camel, llama, alpaca), monkeys (e.g. gorilla, chimpanzee), ursids (e.g. bear), birds, dog, cat, mouse, rat, fish, dolphin, whale and shark. A subject can be a male or female (eg female, a pregnant woman). A subject can be any age (for example, an embryo, fetus, infant, child, adult).

[052] The nucleic acid can be isolated from any type of sample or suitable biological species (for example, a test sample). A sample or test sample can be any species that is isolated or obtained from a subject or part thereof (eg, a human subject, a pregnant woman, a fetus). Non-limiting examples of species include fluid or tissue from a subject, including, without limitation, blood or a blood-derived product (e.g., serum, plasma, or the like), umbilical cord blood, chorionic villi, amniotic fluid, cerebrospinal fluid , spinal fluid, lavage fluid (e.g., bronchoalveolar, gastric, peritoneal, ductal, ear, arthroscopic), biopsy sample (e.g., from pre-implantation embryo), cellocentesis sample, cells (blood cells , placental cells, embryonic or fetal cells, fetal nucleated cells or fetal cell remnants) or parts thereof (e.g., mitochondrial, nucleus, extracts, or the like), female reproductive tract washes, urine, feces, sputum, saliva , nasal mucus, prostate fluid, lavage, semen, lymphatic fluid, bile, tears, perspiration, breast milk, breast fluid, or the like or combinations thereof. In some embodiments, a biological sample is a cervical swab from a subject. In some embodiments, a biological sample can be blood and sometimes plasma or serum. The term "blood" as used herein refers to a blood sample or preparation from a pregnant woman or a woman being tested for pregnancy. The term includes whole blood, blood products, or any fraction of blood, such as serum, plasma, buffy coat, or the like, as conventionally defined. Blood or its fractions often comprise nucleosomes (eg, maternal and/or fetal nucleosomes). Nucleosomes comprise nucleic acids and are sometimes cell-free or intracellular. Blood also comprises buffy coats. Buffy coats are sometimes isolated using a Ficoll gradient. Buffy coats can comprise white blood cells (e.g., leukocytes, T-cells, B-cells, platelets, and the like). In certain embodiments buffy coats comprise maternal and/or fetal nucleic acid. Blood plasma refers to the fraction of whole blood resulting from centrifugation of blood treated with anticoagulants. Blood serum refers to the watery portion of fluid remaining after a blood sample has clotted. Fluid or tissue samples are often collected according to standard protocols that hospitals or clinics commonly follow. For blood, an adequate amount of peripheral blood (eg, between 3 to 40 milliliters) is often collected and can be stored according to standard procedures before or after preparation. A sample of fluid or tissue from which the nucleic acid is extracted can be acellular (e.g., cell-free). In some embodiments, a fluid or tissue sample may contain cellular elements or cellular debris. In some embodiments cancer cells or fetal cells may be included in the sample.

[053] A sample is often heterogeneous, which means that more than one type of nucleic acid species is present in the sample. For example, heterogeneous nucleic acid may include, but is not limited to, (i) maternally derived and fetally derived nucleic acid, (ii) cancer and non-cancer nucleic acid, (iii) host and pathogen nucleic acid, and more generally , (IV) wild-type and mutated nucleic acid. A sample may be heterogeneous because more than one cell type is present, such as a fetal cell and a maternal cell, a cancer and non-cancer cell, or a host and pathogenic cell. In some embodiments, a minority of nucleic acid species and a majority of nucleic acid species are present.

[054] For prenatal applications of the technology described herein, a fluid or tissue sample may be collected from a woman with an appropriate gestational age for testing, or from a woman being tested for possible pregnancy. Appropriate gestational age may vary according to the prenatal test to be performed. In certain embodiments, a pregnant woman is sometimes in the first trimester of pregnancy, sometimes in the second trimester of pregnancy, or sometimes in the third trimester of pregnancy. In certain embodiments, a fluid or tissue is collected from a pregnant woman between about 1 to about 45 weeks of fetal gestation (e.g., at 1-4, 4-8, 812, 12-16, 16-20, 20 - 24, 24-28, 28-32, 32-36, 36-40 or 4044 weeks of fetal gestation), and sometimes between about 5 to about 28 weeks of fetal gestation (e.g. at 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26 or 27 weeks of fetal gestation). In certain embodiments a fluid or tissue sample is collected from a pregnant woman during or immediately after (eg, 0 to 72 hours after) delivery (eg, vaginal or non-vaginal birth (eg, surgical delivery)).

Acquisition of blood samples and DNA extraction

[055] Methods here often include separating, enriching and analyzing fetal DNA found in maternal blood, as a non-invasive means to detect the presence or absence of a maternal and/or fetal genetic variation and/or to monitor the health of a fetus and/or a pregnant woman during and sometimes after pregnancy. As such, the first steps of practicing certain methods here often include obtaining a blood sample from a pregnant woman and extracting DNA from a sample.

Acquisition of blood samples

[056] A blood sample may be obtained from a pregnant woman of an appropriate gestational age for testing using a method of present technology. The appropriate gestational age may vary depending on the disorder being tested as discussed below. Drawing blood from a woman is often performed according to the standard protocol that hospitals or clinics usually follow. An adequate amount of peripheral blood, for example typically between 5 to 50 mL, is often collected and can be stored according to standard procedure prior to further preparation. Blood samples may be collected, stored, or transported in a manner that minimizes degradation or the quality of the nucleic acid present in the sample.

Preparation of blood samples

[057]An analysis of fetal DNA found in maternal blood can be performed using, for example, whole blood, serum or plasma. Methods for preparing serum or plasma from maternal blood are known. For example, blood from a pregnant woman can be placed in a tube containing EDTA or a specialized commercial product such as Vacutainer SST (Becton Dickinson, Franklin Lakes, NJ) to prevent blood clotting, and then the plasma can be be obtained from whole blood by means of centrifugation. Serum can be obtained with or without centrifugation following blood clotting. If centrifugation is used then it is typically, but not exclusively, conducted at a suitable speed, eg 1500-3000g times. Plasma or serum may undergo additional centrifugation steps before being transferred to a new tube for DNA extraction.

[058] In addition to the acellular portion of whole blood, DNA can also be recovered from the cellular fraction, enriched in the buffy coat, which can be obtained by following centrifugation of a whole blood sample from a woman and removing the plasma .

DNA extraction

[059] There are several known methods for extracting DNA from a biological sample including blood. General DNA preparation methods (eg described by Sambrook and Russell, Molecular Cloning: A Laboratory Manual 3rd ed, 2001) can be followed; various commercially available reagents or kits, such as Qiagen's QIAamp Circulating Nucleic Acid Kit, QIAamp DNA Mini Kit or QiAmp Blood DNA Mini Kit (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega , Madison, Wis.), and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, NJ), can also be used to obtain DNA from a blood sample from a pregnant woman. Combinations of more than one of these methods may also be used.

[060] In some embodiments, the sample may first be enriched or relatively enriched as fetal nucleic acid by one or more methods. For example, discrimination of maternal and fetal DNA can be performed using the compositions and processes of the present technology alone or in combination with other discrimination factors. Examples of these factors include, but are not limited to, single nucleotide differences between X and Y chromosomes, specific Y chromosome sequences, polymorphisms located elsewhere in the genome, size differences between maternal and fetal DNA, and differences in methylation pattern between maternal and fetal tissues.

[061] Other methods for enriching an extract of a particular species of nucleic acid are described in PCT patent application number PCT/US07/69991, filed on May 30, 2007, PCT patent application number PCT/US2007/071.232 , filed June 15, 2007, US provisional application numbers 60/968,876 and 60/968,878 (filed by the applicant), (PCT Patent Application number PCT/EP05/012,707, filed November 20, 2005), which are all incorporated herein by reference. In certain embodiments, maternal nucleic acid is selectively removed (either partially, substantially, completely or nearly completely) from the sample.

[062] The terms "nucleic acid" and "nucleic acid molecule" may be used interchangeably throughout the invention. The terms refer to nucleic acids of any composition, such as DNA (e.g., complementary DNA (AcDNA), genomic DNA (gDNA) and the like), RNA (e.g., messenger RNA (mRNA), short inhibitory RNA ( siRNA), ribosomal RNA (rRNA), tRNA, microRNA, RNA highly expressed by the fetus or placenta, and the like), and/or DNA or RNA analogs (e.g., containing base analogs, sugar analogs, and/or a structure non-native and similar), RNA/DNA hybrids and polyamide nucleic acids (PNAs), all of which may be in single-stranded or double-stranded form, and, unless otherwise limited, may encompass the known analogues of natural nucleotides that can function in a similar way to naturally occurring nucleotides. A nucleic acid may be, or may be from, a plasmid, phage, autonomously replicating sequence (ARS), centromere, artificial chromosome, chromosome, or other nucleic acid capable of replicating or being replicated in vitro or in a host cell, a cell, a cell nucleus, or cytoplasm of a cell, in certain embodiments. A nucleic acid template in some embodiments may be a single chromosome (for example, a nucleic acid sample may be from a chromosome of a sample obtained from a diploid organism). Unless specifically limited, the term encompasses nucleic acids containing known analogues of natural nucleotides that have similar binding properties as the reference nucleic acid and are metabolized similarly to naturally occurring nucleotides. Unless otherwise noted, a particular nucleic acid sequence also implicitly encompasses conservatively modified variants thereof (e.g., degenerate codon substitutions), alleles, orthologs, single nucleotide polymorphisms (SNPs), and complementary sequences, as the explicitly specified sequence. Specifically, degenerate codon substitutions can be obtained by generating sequences in which the third position of one or more selected codons (or all) is replaced by deoxyinosine and/or mixed base residues. The term nucleic acid is used interchangeably with the locus, gene, cDNA, and mRNA encoded by a gene. The term may also include, as equivalents, derivatives, variants and analogues of RNA or DNA synthesized from nucleotide analogues, single-stranded polynucleotides ("sense" or "foresense", "+" strand or "-" strand, frame readable "forward" or frame readable "reverse") and double-stranded. The term "gene" means the segment of DNA involved in the production of a polypeptide chain; it includes regions that precede and follow the coding region (leader and trailer) involved in the transcription/translation of the gene product and the regulation of transcription/translation, as well as intervening sequences (introns) between individual coding segments (exons). Deoxyribonucleotides include deoxyadenosine, deoxycytidine, deoxyguanosine, and deoxythymidine. For RNA, the base cytosine is replaced by uracil A nucleic acid template can be prepared using a nucleic acid obtained from a subject as a template.

Nucleic acid isolation and processing

[063] The nucleic acid can be derived from one or more sources (e.g., cells, serum, plasma, buffy coat, lymphatic fluid, skin, dirt, and the like) by methods known in the art. Any suitable method can be used to isolate, extract and/or purify DNA from a biological sample (e.g., from blood or a blood-derived product), non-limiting examples of which include DNA preparation methods ( for example, described by Sambrook and Russell, Molecular Cloning: A Laboratory Manual 3d ed, 2001), various commercially available reagents or kits, such as Qiagen's QIAamp Circulating Nucleic Acid Kit, QIAamp DNA Mini Kit, or QIAamp DNA Mini Kit. QiAmp blood (Qiagen, Hilden, Germany), GenomicPrep™ Blood DNA Isolation Kit (Promega, Madison, Wis.), and GFX™ Genomic Blood DNA Purification Kit (Amersham, Piscataway, NJ), the like or combinations thereof.

[064] Cell lysis procedures and reagents are known in the art and can generally be performed by chemical lysis methods (for example, detergent, hypotonic solutions, enzymatic procedures, and the like, or combination thereof), physical (for example , french press, sonication and similar) or electrolytic. Any suitable lysis procedure can be used. For example, chemical methods commonly employ lysing agents to disrupt cells and extract the cell's nucleic acids followed by treatment with chaotropic salts. Physical methods such as freeze/thaw followed by milling, the use of cell presses and the like are also useful. High salt lysis procedures are also commonly used. For example, an alkaline lysis procedure can be used. The latter procedure traditionally incorporates the use of phenol-chloroform solutions, and an alternative phenol-chloroform-free procedure involving three solutions may be used. In the latter processes, a solution may contain 15 mM Tris, pH 8.0; 10 mM EDTA and 100 ug/ml RNAase A; a second solution may contain 0.2N NaOH and 1% SDS; and a third solution may contain 3M KOAc, pH 5.5. These procedures can be found in Current Protocols in Molecular Biology, John Wiley & Sons, New York, 6.3.16.3.6 (1989), incorporated herein in its entirety.

[065] The nucleic acid can be isolated at a different time point compared to another nucleic acid, where each of the samples is from the same or different source. A nucleic acid can be from a nucleic acid library, such as a cDNA or RNA library, for example. A nucleic acid can be a result of nucleic acid purification or isolation and/or amplification of nucleic acid molecules from the sample. The nucleic acid provided for processes described herein may contain the nucleic acid from one sample or from two or more samples (e.g., from 1 or more, 2 or more, 3 or more, 4 or more, 5 or more, 6 or more , 7 or more, 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, 14 or more, 15 or more, 16 or more, 17 or more, 18 or more, 19 or more, or 20 or more samples).

[066] Nucleic acids may include extracellular nucleic acid in certain embodiments. The term "extracellular nucleic acid", as used herein, can refer to nucleic acid isolated from a source having substantially no cells and is also referred to as "cell-free" nucleic acid, "circulating cell-free nucleic acid" (for (e.g., CCF fragments) and/or "circulating cell-free nucleic acid". Extracellular nucleic acid may be present in and obtained from blood (eg, from the blood of a pregnant woman). Extracellular nucleic acid often does not include detectable cells and may contain cellular elements or cellular debris. Non-limiting examples of acellular sources of extracellular nucleic acid are blood, blood plasma, blood serum and urine. As used herein, the term "obtaining nucleic acid from the cell-free circulating sample" includes obtaining a sample directly (e.g., collecting a sample, e.g., a test sample) or obtaining a sample from one that has collected a sample. Without being bound by theory, extracellular nucleic acid can be a product of cellular breakdown and cellular apoptosis, which provides a basis for extracellular nucleic acid often having a range of lengths across a spectrum (e.g., a "ladder") .

[067] Extracellular nucleic acid may include different species of nucleic acid, and therefore is referred to herein as "heterogeneous", in certain embodiments. For example, blood serum or plasma from a person who has cancer may include nucleic acid from cancer cells and nucleic acid from non-cancer cells. In another example, the blood serum or plasma of a pregnant woman can include maternal nucleic acid and fetal nucleic acid. In some cases, the fetal nucleic acid sometimes is from about 5% to about 50% of the total nucleic acid (e.g. about 4, 5, 6, 7, 8, 9, 10, 11, 12 , 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 , 38,39,40,41,42,43, 44, 45, 46, 47, 48, or 49% of the total nucleic acid is fetal nucleic acid). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 500 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99 or 100% fetal nucleic acid is of a length of about 500 base pairs or less). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 250 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99 or 100% fetal nucleic acid is of a length of about 250 base pairs or less). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 200 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99 or 100% fetal nucleic acid is of a length of about 200 base pairs or less). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 150 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99 or 100% fetal nucleic acid is of a length of about 150 base pairs or less). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 100 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99, or 100% fetal nucleic acid is of a length of about 100 base pairs or less). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 50 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99 or 100% fetal nucleic acid is of a length of about 50 base pairs or less). In some embodiments, the majority of the nucleic acid of the fetal nucleic acid is of a length of about 25 base pairs or less (e.g., about 80, 85, 90, 91, 92, 93, 94, 95, 96 , 97, 98, 99, or 100% fetal nucleic acid is of a length of about 25 base pairs or less).

[068] The nucleic acid can be provided for carrying out methods described herein without processing the sample(s) containing the nucleic acid, in certain embodiments. In some embodiments, the nucleic acid is provided for carrying out methods described herein, after processing the sample(s) containing the nucleic acid. For example, a nucleic acid can be extracted, isolated, purified, partially purified or amplified from the sample(s). The term "isolated" as used herein refers to nucleic acid removed from its original environment (e.g., the natural environment if it is naturally occurring, or a host cell is exogenously expressed), and therefore is altered by human intervention (e.g., "by the hand of man") from its original environment. The term "isolated nucleic acid", as used herein, can refer to nucleic acid removed from a subject (e.g., a human subject). An isolated nucleic acid may be provided with fewer non-nucleic acid components (eg, proteins, lipids) than the amount of components present in a source sample. A composition comprising isolated nucleic acid can be from about 50% to greater than 99% free acid free of non-nucleic acid components. A composition comprising isolated nucleic acid can be about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or greater than 99% free of acid components non-nucleic. The term "purified" as used herein may refer to a nucleic acid as long as it contains less of the non-nucleic acid components (e.g., protein, lipid, carbohydrate) than the amount of non-nucleic acid components. nucleic acids present before subjecting the nucleic acid to a purification process. A composition comprising purified nucleic acid can be about 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or greater than 99% free of other non-nucleic acid components. The term "purified" as used herein may refer to a nucleic acid as long as it contains fewer species of nucleic acid than in the sample source from which the nucleic acid is derived. A composition comprising purified nucleic acid can be about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or greater than 99% free of other species of nucleic acid For example, fetal nucleic acid can be purified from a mixture comprising maternal and fetal nucleic acid. In certain examples, nucleosomes comprising small fragments of fetal nucleic acid can be purified from a mixture of larger nucleosome complexes comprising larger fragments of maternal nucleic acid.

[069] In some embodiments nucleic acids are fragmented or cleaved before, during or after a method described herein. Fragmented or cleaved nucleic acid can have a nominal, medium, or short length of about 5 to about 10,000 base pairs, about 100 to about 1000 base pairs, about 100 to about 500 base pairs, or about from 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 200, 300, 400, 500, 600, 700 , 800, 900, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000 or 9000 base pairs. Fragments can be generated by a suitable method known in the art, and the average, low, or nominal length of nucleic acid fragments can be controlled by selecting a suitable fragment generation procedure.

[070] Nucleic acid fragments may contain overlapping nucleotide sequences, and such overlapping sequences may facilitate the construction of an unfragmented counter nucleic acid nucleotide sequence, or segment thereof. For example, one fragment may have x and y subsequences and another fragment may have y and z subsequences, where x, y, and z are nucleotide sequences that may be 5 nucleotides in length or longer. Y-overlapping sequence can be used to facilitate the construction of the x-y-z nucleotide sequence in the nucleic acid of a sample in certain embodiments. Nucleic acid can be partially fragmented (e.g., from an incomplete or terminated specific cleavage reaction) or fully fragmented in certain embodiments.

[071] In some embodiments, nucleic acid is fragmented or cleaved by a suitable method, non-limiting examples of which include physical methods (e.g., shear, e.g., sonication, French press, heat, UV irradiation, the like), enzymatic processes (e.g., enzymatic cleavage agents (e.g., a suitable nuclease, a suitable restriction enzyme, a suitable methylation-sensitive restriction enzyme)), chemical methods (e.g., alkylation, DMS, piperidine, acid hydrolysis, basic hydrolysis, heat, the like, or combinations thereof), processes described in US Patent Application Publication No. 20050112590, the like, or combinations thereof.

[072] As used herein, "fragmentation" or "cleavage" refers to a procedure or conditions in which a nucleic acid molecule, such as a gene template of the nucleic acid molecule or its amplified product, can be separated into two or more smaller nucleic acid molecules. Such fragmentation or cleavage may be sequence-specific, base-specific, or non-specific, and may be accomplished by any of a variety of methods, reagents or conditions, including, for example, chemical, enzymatic, physical fragmentation.

[073] As used herein, "fragments", "cleavage products", "cleaved products" or their grammatical variants thereof, refers to nucleic acid molecules resulting from a fragmentation or cleavage of a gene model of the acid molecule nucleic acid or amplified product thereof. While such fragments or cleavage products may refer to all nucleic acid molecules resulting from a cleavage reaction, typically such fragments or cleavage products only refer to nucleic acid molecules resulting from a fragmentation or cleavage of a gene template. of the nucleic acid molecule or segment of an amplified product containing the same corresponding nucleotide sequence of a model nucleic acid molecule gene. The term "amplification", as used herein, refers to subjecting a target nucleic acid in a sample to a process that linearly or exponentially generates amplicon nucleic acids having the same or substantially the same nucleotide sequence as the target nucleic acid. , or its thread.

[074] In certain embodiments, the term "amplified" refers to a method comprising a polymerase chain reaction (PCR). For example, an amplified product may contain one or more nucleotides more than the amplified nucleotide region of a template nucleic acid sequence (e.g., a primer may contain "extra" nucleotides, such as a transcription initiation sequence, in addition to nucleotides complementary to a template nucleic acid molecule gene, resulting in an amplified product containing "extra" nucleotides or nucleotides not corresponding to the amplified nucleotide region of the template nucleic acid molecule gene). Accordingly, fragments may include resulting fragments of segments or parts of amplified nucleic acid molecules containing, at least in part, nucleotide sequence information from or based on the representative model nucleic acid molecule.

[075] As used herein, the term "complementary cleavage reactions" refers to cleavage reactions that are performed on the same nucleic acid using different cleavage reagents or by changing the cleavage specificity of the same cleavage reagent, such that the standards Alternative cleavage sequences of the same target or reference nucleic acid or protein are generated. In certain embodiments, the nucleic acid can be treated with one or more specific cleavage agents (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more specific cleavage agents) in a or more reaction portions (eg, the nucleic acid is treated with each specific cleavage agent in a separate portion). The term "specific cleavage agent" as used herein refers to an agent, sometimes a chemical or an enzyme, that can cleave a nucleic acid at one or more specific sites.

[076] The nucleic acid may also be exposed to a process that modifies certain nucleotides in the nucleic acid before providing nucleic acid for a method described herein. A process that selectively modifies nucleic acid based on its nucleotide methylation state can be applied to nucleic acid, for example. Furthermore, conditions such as high temperature, ultraviolet radiation, X-radiation, can induce changes in the sequence of a nucleic acid molecule. The nucleic acid may be provided in any suitable form useful for conducting the proper sequence analysis.

[077] The nucleic acid can be single-stranded or double-stranded. Single-stranded DNA, for example, can be generated by denaturing double-stranded DNA by heating or by treating it with alkali, for example. In certain embodiments, the nucleic acid is in a LASSO D structure, formed by strand invasion of a duplex DNA molecule by an oligonucleotide or a DNA-like molecule, such as peptide nucleic acid (PNA). Formation of LASSO D can be facilitated by the addition of E.coli RecA protein and/or by altering the salt concentration, for example, using methods known in the art.

genomic targets

[078] In some embodiments, target nucleic acids, also referred to herein as target fragments, include fragments of polynucleotides from a particular genomic region or plurality of genomic regions (e.g., single chromosome, set of chromosomes, and/or certain chromosomal regions ). In some embodiments, such genomic regions may be associated with fetal genetic abnormalities (e.g., aneuploidy), as well as other genetic variations, including, but not limited to, mutations (e.g., point mutations), insertions, additions, deletions, translocations, trinucleotide repeat disorders, and/or single nucleotide polymorphisms (SNPs). In some embodiments, reference nucleic acids, also referred to herein as reference fragments, include polynucleotide fragments from a particular genomic region or plurality of genomic regions not associated with fetal genetic abnormalities. In some embodiments, target and/or reference nucleic acids (i.e., target fragments and/or reference fragments) include nucleotide sequences that are substantially unique to the chromosome of interest or reference chromosome (i.e., identical nucleotide sequences or substantially similar nucleotide sequences are not found elsewhere in the genome).

[079] In some embodiments, fragments from a plurality of genomic regions are evaluated. In some embodiments, target fragments and reference fragments from a plurality of genomic regions are evaluated. In some embodiments, fragments from a plurality of genomic regions are evaluated to determine the presence, absence, amount (e.g., relative amount) or quotient of a chromosome of interest, for example. In some embodiments, a chromosome of interest is a chromosome suspected of having aneuploidy and may be referred to herein as a "test chromosome". In some embodiments, fragments from a plurality of genomic regions are evaluated for a presumed euploid chromosome. Such a chromosome may be referred to herein as a "reference chromosome". In some embodiments, a plurality of test chromosomes are evaluated. In some embodiments, the test chromosomes are selected from among chromosome 13 (Chr13), chromosome 18 (Chr18), and chromosome 21 (Chr21). In some embodiments, the reference chromosomes are selected from chromosomes 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 , 20, 21, 22, X and Y, and sometimes the reference chromosomes are selected autosomes (ie not X and Y). In some embodiments, chromosome 20 (Chr20) is selected as a reference chromosome. In some embodiments, chromosome 14 is selected as a reference chromosome. In some embodiments, chromosome 9 is selected as a reference chromosome. In some embodiments, a test chromosome and a reference chromosome are from the same individual. In some embodiments, a test chromosome and a reference chromosome are from different individuals.

[080] In some embodiments, fragments from at least one genomic region are evaluated for a test and/or reference chromosome. In some embodiments, fragments from at least 10 genomic regions (e.g., about 20, 30, 40, 50, 60, 70, 80, or 90 genomic regions) are evaluated for a test chromosome and/or a reference chromosome. In some embodiments, fragments from at least 100 genomic regions (e.g., about 200, 300, 400, 500, 600, 700, 800, or 900 genomic regions) are evaluated for a test chromosome and/or a reference chromosome . In some embodiments, fragments from at least 1000 genomic regions (e.g., about 2000, 3000, 4000, 5000, 6000, 7000, 8000, or 9000 genomic regions) are evaluated for a test chromosome and/or a reference chromosome. . In some embodiments, fragments from at least 10,000 genomic regions (e.g., about 20,000, 30,000, 40,000, 50,000, 60,000, 70,000, 80,000, or 90,000 genomic regions) are evaluated for a test chromosome and/or a reference chromosome. In some embodiments, fragments from at least 100,000 genomic regions (e.g., about 200,000, 300,000, 400,000, 500,000, 600,000, 700,000, 800,000, or 900,000 genomic regions) are evaluated for a test chromosome and/or reference chromosome.

Enrichment and separation of nucleic acid subpopulations

[081] In some embodiments, the nucleic acid (e.g., extracellular nucleic acid) is enriched or relatively enriched by a subpopulation or species of nucleic acid. Nucleic acid subpopulations can include, for example, fetal nucleic acid, maternal nucleic acid, fragments comprising nucleic acid of a particular length or range of lengths, or nucleic acid from a specific region of the genome (e.g., single chromosome, a set of of chromosomes, and/or certain chromosomal regions). Such enriched samples can be used in conjunction with a method provided herein. Thus, in certain embodiments, methods of the technology comprise an additional step of enriching a subpopulation of nucleic acid in a sample, such as, for example, fetal nucleic acid. In certain embodiments, a method for determining fetal fraction described herein can also be used to enrich fetal nucleic acid. In certain embodiments, maternal nucleic acid is selectively removed (partially, substantially, completely or almost completely) from the sample. In certain embodiments, enriching low copy number nucleic acid species (e.g., fetal nucleic acid) can improve quantitative sensitivity. Methods for enriching a sample for a particular species of nucleic acid are described, for example, in US Patent No. 6,927,028, International Patent Application Publication No. WO2007/140417, International Patent Application Publication No. WO2007 /147063, International Patent Application Publication No. WO2009/032779, International Patent Application Publication No. WO2009/032781, International Patent Application Publication No. WO2010/033639, International Patent Application Publication No. W02011/034631 , International Patent Application Publication No. WO2006/056480, and International Patent Application Publication No. WO2011/143659, all of which are incorporated herein by reference.

[082] In some embodiments, the nucleic acid is enriched with certain target fragment species and/or reference fragment species. In certain embodiments, the nucleic acid is enriched for a specific nucleic acid fragment length or range of fragment lengths using one or more length-based separation methods described below. In certain embodiments, the nucleic acid is enriched for fragments from a selected genomic region (e.g., chromosome) using one or more sequence-based separation methods described herein and/or known in the art. Certain methods for enriching a subpopulation of nucleic acid (eg, fetal nucleic acid) in a sample are described in detail below.

[083] Some methods for enriching a subpopulation of nucleic acid (e.g., fetal nucleic acid) that can be used with a method described herein include methods that exploit differences between maternal and fetal epigenetic nucleic acid. For example, fetal nucleic acid can be differentiated and separated from maternal nucleic acid on the basis of methylation differences. Methods of enriching fetal nucleic acid based on methylation are described in US Patent Application Publication No. 2010/0105049, which is incorporated herein by reference. Such methods sometimes involve attaching a sample nucleic acid to a methylation-specific binding agent (CpG-methyl binding protein (MBD), methylation-specific antibodies, and the like) separating bound nucleic acid from unbound nucleic acid. bound based on differential methylation state. Such methods may also include the use of methylation-sensitive restriction enzymes (as described above, e.g., Hhal and Hpall), which allow the enrichment of fetal nucleic acid regions in a maternal sample by selectively digesting the maternal sample's nucleic acid. with an enzyme that selectively and completely or substantially digests maternal nucleic acid to enrich the sample for at least one region of fetal nucleic acid.

[084] Another method for enriching a subpopulation of nucleic acid (e.g., fetal nucleic acid) that can be used with a method described herein is a restriction endonuclease-enhanced polymorphic sequence approach such as a method described in the publication of the US Patent Application No. 2009/0317818, which is incorporated herein by reference. Such methods include cleaving nucleic acid comprising a non-target allele with a restriction endonuclease that recognizes nucleic acid comprising the non-target allele but not the target allele; and amplifying the cleaved nucleic acid, but not the cleaved nucleic acid, wherein the amplified, uncleaved nucleic acid represents the enriched target nucleic acid (e.g., fetal nucleic acid) in relation to the non-target nucleic acid (e.g., fetal nucleic acid) maternal). In certain embodiments, the nucleic acid can be selected such that it comprises an allele having a polymorphic site that is susceptible to selective digestion by a cleavage agent, for example.

[085] Some methods for enriching a subpopulation of nucleic acid (eg, fetal nucleic acid) that can be used with a method described herein include selective enzymatic degradation approaches. Such methods involve protecting target sequences from exonuclease digestion thereby facilitating the elimination in a sample of unwanted sequences (eg, maternal DNA). For example, in one approach, sample nucleic acid is denatured to generate single-stranded nucleic acid, single-stranded nucleic acid is contacted with at least one target-specific primer pair under suitable hybridization conditions, paired primers are extended by nucleotide polymerization generating double-stranded target sequences, and digesting single-stranded nucleic acid using a nuclease that digests single-stranded (i.e., non-targeting) nucleic acid. In certain embodiments, the method can be repeated for at least one additional cycle. In certain embodiments, the same target specific primer pair is used to inject each of the first and second extension cycles, and in certain embodiments, different target specific primer pairs are used for the first and second cycles.

[086] In some embodiments, the nucleic acid is enriched with fragments from a selected genomic region (eg, chromosome) using one or more sequence-based separation methods described herein. In some embodiments, the nucleic acid is enriched with a specific polynucleotide fragment length or range of fragment lengths and fragments from a selected genomic region (e.g., chromosome) using a combination of length-based sequence-based separation methods . Such sequence-based and length-based separation methods are described in greater detail below.

[087] Some methods for enriching a subpopulation of nucleic acid (e.g., fetal nucleic acid) that can be used with a method described herein include massively parallel signature sequencing (MPSS). MPSS is typically a solid phase method that uses adapter ligation (i.e. labeling), followed by adapter decoding, and reading the nucleic acid sequence in small increments. Tagged PCR products are typically amplified such that each nucleic acid generates a PCR product with an individual tag. Labels are often used to link PCR products to microbeads. After several rounds of binding-based sequencing, for example, a sequence signature can be identified from each bead. Each signature string (marked with MPSS) in an MPSS dataset is analyzed, compared to all other signatures, and all identical signatures are counted.

[088] In certain embodiments, certain enrichment methods (eg, certain MPS and/or MPSS-based enrichment methods) may include amplification-based approaches (eg, PCR). In certain embodiments, loci-specific amplification methods can be used (eg, loci-specific amplification primers). In certain embodiments, a multiplex SNP allele PCR approach can be used. In certain embodiments, a multiplex SNP allele PCR approach can be used in combination with uniplex sequencing. For example, such an approach may involve the use of multiplex PCR (eg, MASSARRAY system) and incorporation of capture probe sequences into amplicons followed by sequencing using, for example, the MPSS Illumina system. In certain embodiments, a multiplex SNP allele PCR approach can be used in combination with a three-primer system of indexed sequencing. For example, such an approach may involve the use of multiplex PCR (e.g. MASSARRAY system) with primers having a first capture probe incorporated into certain specific forward PCR primers and adapter sequences incorporated into loci-specific reverse PCR primers for, thereby generating amplification products, followed by a secondary PCR to incorporate reverse capture sequences and molecular index barcodes for sequencing using, for example, the Illumina MPSS system. In certain embodiments, a multiplex SNP allele PCR approach can be used in combination with a four primer system and indexed sequencing. For example, such an approach may involve the use of multiplex PCR (e.g. MASSARRAY system) with primers having adapter sequences incorporated into both loci-specific forward and reverse loci-specific PCR primers and followed by a secondary PCR to incorporate both forward and reverse capture sequences and molecular index barcodes for sequencing using, for example, the Illumina MPSS system. In certain embodiments, a microfluidic approach can be used. In certain embodiments, matrix-based microfluids can be used. For example, such an approach might involve the use of a microfluidic matrix (eg, Fluidigm) for low plex amplification and incorporation of index and capture probes, followed by sequencing. In certain embodiments, an emulsion microfluidic approach may be used, such as, for example, digital dot PCR.

[089] In certain embodiments, universal amplification methods may be used (for example, using universal or non-loci-specific amplification primers). In certain embodiments, universal amplification methods can be used in combination with pulldown approaches. In certain embodiments, a method may include biotinylated ultramer pulldown (e.g., Agilent or IDT biotinylated pulldown assays) of a universally amplified sequencing library. For example, such an approach might involve preparing a standard library, enriching selected regions by a pulldown assay, and a secondary universal amplification step. In certain embodiments, pulldown approaches can be used in combination with linkage-based methods. In certain embodiments, a method may include biotinylated ultramer pulldown with sequence-specific adapter ligation (e.g., HALOPLEX PCR, from Halo Genomics). For example, such an approach might involve the use of selector probes to capture restriction enzyme-digested fragments, followed by ligation of captured products to an adapter, and universal amplification followed by sequencing. In certain embodiments, pulldown approaches can be used in combination with extension and binding-based methods. In certain embodiments, a method may include molecular inversion probe (MIP) binding and extension. For example, such an approach might involve the use of molecular inversion probes in combination with sequence adapters, followed by universal amplification and sequencing. In certain embodiments, complementary DNA can be synthesized and sequenced without amplification.

[090] In certain embodiments, extension and linkage approaches can be performed without a pulldown component. In certain embodiments, a specific method may include loci-specific forward and reverse primer hybridization, extension, and ligation. Such methods may also include universal amplification or complementary DNA synthesis without amplification, followed by sequencing. Such methods can reduce or exclude base sequences during analysis, in certain embodiments.

[091] In certain embodiments, pulldown approaches can be used with an optional amplification component or without an amplification component. In certain embodiments, a method may include a modified pulldown assay and binding with total incorporation of capture probes without universal amplification. For example, such an approach may involve the use of modified selector probes to capture fragments digested by capture restriction enzymes, followed by ligation of captured products to an optional adapter, amplification, and sequencing. In certain embodiments, a method may include a biotinylated pulldown assay and with extension and adapter sequence ligation in combination with circular single-stranded ligation. For example, such an approach might involve the use of selector probes to capture regions of interest (i.e., target sequences), probe extension, adapter ligation, single-stranded circular ligation, optional amplification, and sequencing. In certain embodiments, analysis of sequencing results can separate target sequences from background.

[092] In some embodiments, the nucleic acid is enriched with fragments from a selected genomic region (e.g., chromosome) using one or more sequence-based separation methods described herein. Sequence-based separation is generally based on nucleotide sequences present in the fragments of interest (e.g., target and/or reference fragments) and substantially not present in other fragments in the sample or present in an insubstantial amount of other fragments ( for example, 5% or less). In some embodiments, sequence-based separation can generate separate target fragments and/or separate reference fragments. Separate target fragments and/or separate reference fragments are often isolated from the remaining fragments in the nucleic acid sample. In certain embodiments, the separate target fragments and the separate reference fragments are also isolated from one another (e.g., isolated in separate compartments in the assay). In certain embodiments, separate target fragments and separate reference fragments are isolated together (e.g., isolated in the same assay compartment). In some embodiments, unlinked fragments can be differentially removed or degraded or digested.

[093] In some embodiments, a selective nucleic acid capture process is used to separate target and/or reference fragments from the nucleic acid sample. Commercially available nucleic acid capture systems include, for example, the NimbleGen sequence capture system (Roche NimbleGen, Madison, WI); Illumina's BEADARRAY platform (Illumina, San Diego, CA); Affymetrix GeneChip platform (Affymetrix, Santa Clara, CA); Agilent SureSelect target enrichment system (Agilent Technologies, Santa Clara, CA); and related platforms. Such methods typically involve hybridizing a capture oligonucleotide to a segment or the entire nucleotide sequence of a target or reference fragment and may include the use of a solid phase (e.g., solid phase array) and/or a solution-based platform. Capture oligonucleotides (sometimes referred to as "bait") can be selected or designed such that they preferentially hybridize with nucleic acid fragments from selected genomic regions or loci (e.g., one of chromosomes 21, 18, 13, X or Y, or a reference chromosome). In certain embodiments, a hybridization-based method (e.g., using oligonucleotide arrays) can be used to enrich nucleic acid sequences from certain chromosomes (e.g., a potentially aneuploidy chromosome, reference chromosome, or other chromosome of interest) or their segments of interest.

[094] Capture oligonucleotides typically comprise a nucleotide sequence capable of hybridizing or annealing a nucleic acid fragment of interest (eg, target fragment, reference fragment) or a portion thereof. A capture oligonucleotide can be natural or synthetic and can be DNA or RNA based. Capture oligonucleotides can allow specific separation of, for example, a target and/or reference fragment from other fragments in a nucleic acid sample. The term "specific" or "specificity", as used herein, refers to the binding or hybridization of a molecule to another molecule, such as an oligonucleotide to a target polynucleotide. "Specific" or "specificity" refers to the recognition, contact, and formation of a stable complex between the two molecules, as compared to substantially less recognition, contact, or complex formation of either of these two molecules with other molecules. As used herein, the term "annealing" refers to the formation of a stable complex between the two molecules. The terms "capture oligonucleotide", "capture oligonucleotide", "oligo", or "oligonucleotide" may be used interchangeably throughout the document when referring to capturing oligonucleotides. The following characteristics of oligonucleotides can be applied to primers and other oligonucleotides such as probes provided herein.

[095] A capture oligonucleotide can be designed and synthesized using a suitable process, and can be of any length suitable for hybridizing with a nucleotide sequence of interest and performing separation and/or analysis procedures described herein. Oligonucleotides can be designed based on a nucleotide sequence of interest (eg target fragment sequence, reference fragment sequence). An oligonucleotide, in some embodiments, can be from about 10 to about 300 nucleotides, about 10 to about 100 nucleotides, about 10 to about 70 nucleotides, about 10 to about 50 nucleotides, about 15 to about 30 nucleotides, or about 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 or 100 nucleotides in length. An oligonucleotide can be composed of naturally occurring and/or non-naturally occurring nucleotides (e.g., labeled nucleotides), or a mixture of these. Oligonucleotides suitable for use with the embodiments described herein can be synthesized and labeled using known techniques. Oligonucleotides can be chemically synthesized according to the solid phase phosphoramidite triester method first described by Beaucage and Caruthers (1981) Tetrahedron Letts. 22: 18591862, using an automatic synthesizer, and/or as described in Needham-VanDevanter et al. (1984) Nucleic Acids Res. 12: 6159-6168. Purification of oligonucleotides can be accomplished by native acrylamide gel electrophoresis or by anion exchange high performance liquid chromatography (HPLC), for example, as described in Pearson and Regnier (1983) J. Chrom. 255: 137-149.

[096] All or a portion of an oligonucleotide sequence (naturally occurring or synthetic) may be substantially complementary to a target and/or reference fragment sequence or portion thereof, in some embodiments. As referred to herein, "substantially complementary" in relation to sequences refers to nucleotide sequences that will hybridize to one another. The stringency of the hybridization conditions can be altered to tolerate varying amounts of sequence mismatch.

[097] Target/reference oligonucleotide sequences are included that are 55% or more, or more than 56%, 57% or more, 58% or more, 59% or more, 60% or more, 61% or more, 62 % or more, 63% or more, 64% or more, 65% or more, 66% or more, 67% or more, 68% or more, 69% or more, 70% or more, 71% or more, 72 % or more, 73% or more, 74% or more, 75% or more, 76% or more, 77% or more, 78% or more, 79% or more, 80% or more, 81% or more, 82 % or more, 83% or more, 84% or more, 85% or more, 86% or more, 87% or more, 88% or more, 89% or more, 90% or more, 91% or more, 92 % or more, 93% or more, 94% or more, 95% or more, 96% or more, 97% or more, 98% or more, or 99% or more complement each other.

[098] Oligonucleotides that are substantially complementary to a nucleic acid sequence of interest (for example, target fragment sequence, reference fragment sequence) or portion thereof are also substantially similar to the complement of the target nucleic acid sequence or relevant portion same (e.g., substantially similar to the sense nucleic acid strand). One test to determine whether two nucleotide sequences are substantially similar is to determine the percentage of identical nucleotide sequences. As referred to herein, "substantially similar" with respect to sequences refers to nucleotide sequences that are 55% or more, 56% or more, 57% or more, 58% or more, 59% or more, 60% or more , 61% or more, 62% or more, 63% or more, 64% or more, 65% or more, 66% or more, 67% or more, 68% or more, 69% or more, 70% or more , 71% or more, 72% or more, 73% or more, 74% or more, 75% or more, 76% or more, 77% or more, 78% or more, 79% or more, 80% or more , 81% or more, 82% or more, 83% or more, 84% or more, 85% or more, 86% or more, 87% or more, 88% or more, 89% or more, 90% or more , 91% or more, 92% or more, 93% or more, 94% or more, 95% or more, 96% or more, 97% or more, 98% or more or 99% or more identical to each other.

[099] Annealing conditions (for example, hybridization conditions) can be determined and/or adjusted, depending on the characteristics of the oligonucleotides used in an assay. Oligonucleotide sequence and/or length can sometimes affect hybridization to a nucleic acid sequence of interest. Depending on the degree of mismatch between an oligonucleotide and the nucleic acid of interest, low, medium, or high stringency conditions can be used to perform annealing. As used herein, the term "stringent conditions" refers to hybridization and wash conditions. Methods for optimizing the hybridization reaction and temperature condition are known in the art, and can be found in Current Protocols in Molecular Biology, John Wiley & Sons, N.Y., 6.3.1-6.3.6 (1989). Aqueous and non-aqueous methods are described in that reference and either can be used. Non-limiting examples of stringent hybridization conditions are hybridization in 6X sodium chloride/sodium citrate (SSC) at about 45°C, followed by one or more washes in 0.2X SSC, 0.1% SDS at 50 °C. Another example of stringent hybridization conditions is hybridization in 6X sodium chloride/sodium citrate (SSC) at about 45°C, followed by one or more washes in 0.2X SSE, 0.1% SDS at 55° W. An additional example of stringent hybridization conditions is hybridization in 6X sodium citrate/sodium chloride (SSC) at about 45°C, followed by one or more washes in 0.2X SSE, 0.1% SDS at 60°C. Often, stringent hybridization conditions are hybridization in 6X sodium citrate/sodium chloride (SSC) at about 45°C, followed by one or more washes in 0.2X SSE, 0.1% SDS at 65°C . Most often, the stringent conditions are 0.5M sodium phosphate, 7% SDS at 65°C, followed by one or more washes in 0.2X SSC, 1% SDS at 65°C. Stringent hybridization temperatures can also be changed (ie, lowered) with the addition of certain organic solvents, eg formamide. Organic solvents such as formamide reduce the thermal stability of double-stranded polynucleotides, so that hybridization can be performed at lower temperatures, while still maintaining stringent conditions that extend the shelf life of nucleic acids that may be thermolabile.

[100] As used herein, the phrase "hybridization" or grammatical variations thereof, refers to the annealing of a first nucleic acid molecule with a second nucleic acid molecule under low, medium, or high stringency conditions, or under synthetic conditions. of nucleic acid. Hybridization can include instances where a first nucleic acid molecule hybridizes to a second nucleic acid molecule, where the first and second nucleic acid molecules are complementary. As used herein, "specifically hybridizes" refers to the preferential hybridization under conditions of nucleic acid synthesis of an oligonucleotide to a nucleic acid molecule having a sequence complementary to the oligonucleotide compared to hybridization to a nucleic acid molecule having a sequence not complementary. For example, specific hybridization includes hybridizing a capture oligonucleotide to a target fragment sequence that is complementary to the oligonucleotide.

[101] In some embodiments, one or more capture oligonucleotides are associated with an affinity ligand, such as a member of a binding pair (e.g., biotin) or antigen that can bind a capture agent, such as avidin, streptavidin, an antibody, or a receptor. For example, a capture oligonucleotide can be biotinylated in such a way that it can be captured onto a streptavidin-coated bead.

[102] In some embodiments, one or more capture oligonucleotides and/or capture agents are effectively attached to a solid support or substrate. A solid support or substrate can be any solid surface to which a capture oligonucleotide can be directly or indirectly attached, including, but not limited to, microarray-provided surfaces and wells, and particles, such as beads (e.g., paramagnetic beads , magnetic granules, microgranules, nanogranules), microparticles and nanoparticles. Solid supports may also include, for example, chips, columns, optical fibers, wipes, filters (e.g., flat surface filters), one or more capillaries, glass, and modified or functionalized glass (e.g., controlled pore glass (e.g., controlled pore glass). CPG)), quartz, mica, diazotized membranes (paper or nylon), polyformaldehyde, cellulose, cellulose acetate, paper, ceramics, metals, metalloids, semiconductor materials, quantum dot, granules or coated particles, other chromatographic materials, magnetic particles; plastics (including acrylic, polystyrene, styrene copolymers or other materials, polybutylene, polyurethanes, Teflon™, polyethylene, polypropylene, polyamide, polyester, polyvinylidene difluoride (PVDF), and the like), polysaccharides, nylon or nitrocellulose, resins, silica or silicon-based materials including silicon, silica gel, and modified silicon, Sephadex®, Sepharose®, carbon, metals (e.g., steel, gold, silver, aluminum, silicon, and copper), inorganic glasses, conductive polymers (including polymers such as polypyrrole and polyindole); micro- or nano-structured surfaces such as nucleic acid matrices, nanotubes, nanowires or decorated nanoparticulate surfaces; or porous surfaces or gels, such as methacrylates, acrylamides, sugar polymers, cellulose, silicates, or other fibrous or ribbon polymers. In some embodiments, the solid support or substrate can be coated using passive or chemically derivatized coatings with any number of materials, including polymers, such as dextrans, acrylamides, gelatins, or agarose. Granules and/or particles can be free or in connection with each other (eg synthesized). In some embodiments, the solid phase can be a collection of particles. In some embodiments, the particles can comprise silica and the silica can comprise silica dioxide. In some embodiments the silica may be porous and in certain embodiments the silica may be non-porous. In some embodiments, the particles further comprise an agent that imparts a paramagnetic property to the particles. In certain embodiments, the agent comprises a metal, and in certain embodiments, the agent is a metal oxide, (e.g., iron or iron oxides, where the iron oxide contains a mixture of Fe2+ and Fe3+). Oligonucleotides can be attached to the solid support through covalent bonds or by non-covalent interactions, and can be attached to the solid support directly or indirectly (for example, through an intermediate agent such as a spacer molecule or biotin). A probe can be attached to the solid support before, during or after nucleic acid capture.

[103] In some embodiments, the nucleic acid is enriched for a particular nucleic acid fragment length, range of lengths, or lengths under or over a particular threshold or cutoff using one or more length-based separation methods. Nucleic acid fragment length typically refers to the number of nucleotides in the fragment. Nucleic acid fragment length is also sometimes referred to as nucleic acid fragment size. In some embodiments, a length-based separation method is performed without measuring individual fragment lengths. In some embodiments, a length-based separation method is performed in conjunction with a method for determining the length of individual fragments. In some embodiments, length-based separation refers to a size fractionation procedure, where all or part of the fractionated set can be isolated (e.g., retained) and/or analyzed. Size fractionation procedures are known in the art (e.g., matrix separation, molecular sieve separation, gel electrophoresis separation, column chromatography separation (e.g., size exclusion columns), and approaches based on microfluids). In certain embodiments, length-based separation approaches may include fragment circularization, chemical treatment (e.g., formaldehyde, polyethylene glycol (PEG)), mass spectrometry, and/or size-specific nucleic acid amplification, for example.

[104] In some embodiments, nucleic acid fragments of a certain length, range of lengths, or lengths below or above a particular threshold or cutoff, are separated from the sample. In some embodiments, fragments with a length under a particular threshold or cutoff (e.g., 500 bp, 400 bp, 300 bp, 200 bp, 150 bp, 100 bp) are referred to as "short" fragments and fragments with a length along of a particular boundary or cut (eg, 500bp, 400bp, 300bp, 200bp, 150bp, 100bp) are referred to as "long" fragments. In some embodiments, fragments of a certain length, range of lengths, or lengths below or above a particular threshold or cutoff, are retained for analysis as fragments of a different length or range of lengths, or lengths above or below the threshold or cutoff. are not retained for analysis. In some embodiments, fragments that are smaller than about 500 bp are retained. In some embodiments, fragments that are smaller than about 400 bp are retained. In some embodiments, fragments that are smaller than about 300 bp are retained. In some embodiments, fragments that are smaller than about 200 bp are retained. In some embodiments, fragments that are smaller than about 150 bp are retained. For example, fragments that are less than about 190 bp, 180 bp, 170 bp, 160 bp, 150 bp, 140 bp, 130 bp, 120 bp, 110 bp and 100 bp are retained. In some embodiments, fragments that are about 100 bp to about 200 bp are retained. For example, fragments that are about 190 bp, 180 bp, 170 bp, 160 bp, 150 bp, 140 bp, 130 bp, 120 bp, or 110 bp are retained. In some embodiments, fragments that are in the range of about 100 bp to about 200 bp are retained. For example, fragments that are in the range of about 110 bp to about 190 bp, 130 bp to about 180 bp, 140 bp to about 170 bp, 140 bp to about 150 bp, 150 bp to about 160 bp , or 145 bp to about 155 bp are retained. In some embodiments, fragments that are about 10 bp to about 30 bp shorter than other fragments of a certain length or range of lengths are retained. In some embodiments, fragments from about 10 bp to about 20 bp shorter than other fragments of a certain length or range of lengths are retained. In some embodiments, fragments that are about 10 bp to about 15 bp shorter than other fragments of a certain length or range of lengths are retained.

[105] In some embodiments, the nucleic acid is enriched to a particular nucleic acid fragment length, range of lengths, or lengths above or below a particular threshold or cutoff using one or more bioinformatics-based methods (e.g., in silica) methods. For example, fragments (reads) of nucleotide sequences can be obtained for nucleic acid fragments using a suitable nucleotide sequencing process. In some cases, such as when a paired-end sequencing method is used, the length of a particular fragment can be determined based on the fragment positions (reads) of mapped sequences obtained from each end of the fragment. Sequence reads used for a particular analysis (e.g., determining the presence or absence of a genetic variation) can be enriched or filtered according to one or more selected fragment lengths or fragment fragment length threshold values. correspondents as described in greater detail here.

[106]Certain length-based separation methods that can be used with the methods described here sometimes use a sequence-selective tagging approach, for example. The term "sequence tagging" refers to the incorporation of a recognized and distinct sequence into a nucleic acid or population of nucleic acids. The term "sequence tag" as used herein has a different meaning than the term "sequence tag" described later herein. In such sequence labeling methods, fragment-sized nucleic acid species (eg, short fragments) are subjected to selective sequence labeling in a sample that includes both short and long nucleic acids. Such methods typically involve performing a nucleic acid amplification reaction using a set of nested primers that include nested primers and nested primers. In certain embodiments, one or both of the internals can be labeled to thereby introduce a tag into the target amplification product. Outside primers are generally not annealed with the short fragments that carry the target (internal) sequence. Internal primers can anneal the short fragments and generate an amplification product that carries a tag and the target sequence. Typically, labeling of long fragments is inhibited through a combination of mechanisms including, for example, blocked extension of inner primers by prior annealing and extension of outer primers. Enrichment of the tagged fragments can be achieved by any of a variety of methods including, for example, exonuclease digestion of single-stranded nucleic acid and amplification of the tagged fragments using amplification primers specific for at least one tag.

[107] Another length-based separation method that can be used with the methods described herein is to subject a nucleic acid sample to polyethylene glycol (PEG) precipitation. Examples of methods include those described in International Publication Application Nos. WO2007/140417 and WO2010/115016. This method generally entails contacting a nucleic acid sample with PEG in the presence of one or more monovalent salts under conditions sufficient to substantially precipitate large nucleic acids without substantially precipitating small nucleic acids (e.g., less than 300 nucleotides).

[108] Another size-based enrichment method that can be used with the methods described here involves ligation circularization, for example using circligase. Short nucleic acid fragments can typically be circularized more efficiently than long fragments. Non-circularized sequences can be separated from circularized sequences, and the short enriched fragments can be used for further analysis.

Determining the length of the fragment

[109] In some embodiments, the length is determined for one or more nucleic acid fragments. In some embodiments, the length is determined for one or more target fragments, thereby identifying one or more target fragment size species. In some embodiments, the length is determined for one or more target fragments and one or more reference fragments, thereby identifying one or more target fragment length species and one or more reference fragment length species. In some embodiments, the length of the fragment is determined by measuring the length of a probe that hybridizes to the fragment, which is discussed in more detail below. Nucleic acid fragment or probe length can be determined using any method suitable in the art for determining nucleic acid fragment length, such as, for example, a mass sensitive procedure (e.g. mass spectrometry (e.g. , matrix-assisted laser ionization/desorption mass spectrometry (MALDI) and electrospray mass spectrometry (ES)), electrophoresis (e.g., capillary electrophoresis), microscopy (scanning tunneling microscopy, atomic force microscopy), measuring the length using a nanopore, and sequence-based length determination (e.g., paired-end sequencing). In some embodiments, the length of the probe or fragment can be determined without using a sequence-based separation method. fragment charge In some embodiments, the length of the fragment or probe can be determined without using an electrophoresis procedure In some embodiments, the length of the fragment or probe can be determined without using a nucleotide sequencing procedure.

Mass spectrometry

[110] In some embodiments, mass spectrometry is used to determine the length of the nucleic acid fragment. Mass spectrometry methods are typically used to determine the mass of a molecule, such as a nucleic acid fragment. In some embodiments, the length of the nucleic acid fragment can be extrapolated from the mass of the fragment. In some embodiments, a predicted range of the length of the nucleic acid fragment can be extrapolated from the mass of the fragment. In some embodiments, the length of the nucleic acid fragment can be extrapolated from the mass of a probe that hybridizes to the fragment, which is described in greater detail below. In some embodiments, the presence of a target and/or reference nucleic acid of a given length can be verified by comparing the mass of the detected signal with the expected mass of the target and/or reference fragment. Relative signal strength, e.g. peak mass in a spectrum for a particular fragment length and/or nucleic acid fragment, can sometimes indicate the relative population of fragment species among other nucleic acids present in the sample (see , for example, Jurinke et al (2004) 26, Mol. Biotechnol., 147-164 ).

[111]Mass spectrometry generally works by ionizing chemical compounds to generate charged molecules or fragments of molecules and measuring their mass-to-charge ratios. A typical mass spectrometry procedure involves several steps, including (1) loading a sample into the mass spectrometry instrument followed by vaporization, (2) ionization of sample components by any of a variety of methods (e.g., impacting with an electron beam), resulting in charged particles (ions), (3) separation of ions according to their mass to charge ratio in an analyzer by electromagnetic fields, (4) detection of ions (e.g., by a quantitative method), and (5) signal processing of the ions in the mass spectrum.

[112]Mass spectrometry methods are well known in the art (see, for example, Burlingame et al. Anal. Chem. 70:647R-716R (1998)), and include, for example, quadrupole mass spectrometry, ion capture mass spectrometry, time-of-flight mass spectrometry, gas chromatography mass spectrometry, and tandem mass spectrometry can be used with the methods described here. The basic processes associated with a mass spectrometry method are the generation of sample-derived gas phase ions, and the measurement of their masses. The movement of ions in the gas phase can be precisely controlled using electromagnetic fields generated in the mass spectrometer. The movement of ions in these electromagnetic fields is proportional to the m/z (mass to charge ratio) of the ion and this forms the basis of measuring m/z and hence the mass of a sample. The movement of ions in these electromagnetic fields allows the containment and focus of the ions which explains the high sensitivity of mass spectrometry. During the course of m/z measurement, ions are transmitted with high efficiency to particle detectors which register the arrival of these ions. The amount of ions in each m/z is shown by peaks on a graph where the x-axis is m/z and the y-axis is relative abundance. Different mass spectrometers have different levels of resolution, that is, the ability to resolve peaks between closely related ions by mass. Resolution is defined as R = m/delta m, where m is the mass of the ion and delta m is the difference in mass between two peaks in the mass spectrum. For example, a mass spectrometer with a resolution of 1000 can resolve an ion with m/z of 100.0 from an ion with m/z of 100.1. Certain mass spectrometry methods can utilize various combinations of ion sources and mass analyzers which allows flexibility in designing custom detection protocols. In some embodiments, mass spectrometers can be programmed to transmit all ions from the ion source into the mass spectrometer, either sequentially or at the same time. In some embodiments, a mass spectrometer can be programmed to select ions of a particular mass for transmission into the mass spectrometer while blocking other ions.

[113]Various types of mass spectrometers are available or can be produced in various configurations. In general, a mass spectrometer has the following major components: a sample inlet, an ion source, a mass analyzer, a detector, a vacuum system, and instrument control system, and a data system. Differences in sample input, ion source, and mass analyzer generally define instrument type and capabilities. For example, an input can be a capillary liquid chromatography source or it can be a direct probe or stage as used in matrix-assisted laser desorption. Common ion sources are, for example, electrospray, including nanospray and microspray, or matrix-assisted laser desorption. Mass analyzers include, for example, a quadrupole-type mass filter, ion capture mass analyzer, and time-of-flight mass analyzer.

[114]The ion formation process is the starting point for mass spectral analysis. Several ionization methods are available and the choice of ionization method depends on the sample used for the analysis. For example, for polypeptide analysis a relatively mild ionization procedure such as electrospray ionization (ESI) may be desirable. For ESI, a solution containing the sample is passed through a fine needle with a high potential that creates a strong electric field that results in a fine spray of highly charged droplets that is directed into the mass spectrometer. Other ionization procedures include, for example, fast atom bombardment (FAB), which uses a high-energy beam of neutral atoms to attack a solid sample causing desorption and ionization. Matrix-assisted laser desorption ionization (MALDI) is a method in which a laser pulse is used to attack a sample that has been crystallized in a matrix of a UV-absorbing compound (e.g., 2,5-dihydroxybenzoic acid, alpha-cyano-4-hydroxycinnamic acid, 3-hydroxypicolinic acid (3-HPA), diammonium citrate (DAC) and combinations thereof). Other ionization procedures known in the art include, for example, plasma and glow discharge, plasma desorption ionization, resonance ionization, and secondary ionization.

[115]A variety of mass analyzers are available that can be paired with different ion sources. Different mass analyzers have different advantages as known in the art and as described herein. The mass spectrometer and methods chosen for detection depend on the particular assay, for example a more sensitive mass analyzer may be used when a small amount of ions is generated for detection. Various types of mass analyzers and mass spectrometry methods are described below.

[116]Ion mobility (IM) mass spectrometry is a gas phase separation method. IM separates gas phase ions based on their collision cross section and can be coupled with time-of-flight (TOF) mass spectrometry. IM-MS is discussed in more detail by Verbeck et al., in Journal of Biomolecular Techniques 13 Vol., Issue 2, 56-61).

[117]Quadrupole-type mass spectrometry uses a quadrupole-type mass analyzer or filter. This type of mass analyzer consists of four rods arranged in two sets of two electrically connected rods. A combination of rf and dc voltages are applied to each pair of rods which produce fields that cause the ions to oscillate in motion as they travel from the beginning of the mass filter to the end. The result of these fields is the production of a high-pass mass filter on one pair of rods and a low-pass filter on the other pair of rods. The overlap between the high-pass and low-pass filters leads to a defined m/z that can pass both filters and traverse the length of the quadrupole. This m/z is selected and remains stable in the quadrupole mass filter, while all other m/z have unstable trajectories and do not remain in the mass filter. A mass spectrum results by increasing the applied fields such that an increase in m/z is selected to pass through the mass filter and reach the detector. In addition, quadrupoles can also be configured to contain and transmit all m/z ions by applying an rf-only field. This allows the quadrupole to function as a lens or focusing system in regions of the mass spectrometer where transmission is required without mass filtering.

[118] A quadrupole type mass analyzer, as well as the other mass analyzers described here, can be programmed to analyze a defined m/z or range of mass. Once the desired mass range of the nucleic acid fragment is known, in some cases a mass spectrometer can be programmed to transmit ions of the correct projected mass range while excluding ions of a higher or lower mass range. The ability to select a mass range can decrease background noise in the assay and thereby increase the signal-to-noise ratio. Thus, in some cases, a mass spectrometer can perform a separation step, as well as the detection and identification of certain nucleic acid fragments distinguishable from the mass.

[119]Ion capture mass spectrometry uses an ion capture mass analyzer. Typically, fields are applied so that all m/z ions are initially captured and oscillate in the mass analyzer. Ions enter the ion trap from the ion source through a focusing device, such as an octapole lens system. Ion capture occurs in the capture region before excitation and ejection through an electrode to the detector. Mass analysis can be performed by sequentially applying voltages that increase the amplitude of the oscillations in a way that ejects ions of increasing m/z out of the trap and into the detector. In contrast to quadrupole mass spectrometry, all ions are retained in the mass analyzer fields, except those with m/z selected. Control of the number of ions can be accomplished by varying the time during which the ions are injected into the trap.

[120] Time-of-flight mass spectrometry uses a time-of-flight mass analyzer. Typically, an ion is first given a fixed amount of kinetic energy by accelerating it in an electric field (generated by high voltage). In the acceleration sequence, the ion enters a "drag" free field region, where it travels at a velocity that is inversely proportional to m/z. Therefore, ions with low m/z travel faster than ions with high m/z. The time required for the ions to travel the length of the free field region is measured and used to calculate m/z of the ion.

[121] Gas chromatography coupled with mass spectrometry can often be a real-time target. The gas chromatography (GC) portion of the system separates the chemical mixture into pulses of analyte and the mass spectrometer (MS) identifies and quantifies the analyte.

[122]Tandem mass spectrometry can use combinations of the mass analyzers described above. Tandem mass spectrometers can use a first mass analyzer to separate ions according to their m/z in order to isolate an ion of interest for further analysis. The isolated ion of interest is then split into fragmented ions (called collisionally activated dissociation or collisionally induced dissociation) and the fragmented ions are analyzed by the second mass analyzer. These types of tandem mass spectrometry systems are called tandem in space systems because the two mass analyzers are separated in space, usually by a collision cell. Tandem mass spectrometry systems also include tandem in time systems in which a mass analyzer is used, however, the mass analyzer is used sequentially to isolate an ion, induce fragmentation, and then perform the analysis. mass.

[123] Tandem mass spectrometers in the space category have more than one mass analyzer. For example, a tandem quadrupole mass spectrometer system might have a first quadrupole mass filter, followed by a collision cell, followed by a second quadrupole mass filter, and then the detector. Another arrangement is to use a quadrupole mass filter for the first mass analyzer and a time-of-flight mass analyzer for the second mass analyzer, with a collision cell separating the two mass analyzers. Other tandem systems are known in the art, including quadrupole-sector, tandem-sector, and time-of-flight mass spectrometry containing electrostatic mirrors.

[124] Tandem mass spectrometers in the time category have a mass analyzer that performs different functions at different times. For example, an ion capture mass spectrometer can be used to capture ions of all m/z. A series of rf sweep functions are applied that eject ions from all m/z of the trap except m/z of the ions of interest. After the m/z of interest has been isolated, an rf pulse is applied to produce collisions with the gas molecules in the trap to induce fragmentation of the ions. Then the m/z values of the fragmented ions are measured by the mass analyzer. Ion cyclotron resonance instruments, also known as Fourier transform mass spectrometers, are an example of tandem-time systems.

[125]Various types of tandem mass spectrometry experiments can be performed by controlling the ions that are selected at each stage of the experiment. Different types of experiments use different modes of operation, sometimes called "sweeps", of the mass analyzers. In a first example, called a mass spectrum scan, the first mass analyzer and collision cell transmit all ions for mass analysis to the second mass analyzer. In a second example, called a product ion scan, the ions of interest are selected en masse in the first mass analyzer and then fragmented in the collision cell. The formed ions are then analyzed by scanning the second mass analyzer. In a third example, called a precursor ion scan, the first mass analyzer is scanned to sequentially transmit the analyzed ions en masse into the collision cell for fragmentation. The second mass analyzer mass selects the product ion of interest for transmission to the detector. Therefore, the detector signal is the result of all the precursor ions that can be fragmented into a common product ion. Other experimental formats include neutral loss scans where a constant mass difference is accounted for in the mass scans.

[126] For quantification, controls can be used that can provide a signal regarding the amount of the nucleic acid fragment, for example, that is present or introduced. A control to allow conversion of relative mass signals to absolute amounts can be obtained by adding a known amount of a mass marker or mass marker to each sample prior to detection of the nucleic acid fragments. See, for example, Ding and Cantor (2003) PNAS U S A. Mar 18; 100(6):3059-64. Any mass label that does not interfere with fragment detection can be used to normalize the mass signal. Such patterns typically have separation properties that are different from those of any of the molecular markers in the sample, and may have the same or different mass signatures.

[127] A separation step can sometimes be used to remove salts, enzymes, or other buffer components from the nucleic acid sample. Various methods well known in the art, such as chromatography, gel electrophoresis or precipitation, can be used to clean the sample. For example, size exclusion chromatography or affinity chromatography can be used to remove salt from a sample. The choice of separation method may depend on the amount of a sample. For example, when small amounts of sample are available or a miniaturized apparatus is used, a micro-affinity chromatography separation step can be used. Furthermore, whether a separation step is desired, and the choice of separation method, may depend on the detection method used. Salts can sometimes absorb laser energy in matrix-assisted laser desorption/ionization and result in lower ionization efficiency. Thus, the efficiency of matrix-assisted laser desorption/ionization and electrospray ionization can sometimes be improved by removing salts from a sample.

Electrophoresis

[128] In some embodiments, electrophoresis is used to determine the length of the nucleic acid fragment. In some embodiments, electrophoresis is not used to determine the length of the nucleic acid fragment. In some embodiments, the length of a matched probe (e.g., a matched fitted probe described herein) is determined using electrophoresis. Electrophoresis can also be used, in some embodiments, as a length-based separation method, as described herein. Any electrophoresis method known in the art, in which nucleic acids are separated by length, can be used in conjunction with the methods provided herein, which include, but are not limited to, standard electrophoretic techniques and specialized electrophoretic techniques, such as, for example, capillary electrophoresis. Examples of methods for separating nucleic acid and measuring nucleic acid fragment length using standard electrophoretic techniques can be found in the art. A non-limiting example is presented here. After running a nucleic acid sample on an agarose or polyacrylamide gel, the gel can be stained (eg, stained) with ethidium bromide (see, Sambrook and Russell, Molecular Cloning: A Laboratory Manual 3d ed., 2001 ). The presence of a band of the same size as a standard control is an indication of the presence of a particular nucleic acid sequence length, the amount of which can then be compared to the control based on the intensity of the band, thereby detecting and quantifying the length of the nucleic acid sequence of interest.

[129] In some embodiments, capillary electrophoresis is used to separate, identify, and sometimes quantify nucleic acid fragments. Capillary electrophoresis (CE) includes a family of related separation techniques that use narrow-bore fused silica capillaries to separate a complex array of large and small molecules, such as, for example, nucleic acids of varying length. High electric field strengths can be used to separate nucleic acid molecules based on differences in charge, size, and hydrophobicity. Sample introduction is accomplished by immersing the end of the capillary in a sample vial and applying pressure, vacuum, or voltage. Depending on the types of capillaries and electrolytes used, EC technology can be segmented into a number of separation techniques, any of which can be adapted to the methods provided here. Non-limiting examples of these include capillary zone electrophoresis (CZE), also known as free solution CE (FSCE), capillary isoelectric focusing (CIEF), isotachophoresis (ITP), electrokinetic chromatography (EKC), micellar electrokinetic capillary chromatography (MECC or MEKC ), microemulsion electrokinetic chromatography (MEEKC), non-aqueous capillary electrophoresis (NACE) and capillary electrochromatography (CEC).

[130]Any device, machine, or apparatus capable of performing capillary electrophoresis may be used in conjunction with the methods provided herein. In general, the main components of a capillary electrophoresis system are a sample vial, source and destination vials, a capillary, electrodes, a high voltage power supply, a detector, and a data output and handling device. The source vial, destination vial and capillary are filled with electrolyte such as an aqueous buffer solution. To introduce the sample, the capillary inlet is placed in a vial containing the sample and then returned to the source vial (sample is introduced into the capillary via capillary action, pressure or siphoning). Migration of the analytes (ie nucleic acids) is then initiated by an electric field that is applied between the source and destination vials and is supplied to the electrodes by the high voltage power supply. Ions, positive or negative, are pulled through the capillary in the same direction by electro-osmotic flow. Analytes (i.e. nucleic acids) separate as they migrate due to their electrophoretic mobility and are detected near the exit end of the capillary. The detector's output is sent to an output and data handling device, such as a computer or integrator. The data is then displayed as an electropherogram, which can communicate the detector's response as a function of time. Separated nucleic acids can appear as peaks with different migration times on an electropherogram.

[131]Separation by capillary electrophoresis can be detected by various detection devices. Most commercial systems use UV absorbance or UV-Vis as their primary mode of detection. In these systems, a section of the capillary itself is used as the detection cell. The use of in-tube detection allows detection of separate analytes without loss of resolution. In general, capillaries used in capillary electrophoresis can be coated with a polymer to increase stability. The portion of the capillary used for UV detection is often optically transparent. The path length of the detection cell in capillary electrophoresis (~50 micrometers) is much smaller than that of a traditional UV cell (~1 cm). According to Lambert Beer's law, the sensitivity of the detector is proportional to the path length of the cell. To improve sensitivity, the path length can be increased, although this may result in a loss of resolution. The capillary tube itself can be expanded at the sensing point, creating a "bubble cell" with a longer path length, or additional tubing can be added at the sensing point. Both of these methods, however, can reduce the resolution of the separation.

[132] Fluorescence detection can also be used in capillary electrophoresis for samples that naturally fluoresce or are chemically modified to contain fluorescent tags, such as, for example, nucleic acid fragments or labeled probes described herein. This high sensitivity detection mode offers improved high selectivity for these samples. The method requires the light beam to be focused on the capillary. Laser-induced fluorescence can be used in EC systems with detection limits as low as 10-18 to 1021 mol. The sensitivity of the technique is attributed to the high intensity of the incident light and the ability to accurately focus the light onto the capillary.

[133]Various capillary electrophoresis machines are known in the art and can be used in conjunction with the methods provided herein. These include, but are not limited to, CALIPER LAB CHIP GX (Caliper Life Sciences, Mountain View, CA), P/ACE 2000 Series (Beckman Coulter, Brea, CA), HP G1600A CE (Hewlett-Packard, Palo Alto, CA) , AGILENT 7100 CE (Agilent Technologies, Santa Clara, CA), and ABI PRISM genetic analyzer (Applied Biosystems, Carlsbad, CA).

Microscopy

[134] In some embodiments, the length of the nucleic acid fragment is determined using an imaging-based method, such as a microscopy method. In some embodiments, the length of a matched probe (e.g., a trimmed matched probe described herein) is determined using an imaging-based method. In some embodiments, fragment length can be determined by microscopic visualization of nucleic acid fragments (see, for example, US Patent No. 5,720,928). In some embodiments, the nucleic acid fragments are fixed to a surface (e.g., the modified glass surface) in an elongated state, stained, and visualized microscopically. Images of the fragments can be collated and processed (eg measured by length). In some embodiments, image formation and image analysis steps can be automated. Methods for directly visualizing nucleic acid fragments using microscopy are known in the art (see, for example, Li et al., (1999) Nat Genet 23(3):309-13; Aston et al., (1999) Trends Biotechnol 17(7):297-302, Aston et al., (1999) Methods Enzymol., 303:55-73, Jing et al., (1998) Proc Natl Acad Sci USA, 95(14):8046-51; and US Patent No. 5,720,928). Other microscopy methods that may be used with the methods described herein include, without limitation, scanning tunneling microscopy (STM), atomic force microscopy (ATM), scanning force microscopy (SFM), photon scanning microscopy ( PSTM), Scanning Tunneling Potentiometry (STP), Magnetic Force Microscopy (MFM), Scanning Probe Microscopy, Voltage Scanning Microscopy, Photoconductive Atomic Force Microscopy, Electrochemical Scanning Tunneling Microscopy, Electron Microscopy, Electrochemical Scanning Microscopy polarized spin tunneling (SPSTM) scanning, thermal scanning microscopy, joule expansion scanning microscopy, photothermal microspectroscopy, and the like.

[135] In some embodiments, scanning tunneling microscopy (STM) can be used to determine the length of a nucleic acid fragment. STM methods can often generate atomic-level images of molecules, such as nucleic acid fragments. STM can be performed, for example, in air, water, ultra high vacuum, various other liquid or gaseous environments, and can be performed at temperatures ranging from close to zero Kelvin (-273.15°C) to a few hundred degrees Celsius , for example. Components of an STM system typically include a scanning tip, piezoelectric controlled height, x,y scanning agent, coarse sample-to-tip control, vibration isolation system, and computer. STM methods are generally based on the concept of quantum tunneling. For example, when a conducting tip is brought close to the surface of a molecule (e.g., nucleic acid fragment), a bias (i.e., voltage difference) applied between the two can allow electrons to tunnel through the vacuum between they. The resulting tunneling current is a function of the tip position, applied voltage, and the local density of states (LDOS) of the sample. Information is acquired by monitoring the current as the tip position sweeps across the surface, and can be displayed in image form. If the tip is moved across the sample in the x-y plane, changes in surface height and state density cause changes in current. These changes can be mapped onto images. The current change with respect to position can sometimes be measured itself, or the height, z, of the tip corresponding to a constant current can be measured. These two modes are often referred to as constant height mode and constant current mode, respectively.

[136] In some embodiments, atomic force microscopy (AFM) can be used to determine the length of the nucleic acid fragment. AFM is usually a high-resolution type of nanoscale microscopy. Information about an object (eg nucleic acid fragment) is typically gathered by "feeling" the surface with a mechanical probe. Piezoelectric elements that facilitate tiny but exact and precise movements in the electronic control can facilitate very precise scanning. In some variations, electrical potentials can be swept away using conduction cantilevers. Components of an AFM system typically include a support arm with a sharp point (i.e., probe) at its end that is used to scan the surface of a sample (i.e., nucleic acid fragment). The cantilever is typically silicon or silicon nitride with a tip radius of curvature in the order of nanometers. When the tip is placed in close proximity to a sample surface, the forces between the tip and the sample lead to a deflection of the cantilever according to Hooke's law. Depending on the situation, forces that are measured in AFM include, for example, mechanical contact force, van der Waals force, capillary forces, chemical bonding, electrostatic forces, magnetic forces, Casimir forces, solvation forces, and the like. Typically, deflection is measured using a laser spot reflected from the top surface of the cantilever onto an array of photodiodes. Other methods that are used include optical interferometry, capacitive sensing or piezoresistive AFM cantilevers.

nanopore

[137] In some embodiments, the length of the nucleic acid fragment is determined using a nanopore. In some embodiments, the length of a matched probe (e.g., a trimmed matched probe described herein) is determined using a nanopore. A nanopore is a small hole or channel, typically on the order of 1 nanometer in diameter. Certain cellular transmembrane proteins can act as nanopores (eg, alpha-hemolysin). In some embodiments, nanopores can be synthesized (eg, using a silicon platform). Immersing a nanopore in a conducting fluid and applying a potential across it results in a slight electric current due to conduction of ions through the nanopores. The amount of current flowing is sensitive to the size of the nanopore. As a nucleic acid fragment passes through a nanopore, the nucleic acid molecule clogs the nanopore to some degree and generates a change in current. The duration of the current change as the nucleic acid fragment passes through the nanopore can be measured. In some embodiments, the length of the nucleic acid fragment can be determined based on this measurement.

[138] In some embodiments, the length of the nucleic acid fragment can be determined as a function of time. Longer nucleic acid fragments can sometimes take relatively longer to pass through a nanopore and shorter nucleic acid fragments can sometimes take relatively less time to pass through a nanopore. Thus, the relative length of a fragment can be determined based on the nanopore transit time, in some embodiments. In some embodiments, absolute or approximate fragment length can be determined by comparing the nanopore transit time of target fragments and/or reference fragments to the transit times for a set of standards (i.e., with known lengths).

Probes

[139] In some embodiments, the length of the fragment is determined using one or more probes. In some embodiments, probes are designed such that they hybridize to each nucleic acid of interest in a sample. For example, a probe can comprise a polynucleotide sequence that is complementary to a nucleic acid of interest, or it can comprise a series of monomers that can bind a nucleic acid of interest. Probes can be any length suitable to hybridize (e.g., fully hybridize) to one or more nucleic acid fragments of interest. For example, probes can be of any length that extends or spans beyond the length of a nucleic acid fragment to which it hybridizes. Probes can be about 100 bp or more in length. For example, probes can be at least about 200, 300, 400, 500, 600, 700, 800, 900, or 1000 bp in length.

[140] In some embodiments, probes may comprise a polynucleotide sequence that is complementary to a nucleic acid of interest and one or more polynucleotide sequences that are not complementary to a nucleic acid of interest (i.e., non-complementary sequences). Non-complementary sequences can reside, for example, at the 5' end and/or 3' end of the probe. In some embodiments, non-complementary sequences can comprise nucleotide sequences that do not exist in the organism of interest and/or sequences that are not capable of hybridizing to any human genome sequence. For example, non-complementary sequences can be derived from any non-human genome known in the art, such as, for example, non-mammalian animal genomes, plant genomes, fungal genomes, bacterial genomes, or viral genomes. In some embodiments, a non-complementary sequence is from the PhiX 174 genome. In some embodiments, a non-complementary sequence may comprise modified or synthetic nucleotides that are not capable of hybridizing to a complementary nucleotide.

[141]Probes can be designed and synthesized according to methods known in the art and described here for oligonucleotides (eg, capture oligonucleotides). Probes can also include any of the properties known in the art and described herein for oligonucleotides. The probes described herein can be designed such that they comprise nucleotides (e.g. adenine (A), thymine (T), cytosine (C), guanine (G) and uracil (U)), modified nucleotides (e.g. pseudouridine, dihydrouridine, inosine (I), and 7-methylguanosine), synthetic nucleotides, degenerate bases (e.g., 6H, 8H-3,4-hydropyrimido[4,5-c][1,2]oxazin-7-one (P), 2-amino-6-methoxyaminopurine (K), N6-methoxyadenine (Z), and hypoxanthine (I)), universal bases and/or monomers other than nucleotides, modified nucleotides or synthetic nucleotides, or combinations thereof and generally are designed such that they initially have longer lengths than the fragments they hybridize to.

[142] In some embodiments, a probe comprises a plurality of monomers that are capable of hybridizing to any of naturally occurring or modified versions of nucleotides, such as adenine (A), thymine (T), cytosine (C), guanine (G) and uracil (VC). In some embodiments, a probe comprises a plurality of monomers that are capable of hybridizing with at least three of adenine, thymine, cytosine, and guanine. For example, a probe can include a monomer species that is capable of hybridizing to A, T, and C; A, T and G; G, C and T; or G, C and A. In some embodiments, a probe comprises a plurality of monomers that are capable of hybridizing to all of adenine, thymine, cytosine and guanine. For example, a probe may include a monomer species that is capable of hybridizing to all of A, T, C, and G. In some embodiments, the hybridization conditions (e.g., stringent) can be adjusted according to the methods described herein, for example, to facilitate the hybridization of certain species of monomers with various species of nucleotides. In some embodiments, the monomers include nucleotides. In some embodiments, the monomers include naturally occurring nucleotides. In some embodiments, the monomers include modified nucleotides.

[143] In some embodiments, the monomers of a probe include inosine. Inosine is a nucleotide commonly found in tRNA and is capable, in some cases, of hybridizing to A, T and C. Example 9 here describes a method that uses poly-inosine probes for the determination of nucleic acid fragment size. In some embodiments, the polyinosine probes are hybridized to the nucleic acid fragments under low stringency or non-stringent hybridization conditions (e.g., such as low temperature and/or high salt compared to the stringent hybridization conditions herein). described). In some embodiments, the nucleic acid fragments are treated with sodium bisulfite, which causes deamination of unmethylated cytosine residues in fragments to form uracil residues. In some embodiments, sodium bisulfite-treated nucleic acid fragments are amplified (e.g., amplified PCR) prior to sodium bisulfite treatment. In some embodiments, the nucleic acid fragments are ligated to a sequence comprising a universal amplification primer site having no cytosine residues. A second, complementary strand can then be generated, for example using a universal amplification primer and extension reaction. Typically, uracil residues on the first strand generate complementary adenine residues on the second strand. In this way, a second strand having no guanine residues can be generated. Such guanine-free complementary second strands, in some cases, can hybridize with polyinosine probes under stringent hybridization conditions.

[144] In some embodiments, the monomers of a probe include universal base monomers. Typically universal base monomers are nucleobase analogues or synthetic monomers that can nonselectively hybridize to each of the native bases (eg, A, G, C, T). Thus, a probe comprising universal base monomers can sometimes hybridize to a nucleic acid fragment regardless of the nucleotide sequence. Universal bases may include, without limitation, 3-nitropyrole, 4-nitroindole, 5-nitroindole, 6-nitroindole, 3-methyl-7-propynyl isocarbostyril (PIM), 3-methyl isocarbostyril (MICS), and 5-methyl isocarbostyril ( MICS) (see, for example, Nichols et al., (1994). Nature 369, 492-493; Bergstrom et al., (1995) J. Am. Chem. Soc. 117, 1201-1209; Loakes and Brown ( 1994) Nucleic Acids Res. 22, 4039-4043; Lin and Brown (1992) Nucleic Acids Res. 20, 5149-5152; Lin and Brown (1989) Nucleic Acids Res. 17, 10383; Brown and Lin (1991) Carbohydrate Research 216, 129139;Berger et al., (2000) Nucleic Acids Res. 28(15):2911-2914).

[145] In some embodiments, the monomers of a probe include non-nucleotide monomers. In some embodiments, the monomer subunits include a synthetic polymer. In some embodiments, the monomers include pyrrolidone. Pyrrolidone is a synthetic polypyrrolidone polymer monomer and is capable, in some cases, of hybridizing with all of A, T, G and C.

[146] In some embodiments, a method for determining fragment length includes the step of contacting, under annealing conditions, nucleic acid fragments (e.g., target and/or reference fragments), with a plurality of probes that can anneal the fragments, thereby generating fragment-probe species such as, for example, probe-target species and probe-reference species. Hybridization and/or probe conditions (e.g., stringency) can be optimized to favor complete or substantially complete binding with fragment (e.g., high stringency). Full-length or substantially full-length probe-fragment hybridizations generally include duplexes in which the fragment does not comprise unhybridized portions and the probe may comprise unhybridized portions, as described in more detail below.

[147] In some embodiments, such as when the probe length is longer than the fragment length, the target probe species and/or reference probe species may comprise hybridized probe portions (i.e., probe portions single strand; see, for example, Figure 12). Unhybridized probe portions can be at either end of the probe (e.g., 3' or 5' end of a probe) or at both ends of the probe (i.e., 3' and 5' end of a probe), and can comprise any number of monomers. In some embodiments, portions of the unhybridized probe can comprise from about 1 to about 500 monomers. For example, portions of the unhybridized probe can comprise about 5, 10, 20, 30, 40, 50, 100, 200, 300 or 400 monomers.

[148] In some embodiments, portions of unhybridized probe may be removed from the target probe species and/or reference probe species, thereby generating trimmed probes. Removal of unhybridized probe portions can be accomplished by any method known in the art for cleaving and/or digesting a polymer, such as, for example, a method for cleaving or digesting a single-stranded nucleic acid. Portions of unhybridized probe can be removed from the 5' end of the probe and/or the 3' end of the probe. Such methods may include the use of chemical and/or enzymatic cleavage or digestion. In some embodiments, an enzyme capable of cleaving phosphodiester bonds between nucleotide subunits of a nucleic acid is used to remove unhybridized portions of the probe. Such enzymes may include, without limitation, nucleases (e.g., DNAase I, RNAse I), endonucleases (e.g., Mung bean nuclease, S1 nuclease, and the like), restriction nucleases, exonucleases (e.g., exonuclease I, exonuclease III, exonuclease T, exonuclease T7, lambda exonuclease, and the like), phosphodiesterases (e.g., phosphodiesterase II, calf spleen phosphodiesterase, snake venom phosphodiesterase, and the like), deoxyribonucleases (DNAse), ribonucleases (RNase), endonucleases flap, 5' nucleases, 3' nucleases, 3'-5' exonucleases, 5'-3' exonucleases and the like, or combinations thereof. Trimmed probes are generally the same or substantially the same length as the fragment with which they hybridize. Thus, determining the length of a trimmed probe here can provide a measure of the length of the corresponding nucleic acid fragment. Length of the trimmed probe can be measured using any of the methods known in the art or described herein for determining nucleic acid fragment length. In some embodiments, the probes may contain a detectable molecule or entity to facilitate detection and/or length determination (e.g., a fluorophore, radioisotope, colorimetric agent, particle, enzymes, and the like). Trimmed probe length can be evaluated with or without separating products from unhybridized portions after they are removed.

[149] In some embodiments, trimmed probes are dissociated (i.e., separated) from their corresponding nucleic acid fragments. Probes can be separated from their corresponding nucleic acid fragments using any method known in the art, including, but not limited to, heat denaturation. Trimmed probes can be distinguished from corresponding nucleic acid fragments by a method known in the art or described herein for labeling and/or isolating a species of molecule in a mixture. For example, a probe and/or nucleic acid fragment can comprise a detectable property such that a probe is distinguishable from the nucleic acid to which it hybridizes. Non-limiting examples of detectable properties include optical properties, electrical properties, magnetic properties, chemical properties, and time and/or velocity through an aperture of known size. In some embodiments, probes and nucleic acid fragments from the sample are physically separated from each other. Separation can be achieved, for example, using capture ligands, such as biotin or other affinity ligands, and capture agents, such as avidin, streptavidin, an antibody, or a receptor. A probe or nucleic acid fragment may contain a capture ligand having specific binding activity to a capture agent. For example, fragments of a nucleic acid sample can be biotinylated or linked to an affinity ligand using methods well known in the art and separated away from the probes using a pulldown assay with streptavidin coated beads, for example. In some embodiments, a capture ligand and a capture agent or some other moiety (e.g., mass labeling) can be used to add mass to the nucleic acid fragments such that they can be excluded from the mass range of the probes detected in a mass spectrometer. In some embodiments, mass is added to the probes via the monomers themselves and/or addition of a mass marker to shift the mass range away from the mass range for the nucleic acid fragments.

nucleic acid library

[150] In some embodiments, a nucleic acid library is a plurality of polynucleotide molecules (e.g., a nucleic acid sample) that are prepared, assembled, and/or modified by a specific process, non-limiting examples of which include immobilization on a solid phase (e.g., a solid support, e.g., a flow cell, a bead), enrichment, amplification, cloning, detection, and/or for nucleic acid sequencing. In certain embodiments, a nucleic acid library is prepared prior to or during a sequencing process. A nucleic acid library (e.g. sequencing library) can be prepared by a suitable method as is known in the art. A nucleic acid library can be prepared by a targeting or non-targeting preparation process.

[151] In some embodiments, a nucleic acid library can be modified to comprise a chemical moiety (e.g., a functional group) configured for immobilization of the nucleic acids on a solid support. In some embodiments, a nucleic acid library can be modified to comprise a biomolecule (e.g., a functional group) and/or a linkage pair member configured to immobilize the library on a solid support, non-limiting examples of which include globulin thyroxine binding proteins, steroid binding proteins, antibodies, antigens, haptens, enzymes, lecithins, nucleic acids, repressors, protein A, protein G, avidin, streptavidin, biotin, complement component C1q, nucleic acid binding proteins, receptors, carbohydrates, oligonucleotides, polynucleotides, complementary nucleic acid sequences, and the like, and combinations thereof. Some examples of specific binding pairs include, without limitation: an avidin moiety and a biotin moiety; an antigenic epitope and an antibody or immunologically reactive fragment thereof; an antibody and a hapten; a portion of digoxygen and an anti-digoxigen antibody; a fluorescein moiety and an anti-fluorescein antibody; an operator and a repressor; a nuclease and a nucleotide; a lecithin and a polysaccharide; a steroid and a steroid binding protein; an active compound and an active compound receptor; a hormone and a hormone receptor; an enzyme and a substrate; an immunoglobulin and protein A; an oligonucleotide or polynucleotide and its corresponding complement; the like or combinations thereof.

[152] In some embodiments a nucleic acid library can be modified to comprise one or more polynucleotides of known composition, non-limiting examples of which include an identifier (e.g., a tag, an index tag), a capture sequence, a tag, an adapter, an enzyme restriction site, a promoter, an enhancer, an origin of replication, a stem in LASSO, a complementary sequence (e.g., a primer binding site, an annealing site), a suitable integration site (eg, a transposon, a viral integration site), a modified nucleotide, the like, or combinations thereof. Polynucleotides of known sequence can be added at a suitable position, for example at the 5' end, at the 3' end or within a nucleic acid sequence. Polynucleotides of known sequence may be the same or different sequences. In some embodiments a polynucleotide of known sequence is configured to hybridize to one or more oligonucleotides immobilized on a surface (e.g., a flow cell surface). For example, a nucleic acid molecule comprising a known 5' sequence can hybridize to a first plurality of oligonucleotides while the known 3' sequence can hybridize to a second plurality of oligonucleotides. In some embodiments a nucleic acid library may comprise chromosome-specific tags, capture sequences, tags, and/or adapters. In some embodiments a nucleic acid library comprises one or more detectable tags. In some embodiments, one or more detectable tags can be incorporated into a nucleic acid library at a 5' end, at a 3' end, and/or at any nucleotide position in a nucleic acid in the library. In some embodiments a nucleic acid library comprises hybridized oligonucleotides. In certain embodiments hybridized oligonucleotides are labeled probes. In some embodiments a nucleic acid library comprises the oligonucleotide probes hybridized prior to immobilization on a solid phase.

[153] In some embodiments a polynucleotide of known sequence comprises a universal sequence. A universal sequence is a specific nucleotide acid sequence that is integrated into two or more nucleic acid molecules or two or more subsets of nucleic acid molecules where the universal sequence is the same for all subsets of molecules or molecules in which it is integrated. A universal sequence is often designed to hybridize and/or amplify a plurality of different sequences using a single universal primer, which is complementary to a universal sequence. In some embodiments two (e.g., a pair) or more universal sequences and/or universal primers are used. A universal primer often comprises a universal sequence. In some embodiments, adapters (e.g., universal adapters) comprise universal sequences. In some embodiments one or more universal sequences are used to capture, identify and/or detect multiple species or subsets of nucleic acids.

[154] In certain embodiments of preparing a nucleic acid library, (e.g., in certain sequencing synthetic procedures), the nucleic acids are of selected sizes and/or fragmented into lengths of several hundred base pairs, or least (for example, in preparation for library production). In some embodiments, library preparation is performed without fragmentation (e.g., when using ccfDNA).

[155] In certain embodiments, a linkage-based library preparation method is used (eg, ILLUMINA TRUSEQ, Illumina, San Diego CA). Ligation-based library preparation methods often make use of an adapter design (e.g., a methylated adapter) that can incorporate an index sequence in the initial ligation step and often can be used to prepare samples for single-readable sequencing, sequencing trailing edge and multiplexed sequencing. For example, nucleic acids (eg, fragmented nucleic acids or ccfDNA) are sometimes repaired at the end by a fill-in reaction, an exonuclease reaction, or a combination thereof. In some embodiments, the resulting blunt-ended repaired nucleic acid can then be extended by a single nucleotide, which is complementary to a single nucleotide overhang at the 3' end of an adapter/primer. Any nucleotide can be used to extend/trap nucleotides. In some embodiments, preparing the nucleic acid library comprises ligating an adapter oligonucleotide. Adapter oligonucleotides are usually complementary to flow cell anchors, and are sometimes used to immobilize a nucleic acid library to a solid support, such as the inner surface of a flow cell, for example. In some embodiments, an adapter oligonucleotide comprises an identifier, one or more sequencing primer hybridization sites (e.g., complementary sequences for universal sequencing primers, single-ended sequencing primers, paired-end sequencing primers, multiplexed sequencing, and the like) or combinations thereof (e.g., adapter/sequencing, adapter/tag, adapter/tag/sequencing).

[156] An identifier may be a suitable detectable tag incorporated into or attached to a nucleic acid (e.g., a polynucleotide) that allows detection and/or identification of nucleic acids that make up the identifier. In some embodiments, a tag is incorporated into or linked to a nucleic acid during a sequencing method (e.g., by a polymerase). Non-limiting examples of identifiers include nucleic acid tags, nucleic acid indices or barcodes, a radioactive tag (e.g., an isotope), metallic tag, a fluorescent tag, a chemiluminescent tag, a phosphorescent tag, a fluorophore quencher , a dye, a protein (e.g., an enzyme, an antibody or part thereof, a linker, a member of a bond pair), the like, or combinations thereof. In some embodiments, an identifier (e.g., a nucleic acid index or barcode) is a unique, known, and/or identifiable sequence of nucleotide or nucleotide analogues. In some embodiments identifiers are six or more contiguous nucleotides. A wide variety of fluorophores are available with a variety of different emission and excitation spectra. Any suitable type and/or number of fluorophores can be used as an identifier. In some modalities 1 or more, 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 20 or more, 30 or more , or 50 or more different identifiers are used in a method described herein (e.g., a nucleic acid detection and/or sequencing method). In some embodiments, one or two types of identifiers (e.g., fluorescent tags) are attached to each nucleic acid in a library. Detection and/or quantification of a tag can be performed by a suitable method, apparatus or machine, non-limiting examples of which include flow cytometry, quantitative polymerase chain reaction (qPCR), gel electrophoresis, a luminometer, a fluorimeter, a spectrophotometer, a suitable gene chip or microarray analysis, Western blot, mass spectrometry, chromatography, flow cytometry analysis, fluorescence microscopy, a suitable fluorescence or digital imaging method, confocal laser scanning microscopy, laser scanning cytometry, affinity chromatography, manual separation in batch mode, electric field suspension, a suitable nucleic acid sequencing method and/or nucleic acid sequencing apparatus, the like, and combinations thereof.

[157] In some embodiments, a transposon-based library preparation method is used (eg, EPICENTER NEXTERA, Epicenter, Madison, WI). Transposon-based methods typically use in vitro transposition to simultaneously fragment and tag DNA in a single-tube reaction (often allowing the incorporation of platform-specific tags and optional barcodes), and prepare sequencer-ready libraries.

[158] In some embodiments a nucleic acid library or parts thereof are amplified (eg, amplified by a PCR-based method). In some embodiments a sequencing method comprises amplifying a nucleic acid library. A nucleic acid library can be amplified before or after immobilization on a solid support (eg, a solid support of a flow cell). Nucleic acid amplification includes the process of amplifying or increasing the numbers of a nucleic acid template and/or its complement that are present (e.g., a nucleic acid library), by making one or more copies of the template and /or its complement. Amplification can be performed by a suitable method. A nucleic acid library can be amplified by a thermocycling method or an isothermal amplification method. In some embodiments a rolling circle amplification method is used. In some embodiments the amplification takes place on a solid support (e.g., inside a flow cell) where a nucleic acid library or part thereof is immobilized. In certain sequencing methods, a nucleic acid library is added to a flow cell and immobilized by hybridization to anchors under appropriate conditions. This type of nucleic acid amplification is often referred to as solid phase amplification. In some embodiments of solid phase amplification, all or a portion of the amplified products are synthesized by an extension starting from an immobilized primer. Solid phase amplification reactions are analogous to conventional solution phase amplifications, except that at least one of the amplification oligonucleotides (eg, primers) is immobilized on a solid support.

[159] In some embodiments solid phase amplification comprises a nucleic acid amplification reaction comprising only one species of oligonucleotide primer immobilized on a surface. In certain embodiments solid phase amplification comprises a plurality of different species of immobilized oligonucleotide primers. In some embodiments solid phase amplification may comprise a nucleic acid amplification reaction comprising one species of oligonucleotide primer immobilized on a solid surface and a second, different species of oligonucleotide primer in solution. Several different kinds of immobilized primers or solution-based primers can be used. Non-limiting examples of solid phase nucleic acid amplification reactions include interfacial amplification, bridging amplification, emulsion PCR, WildFire amplification (e.g., US patent publication US20130012399), the like, or combinations thereof.

sequencing

[160] In some embodiments, nucleic acids (e.g., nucleic acid fragments, nucleic acid sample, cell-free nucleic acid) can be sequenced. In some embodiments, a complete or substantially complete sequence is obtained, and sometimes a partial sequence is obtained. In some embodiments, a nucleic acid is not sequenced, and the sequence of a nucleic acid is not determined by a sequencing method, when performing a method described herein. In some embodiments, the length of the fragment is determined using a sequencing method. In some embodiments, the length of the fragment is determined without using a sequencing method. Sequencing, mapping, and related analytical methods are described herein and are known in the art (e.g., US Patent Application Publication US2009/0029377, incorporated herein by reference). Certain aspects of such processes are described below.

[161] In some embodiments, the length of the fragment is determined using a sequencing method. In some embodiments, fragment length is determined using a paired-end sequencing platform. Such platforms involve sequencing both ends of a nucleic acid fragment. Generally, sequences corresponding to both ends of the fragment can be mapped to a reference genome (eg, a reference human genome). In certain embodiments, both ends are sequenced to a readable length that is sufficient to map, individually for each end of the fragment, to a reference genome. Examples of readable lengths of paired-end sequence are described below. In certain embodiments, all or a portion of sequence reads can be mapped to a reference genome without mismatch. In some embodiments, each fragment (read) is mapped independently. In some embodiments, information from both reads of sequences (i.e., from each end) is factored into the mapping process. The length of a fragment can be determined, for example, by calculating the difference between the genomic coordinates assigned to each fragment (read) of paired end.

[162] In some embodiments, the length of the fragment may be determined using a sequencing process whereby a complete or substantially complete sequence of nucleotides is obtained for the fragment. Such sequencing processes include platforms that generate relatively long readable lengths (eg, Roche 454, Ion Torrent, single molecule (Pacific Biosciences), SMRT real-time technology, and the like).

[163] In some embodiments some or all of the nucleic acids in a sample are enriched and/or amplified (eg, non-specifically, eg, by a PCR-based method) before or during sequencing. In certain embodiments, specific portions of nucleic acid or subsets in a sample are enriched and/or amplified before or during sequencing. In some embodiments, a portion or a subset of a preselected set of nucleic acids is randomly sequenced. In some embodiments, the nucleic acids in a sample are not enriched and/or amplified before or during sequencing.

[164]As used herein, "reads" (i.e., "a read" or "a sequence read") are short nucleotide sequences produced by any sequencing process described or known herein. in technique. Fragments (reads) can be generated from one end of nucleic acid fragments ("single-ended fragments (reads)"), and sometimes are generated from both ends of nucleic acids (e.g., fragments paired-ended reads, double-ended fragments (reads).

[165]The length of a sequence read is often associated with the sequencing technology in particular. High-throughput methods, for example, provide fragments (reads) of sequences that can vary in size from tens to hundreds of base pairs (bp). Nanopore sequencing, for example, can provide fragments (reads) of sequences that can vary in size from tens to hundreds of thousands of base pairs. In some embodiments, the sequence reads are of medium, median, low, or absolute length from about 15 bp to about 900 bp long. In certain embodiments, sequence reads are an average, median, low, or absolute length of about 1000 bp or more.

[166] In some embodiments the nominal, low, average, or absolute length of one-ended reads is sometimes about 1 nucleotide to about 500 contiguous nucleotides, about 15 contiguous nucleotides to about 50 contiguous nucleotides, about 30 contiguous nucleotides to about 40 contiguous nucleotides, and sometimes about 35 contiguous nucleotides or about 36 contiguous nucleotides. In certain embodiments, the nominal, low, average, or absolute length of one-ended reads is from about 20 to about 30 bases, or about 24 to about 28 bases in length. In certain embodiments the nominal, low, average or absolute length of reads from one end is about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48 or 49 bases in length.

[167] In certain embodiments, nominal, low, average, or absolute length of paired-end reads is sometimes from about 10 contiguous nucleotides to about 25 contiguous nucleotides (e.g., about 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or 25 nucleotides in length), about 15 contiguous nucleotides to about 20 contiguous nucleotides, and sometimes is about of 17 contiguous nucleotides, about 18 contiguous nucleotides, about 20 contiguous nucleotides, about 25 contiguous nucleotides, about 36 contiguous nucleotides, or about 45 contiguous nucleotides.

[168]Fragments (reads) are usually representations of nucleotide sequences in a physical nucleic acid. For example, in a fragment (read) that contains an ATGC representation of a sequence, "A" represents an adenine nucleotide, "T" represents a thymine nucleotide, "G" represents a guanine nucleotide, and "C" represents a cytosine nucleotide, into a physical nucleic acid. Fragments (reads) of sequences obtained from the blood of a pregnant woman can be read from a mixture of fetal and maternal nucleic acid. A mixture of relatively short fragments (reads) can be transformed by processes described herein for representing a genomic nucleic acid present in the pregnant woman and/or fetus. A mixture of relatively short reads can be transformed into a representation of a copy number variation (for example, a maternal and/or fetal copy number variation), genetic variation, or an aneuploidy, for example. Fragments (reads) of a mixture of maternal and fetal nucleic acid can be transformed into a representation of a composite chromosome or its segment comprising features of one or both of the maternal and fetal chromosomes. In certain embodiments, "obtaining" nucleic acid sequence reads from a sample from a subject and/or "obtaining" nucleic acid sequence reads from a biological sample from one or more reference persons may involve directly sequencing nucleic acid to obtain sequence information. In some embodiments, "getting" may involve receiving sequence information obtained directly from one nucleic acid by another.

[169] In some embodiments, a fraction of the genome is sequenced, which is sometimes expressed as the amount of genome covered by the determined nucleotide sequences (eg, coverage less than 1). When a genome is sequenced with about 1-fold coverage, about 100% of the genome's nucleotide sequence is represented by fragments (reads). A genome can also be sequenced with redundancy, where a given region of the genome can be covered by two or more fragments (reads) or overlapping fragments (reads) (eg, coverage greater than 1). In some embodiments, the genome is sequenced with about 0.01 fold to about 100 fold coverage, about 0.2 fold to about 20 fold coverage, or about 0.2 fold to about 1 fold coverage ( for example coverage of about 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.1, 0.2, 0.3 , 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 30 , 40, 50, 60, 70, 80, 90 times).

[170] In some embodiments, genome coverage or sequence coverage is proportional to total sequence readable counts. For example, assays that generate and/or analyze the greatest amounts of readable sequence counts are typically associated with the greatest levels of sequence coverage. Assays that generate and/or analyze fewer amounts of readable sequence counts are typically associated with lower levels of sequence coverage. In some embodiments, sequence coverage and/or sequence readable count can be reduced without significantly decreasing the precision (e.g., sensitivity and/or specificity) of a method described herein. A significant decrease in accuracy can be a reduction in accuracy from about 1% to about 20% compared to a method that does not use reduced sequence readable count. For example, a significant decrease in accuracy could be about a 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15% or more reduction. In some embodiments, sequence coverage and/or sequence read count is reduced by about 50% or more. For example, the sequence coverage and/or sequence read count can be reduced by about 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95% or more. In some embodiments, sequence coverage and/or sequence read count is reduced by about 60% to about 85%. For example, the sequence coverage and/or sequence readable count can be reduced by about 61%, 62%, 63%, 64%, 65%, 66%, 67%, 68%, 69%, 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81%, 82%, 83% or 84%. In some embodiments, sequence coverage and/or sequence read count can be reduced by removing certain fragments (reads) of sequences. In some cases, fragments (reads) of fragment sequences longer than a particular length (eg, fragments longer than about 160 bases) are removed.

[171]In some embodiments, a subset of fragments (reads) is selected for analysis, and sometimes a certain portion of fragments (reads) are removed from analysis. Selection of a subset of fragments (reads) can, in certain cases, enrich a nucleic acid species (eg, fetal nucleic acid). Enrichment of fetal nucleic acid reads, for example, often increases the accuracy of the method described here (eg detection of fetal aneuploidy). However, selecting and removing fragments (reads) from an analysis often decreases the accuracy of the method described here (for example, due to an increase in variance). Thus, without being bound by theory, there is generally a balance between increased accuracy associated with fetal readable enrichment and decreased accuracy associated with a reduced amount of reads in methods comprising selecting and/or removing fragments ( reads) (for example, of fragments in a particular size range). In some embodiments, a method comprises selecting a subset of fragments (reads) enriched with fragments (reads) of fetal nucleic acid without significantly decreasing the accuracy of the method. Considering this apparent trade-off, it was determined, as described herein, that the use of a subset of reads of nucleotide sequences (e.g., reads of relatively short fragments), can improve or maintain the accuracy of analyses. fetal genetics. For example, in certain embodiments, about 80% or more of fragments (reads) of nucleotide sequences can be discarded, maintaining sensitivity and specificity values that are similar to values for a comparable method that does not discard such fragments (reads). of nucleotide sequences.

[172] In certain embodiments, a subset of nucleic acid fragments is selected prior to sequencing. In certain embodiments, hybridization-based techniques (e.g., using arrays of oligonucleotides) can be used to first select the nucleic acid sequences of certain chromosomes (e.g., sex chromosomes and/or a potentially aneuploidy chromosome and another chromosome (s) that is not involved in the tested aneuploidy). In some embodiments, the nucleic acid can be size-fractionated (e.g., by gel electrophoresis, size exclusion chromatography, or by a microfluidic-based approach) and, in certain cases, fetal nucleic acid can be enriched by acid selection. nucleic acid with a lower molecular weight (e.g., less than 300 base pairs, less than 200 base pairs, less than 150 base pairs, less than 100 base pairs). In some embodiments, fetal nucleic acid can be enriched by suppressing baseline maternal nucleic acid, such as by adding formaldehyde. In some embodiments, a portion or a subset of a preselected set of nucleic acid fragments is randomly sequenced. In some embodiments, the nucleic acid is amplified prior to sequencing. In some embodiments, a portion or a subset of the nucleic acid is amplified prior to sequencing.

[173] In some embodiments, a sample of nucleic acid from an individual is sequenced. In certain embodiments, the nucleic acids from each of two or more samples are sequenced, where the samples are from one individual or different individuals. In certain embodiments, nucleic acid samples from two or more biological samples are pooled, where each biological sample is from an individual, or two or more individuals, and the pool is sequenced. In the latter embodiments, a nucleic acid sample from each biological sample is often identified by one or more unique identifiers or identification tags.

[174] In some embodiments a sequencing method utilizes identifiers that allow the multiplexing of sequence reactions in a sequencing process. The greater the number of unique identifiers, the greater the number of samples and/or chromosomes for detection, for example, that can be multiplexed in a sequencing process. A sequencing process can be performed using any suitable number of unique identifiers (eg 4, 8, 12, 24, 48, 96, or more).

[175] A sequencing process sometimes makes use of a solid phase, and sometimes the solid phase comprises a flow cell into which nucleic acid from a library can be bound and reagents can be poured and contacted with the bound nucleic acid. A flow cell sometimes includes flow cell ranges, and the use of identifiers can facilitate the analysis of a number of samples in each range. A flow cell is often a solid support that can be configured to retain and/or allow orderly passage of reagent solutions along bound analytes. Flow cells are often planar in shape, optically transparent, usually in the millimeter or sub-millimeter scale, and often have channels or lanes in which the analyte/reagent interaction takes place. In some embodiments the number of samples analyzed in a given flow cell lane is dependent on the number of unique identifiers used during library preparation and/or probe design, single flow cell lane. Multiplexing using 12 tags, for example, allows simultaneous analysis of 96 samples (eg, equal to the number of wells in a 96-well microwell plate) in an 8-lane flow cell. Similarly, multiplexing using 48 tags, for example, allows simultaneous analysis of 384 samples (eg, equal to the number of wells in a 384-well microwell plate) in an 8-lane flow cell. Non-limiting examples of commercially available multiplex sequencing kits include Illumina Multiplexing Sample Preparation Oligonucleotide Kit and Multiplexing Sequencing Primers and PhiX Control Kit (e.g., Illumina catalog numbers PE-400-1001 and PE- 400-1002, respectively).

[176] Any suitable method for sequencing nucleic acids may be used, non-limiting examples of which include Maxim and Gilbert, chain termination methods, sequencing by synthesis, sequencing by ligation, sequencing by mass spectrometry, techniques based on microscopy, the like or combinations thereof. In some embodiments, a first generation technology, such as, for example, Sanger sequencing methods including automated Sanger sequencing methods, including microfluidic Sanger sequencing, can be used in a method provided herein. In some embodiments sequencing technologies that include the use of nucleic acid imaging technologies (eg, transmission electron microscopy (TEM) and atomic force microscopy (AFM)), may be used. In some embodiments, a high-throughput sequencing method is used. High-throughput sequencing methods usually involve amplified DNA templates by cloning individual DNA or molecules that are sequenced in a massively parallel fashion, sometimes within a flow cell. Next generation (e.g., second and third generation) DNA sequencing techniques, capable of sequencing DNA in a massively parallel fashion, can be used for the methods described herein and are collectively referred to herein as "massively parallel sequencing" (MPS ). In some embodiments MPS sequencing methods use a targeted approach, where specific chromosomes, genes or regions of interest are sequenced. In certain embodiments a non-targeted approach is used where most or all of the nucleic acids in a sample are randomly sequenced, amplified and/or captured.

[177] In some embodiments a targeted enrichment, amplification and/or sequencing is used. The guided approach often isolates, selects and/or enriches a subset of nucleic acids in a sample for further processing using sequence-specific oligonucleotides. In some embodiments, a library of sequence-specific oligonucleotides is used to target (e.g., to hybridize) one or more sets of nucleic acids in a sample. Sequence-specific oligonucleotides and/or primers are often selective for particular sequences (eg, unique nucleic acid sequences) present in one or more chromosomes, genes, exons, introns, and/or regulatory regions of interest. Any suitable method or combination of methods can be used for the enrichment, amplification and/or sequencing of one or more subsets of target nucleic acids. In some embodiments, target sequences are isolated and/or enriched by capture in a solid phase (e.g., a flow cell, a bead) using one or more sequence-specific anchors. In some embodiments, target sequences are enriched and/or amplified by a polymerase-based method (e.g., a PCR-based method, by any suitable polymerase-based extension), using sequence-specific primers and/or primer sets. Sequence-specific anchors can often be used as sequence-specific primers.

[178]MPS sequencing sometimes makes use of synthesis sequencing and certain imaging processes. Nucleic acid sequencing technology that may be used in a method described herein is sequencing by synthesis and reversible terminator-based sequencing (e.g., Illumina Genome Analyzer; Genome Analyzer II; HISEQ 2000; HISEQ 2500 (Illumina , San Diego, CA)). With this technology, millions of nucleic acid fragments (eg DNA) can be sequenced in parallel. In an example of this type of sequencing technology, a flow cell is used which contains an optically transparent slide with 8 individual lanes on the surfaces of which oligonucleotide anchors (eg adapter primers) are attached. The flow cell is often a solid support that can be configured to retain and/or allow the orderly passage of reagent solutions along the bound analytes. Flow cells are often planar in shape, optically transparent, usually in the millimeter or sub-millimeter scale, and often have channels or lanes in which the analyte/reagent interaction takes place.

[179]Synthetic sequencing, in some embodiments, comprises iteratively adding (e.g., by covalent addition) a nucleotide to a pre-existing primer or strand of nucleic acid in a template-directed manner. Each iterative addition of a nucleotide is detected and the process is repeated several times until a nucleic acid strand sequence is obtained. The length of a sequence obtained depends, in part, on the number of addition and detection steps that are performed. In some embodiments of synthetic sequencing, one, two, three, or more nucleotides of the same type (eg, A, G, C, or T) are added and detected in one nucleotide addition cycle. Nucleotides can be added by any suitable method (eg, enzymatically or chemically). For example, in some embodiments a polymerase or ligase adds a nucleotide to a pre-existing primer or strand of nucleic acid in a template-directed manner. In some embodiments of synthetic sequencing, different types of nucleotides, nucleotide analogues, and/or identifiers are used. In some embodiments, reversible terminators and/or removable (eg, cleavable) identifiers are used. In some embodiments fluorescent labeled nucleotides and/or nucleotide analogues are used. In certain embodiments sequencing per synthesis comprises a cleavage (e.g., the cleavage and removal of a tag) and/or a washing step. In some embodiments the addition of one or more nucleotides is detected by a suitable method described herein or known in the art, non-limiting examples of which include any suitable imaging apparatus or machine, a suitable camera, a digital camera, a charge-coupled device (CCD) based imaging apparatus (e.g. a CCD camera), a CMOS (complementary silicon metal oxide) based imaging apparatus (e.g. a CMOS camera), a photodiode (e.g. e.g., a photomultiplier tube), electron microscopy, a field-effect transistor (e.g., a DNA field-effect transistor), an ISFET ion sensor (e.g., a CHEMFET sensor), the like, or combinations thereof . Other sequencing methods that can be used to perform the methods here include digital PCR and hybridization sequencing.

[180]Other sequencing methods that can be used to perform the methods here include digital PCR and hybridization sequencing. Digital polymerase chain reaction (digital PCR or dPCR) can be used to directly identify and quantify nucleic acids in a sample. Digital PCR can be performed on an emulsion, in some embodiments. For example, individual nucleic acids are separated, for example in a microfluidic chamber device, and each nucleic acid is individually amplified by PCR. Nucleic acids that can be separated such that there is not more than one nucleic acid per well. In some embodiments, different probes can be used to distinguish different alleles (eg, fetal alleles and maternal alleles). Alleles can be enumerated to determine copy number.

[181] In certain embodiments, hybridization sequencing can be used. The method involves contacting a plurality of polynucleotide sequences with a plurality of polynucleotide probes, each plurality of polynucleotide probes optionally being attached to a substrate. The substrate can be a flat surface with an array of known nucleotide sequences, in some embodiments. The template hybridization pattern can be used to determine the polynucleotide sequences present in the sample. In some embodiments, each of the probes is attached to a bead, for example a magnetic bead or the like. Hybridization on the beads can be identified and used to identify the plurality of polynucleotide sequences in the sample.

[182] In some embodiments, nanopore sequencing can be used in a method described here. Nanopore sequencing is a single molecule sequencing technology whereby a single nucleic acid molecule (eg DNA) is directly sequenced as it passes through a nanopore.

[183] A method, system, or MPS technology platform suitable for carrying out the methods described herein can be used to obtain sequencing nucleic acid fragments (reads). Non-limiting examples of MPS platforms include Illumina/Solex/HiSeq (e.g., Illumina Genome Analyzer; Genome Analyzer II; HISEQ 2000; HISEQ), SOLID, Roche/454, PACBIO and/or SMRT, Helicos True Single Molecule sequencing, Ion Torrent, and Ion semiconductor-based sequencing (e.g., as developed by Life Technologies), WildFire, 5500, 5500xI W, and/or 5500xI W genetic analyzer-based technologies (e.g., as developed and sold by Life Technologies) , US Patent Publication No. US20130012399); Polony sequencing, pyrosequencing, massively parallel signature sequencing (MPSS), RNA polymerase (RNAP) sequencing, LaserGen systems and methods, nanopore-based platforms, chemical sensitive field-effect transistor array (CHEMFET), electron microscopy-based sequencing ( e.g. as developed by ZS Genetics, Halcyon Molecular), nanoball sequencing.

[184] In some embodiments, chromosome-specific sequencing is performed. In some embodiments, chromosome-specific sequencing is performed using DANSR (Digital Selected Region Analysis). Digital analysis of selected regions allows simultaneous quantification of hundreds of sites by cfDNA-dependent catenation of oligonucleotides from two specific sites via an intervening "linker" oligonucleotide to form a PCR template. In some embodiments, chromosome-specific sequencing is accomplished by generating a library enriched for chromosome-specific sequences. In some embodiments, sequence reads are obtained only for a selected set of chromosomes. In some embodiments, sequence reads are obtained only for chromosomes 21, 18, and 13.

Mapping reads

[185]Fragments (reads) of sequences can be mapped, and the number of fragments (reads) mapping to a specific region of nucleic acid (eg, a chromosome, portion, or segment thereof) is referred to as counts. Any suitable mapping method (eg process, algorithm, program, software, module, the like or combination thereof) may be used. Certain aspects of the process mapping processes are described below.

[186]Mapping nucleotide sequence reads (i.e., sequence information from a fragment whose physical genomic position is unknown) can be accomplished in a number of ways, and often comprise alignment of the fragments (reads). ) of sequences obtained with a corresponding sequence in a reference genome. In such alignments, sequence reads are usually aligned with a reference sequence and those that are aligned as being "mapped", "a mapped sequence read" or "a mapped read". In certain embodiments, a fragment (read) of mapped sequence is referred to as a "hit", or "counts". In some embodiments, fragments (reads) of mapped sequences are grouped together according to various parameters and assigned to particular portions, which are discussed in more detail below.

[187] As used herein, the terms "aligned", "aligned" or "aligned" refer to two or more nucleic acid sequences that can be identified as a match (e.g., 100% identity) or partial match . Alignments can be done manually or by a computer (e.g., a software, program, module, or algorithm), non-limiting examples of which include the Efficient Local Alignment of Nucleotide Data (ELAND) computer program distributed as part of the analysis plumber genomics of Illumina. Sequence fragment (read) alignment can be 100% sequence compatibility. In some cases, an alignment is less than a 100% sequence match (ie, non-perfect match, partial match, partial alignment). In some embodiments an alignment is about 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80%, 79%, 78%, 77%, 76% or 75% compatibility. In some embodiments, an alignment comprises a mismatch. In some embodiments, an alignment comprises 1, 2, 3, 4 or 5 mismatches. Two or more sequences can be aligned using either strand. In certain embodiments, a nucleic acid sequence is aligned with the reverse complement of another nucleic acid sequence.

[188]Various computational methods can be used to map each read from a sequence to a chunk. Non-limiting examples of computer algorithms that may be used to align sequences include, without limitation, BLAST, BLITZ, FASTA, BOWTIE 1, BOWTIE 2, ELAND, MAQ, Probematch, SOAP or SEQMAP, or variations thereof, or combinations thereof . In some embodiments, fragments (reads) of sequences can be aligned with sequences in a reference genome. In some embodiments, sequence reads can be found and/or aligned to sequences in nucleic acid databases known in the art, including, for example, GenBank, dbEST, dbSTS, EMBL (European Molecular Biology Laboratory) and DDBJ (DNA Databank of Japan). BLAST or similar tools can be used to search the identified sequences against a sequence database. Search hits can then be used to order identified sequences into suitable portions (described below), for example.

[189] In some embodiments, a read may exclusively or non-exclusively map portions of a reference genome. A fragment (read) is considered to be "uniquely mapped" if it aligns with a single sequence in the reference genome. The fragment (read) is considered as "not uniquely mapped" if it aligns with two or more sequences in the reference genome. In some embodiments, fragments (reads) not exclusively mapped are eliminated from further analysis (eg, quantification). A certain small degree of mismatch (0-1) may be allowed to account for single nucleotide polymorphisms that may exist between the reference genome and the individual sample reads being mapped, in certain embodiments.

[190]In some embodiments, no degree of mismatch is allowed for a fragment (read) mapped to a reference sequence.

[191] As used herein, the term "reference genome" may refer to any particular known genome, sequenced or characterized, or partial or complete, of any organism or virus that can be used to reference identified sequences of a subject. For example, a reference genome used for humans as well as many other organisms can be found at the National Center for Biotechnology Information at www.ncbi.nlm.nih.gov. A "genome" refers to the genetic information of an organism or virus, expressed in nucleic acid sequences. As used herein, a reference sequence or reference genome is often a assembled or partially assembled genomic sequence from an individual or multiple individuals. In some embodiments, a reference genome is a assembled or partially assembled genomic sequence from one or more human individuals. In some embodiments, a reference genome comprises sequences assigned to chromosomes.

[192] In certain embodiments, where a nucleic acid in the sample is from a pregnant woman, a reference sequence sometimes is not from the fetus, the mother of the fetus, or the father of the fetus, and is referred to herein as a " external reference." A maternal reference can be prepared and used in some modalities. When prepared from a pregnant woman reference ("maternal reference sequence") based on an external reference, fragments (reads) of pregnant woman DNA that contain substantially no fetal DNA are often mapped to the external reference sequence. and gathered. In certain embodiments the external reference is DNA from an individual having substantially the same ethnicity as the pregnant woman. A maternal reference sequence may not completely cover the maternal genomic DNA (for example, it may cover about 50%, 60%, 70%, 80%, 90% or more of the maternal genomic DNA), and the maternal reference may not perfectly match the maternal genomic DNA sequence (for example, the maternal reference sequence may include multiple mismatches).

[193] In certain embodiments, mappability is evaluated for a region of the genome (eg, portion, genomic portion, portion). Mapability is the ability to unequivocally align a fragment (read) of nucleotide sequence to a portion of a reference genome, typically up to a specified number of mismatches, including, for example, 0, 1, 2 or more mismatches. For a given genomic region, the expected mappability can be estimated using a sliding window approach of a predefined readable length and the average of the resulting readable level mappability values. Genomic regions containing stretches of single nucleotide sequence sometimes have a high mappability value.

Portions

[194] In some embodiments, fragments (reads) of mapped sequences (ie, sequence tags) are grouped together according to various parameters and assigned to particular portions (eg, portions of a genome reference). Often, individual mapped sequence reads can be used to identify a portion (eg, the presence, absence, or amount of a portion) present in a sample. In some embodiments, the amount of a portion is indicative of the amount of a larger sequence (e.g., a chromosome) in the sample. The term "portion" may also be referred to herein as a "genomic section", "portion" (bin), "region", "partition", "portion of a reference genome", "portion of a chromosome" or "portion genomics". In some embodiments a chunk is an entire chromosome, a segment of a chromosome, a segment of a reference genome, a segment spanning multiple chromosomes, multiple chromosome segments, and/or combinations thereof. In some embodiments, a portion is predefined based on specific parameters (eg indicators). In some embodiments, a portion is arbitrarily or non-arbitrarily defined based on the partition of a genome (e.g., partitioned by size, GC content, contiguous regions, contiguous regions of an arbitrarily defined size, and the like). In some embodiments portions are chosen from discrete genomic portions (bins), genomic portions (bins) having continuous sequences of predetermined length, portion (bin) of variable size, point based views of a flattened coverage map, and /or a combination thereof.

[195] In some embodiments, a portion is delineated based on one or more parameters that include, for example, length or a particular sequence characteristic or characteristics. Portions may be selected, filtered and/or removed from consideration using any suitable criteria known in the art or described herein. In some embodiments, a portion is based on a particular length of genomic sequence. In some embodiments, a method may include analyzing multiple fragments (reads) of sequences mapped to a plurality of portions. Portions can be approximately the same length or portions can be of different lengths. In some embodiments, the portions are of approximately equal length. In some embodiments portions of different lengths are adjusted or weighed. In some embodiments a portion is from about 10 kilobases (kb) to about 20 kb, from about 10 kb to about 100 kb, from about 20 kb to about 80 kb, from about 30 kb to about 70 kb , from about 40 kb to about 60 kb. In some embodiments a chunk is about 10 kb, 20 kb, 30 kb, 40 kb, 50 kb or about 60 kb in length. A portion is not limited to contiguous runs of the sequence. Thus, the portions may be comprised of contiguous and/or non-contiguous sequences. A portion is not limited to a single chromosome. In some embodiments, a portion includes all or part of one chromosome or all or part of two or more chromosomes. In some embodiments, the portions may encompass one, two, or more entire chromosomes. In addition, portions may span jointed or disjointed regions of multiple chromosomes.

[196] In some embodiments, the moieties may be particular chromosome segments on a chromosome of interest, such as, for example, a chromosome, where a genetic variation is evaluated (e.g., an aneuploidy of chromosomes 13, 18, and/or 21, or a sex chromosome). A portion may also be a pathogenic genome (for example, bacteria, fungi, or viruses) or a fragment thereof. The moieties can be genes, gene fragments, regulatory sequences, introns, exons and the like.

[197] In some embodiments, a genome (eg, human genome) is divided into portions based on the information content of particular regions. In some embodiments, partitioning a genome can eliminate similar regions (e.g., identical or homologous regions or sequences) throughout the genome and keep only the unique regions. Regions removed during partition can be within a single chromosome or can span multiple chromosomes. In some embodiments a partitioned genome is trimmed down and optimized for faster alignment, often allowing focus on uniquely identifiable sequences.

[198] In some embodiments, partitioning can decrease weight of similar regions. A process of decreasing the weight of a serving is discussed in more detail below.

[199] In some embodiments, partitioning a genome into regions that transcend chromosomes may be based on gaining information produced in the context of classification. For example, information content can be quantified using a p-value profile measuring the significance of particular genomic loci to distinguish between groups of confirmed normal and abnormal individuals (eg, euploid and trisomy individuals, respectively). In some embodiments, partitioning a genome into regions that transcend chromosomes can be based on any other criteria, such as, for example, speed/convenience, while align marks, GC content (e.g., high or low GC), GC content uniformity, other measures of sequence content (eg, fraction of individual nucleotides, fraction of pyrimidines or purines, fraction of natural versus unnatural nucleic acids, fraction of methylated nucleotides, and CpG content), methylation state, duplex melting temperature, sequencing or PCR responsiveness, uncertainty value assigned to individual portions of a reference genome, and/or a targeted search for particular traits.

[200]A "segment" of a chromosome is usually part of a chromosome, and typically is a different part of a chromosome than a portion. A segment of a chromosome is sometimes in a different region of a chromosome than a chunk, sometimes it does not share a polynucleotide with a chunk, and sometimes it includes a polynucleotide that is on a chunk. A segment of a chromosome often contains a greater number of nucleotides than a chunk (for example, a segment sometimes includes a chunk), and sometimes a segment of a chromosome contains fewer nucleotides than a chunk. portion of (e.g. a segment is sometimes within a portion).

Filtering and/or selection of portions

[201] Portions are sometimes processed (eg, normalized, filtered, selected, the like, or combinations thereof) according to one or more characteristics, parameters, criteria, and/or methods described herein or known in the art. Portions may be processed by any suitable method and according to any suitable parameters. Non-limiting examples of characteristics and/or parameters that can be used to filter and/or select portions include counts, coverage, mappability, variability, an uncertainty level, guanine-cytosine (GC) content, CCF fragment length, and/or or length reading (e.g., a fragment length ratio (FLR), a fetal ratio statistic (FRS)), DNasel sensitivity, methylation status, acetylation, histone distribution, chromatin structure, the like, or combinations of the same. Portions may be filtered and/or selected according to a suitable characteristic or parameter that correlates to a characteristic or parameter listed or described herein. Portions can be filtered and/or selected according to characteristics or parameters that are specific to a portion (e.g., as determined by a single portion according to multiple samples) and/or characteristics or parameters that are specific to a sample (eg, as determined for multiple servings within a sample). In some embodiments portions are filtered and/or removed according to relatively low mappability, relatively high variability, a high level of uncertainty, relatively long CCF fragment lengths (e.g., low FRS, low FLR), relatively high percentage of sequences repetitive, high GC, low GC, low counts, zero counts, high counts, similar, or combinations thereof. In some embodiments portions (eg, a subset of portions) are selected according to the appropriate level of mapability, variability, level of uncertainty, fraction of repetitive sequences, counts, GC content, like, or combination thereof. In some embodiments chunks (eg, a subset of chunks) are selected according to relatively short lengths of CCF fragments (eg, high FRS, high FRS). Counts and/or reads mapped to chunks are sometimes handled (eg, normalized) before and/or after filtering or selecting chunks (eg, a subset of chunks). In some embodiments, counts and/or reads mapped to chunks are not processed before and/or after filtering or selecting chunks (eg, a subset of chunks).

[202]Sequence reads from any suitable number of samples can be used to identify a subset of portions that satisfy one or more of the criteria, parameters and/or characteristics described herein. Fragments (reads) of sequences from a pool of samples from several pregnant women are sometimes used. One or more samples from each of several pregnant women may be targeted (e.g. 1 to about 20 samples from each pregnant woman (e.g. about 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, or 19 samples)), and an adequate number of pregnant women can be targeted (e.g., about 2 to about 10,000 pregnant women (e.g., about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 350, 400, 500, 600, 700, 800, 900, 1000, 2000, 3000, 4000 ,5000 ,6000, 7000, 8000, 9000 pregnant women)). In some embodiments, fragments (reads) of sequences from the same test sample(s) from the same pregnant woman are mapped to portions in the reference genome and are used to generate the subset of portions.

[203] It has been observed that circulating cell-free nucleic acid fragments (CCF fragments) obtained from a pregnant woman generally comprise nucleic acid fragments derived from fetal cells (i.e., fetal fragments) and nucleic acid fragments from maternal cells (i.e., maternal fragments). Sequence reads derived from CCF fragments derived from a fetus are referred to herein as "fetal reads". Sequence reads derived from CCF fragments from the genome of a pregnant woman (eg, a mother) carrying a fetus are referred to herein as "maternal reads". CCF fragments from which fetal reads are obtained are referred to herein as fetal templates and CCF fragments from which maternal reads are obtained are referred to herein as maternal templates.

[204] It has also been observed that in CCF fragments, fetal fragments are generally relatively short (eg, about 200 base pairs in length or less) and that maternal fragments include both relatively short fragments and relatively longer fragments. A subset of portions that map to a significant amount of relatively short fragment reads can be selected and/or identified. Without being bound by theory, it is expected that the reads mapped to such portions will be enriched with fetal reads, which may improve the accuracy of a fetal genetic analysis (e.g., detecting the presence or absence of a genetic variation of the fetus (eg fetal chromosome aneuploidy (eg T21, T18 and/or T13))).

[205] A significant number of reads is often overlooked, however, when a fetal genetic analysis is based on a subset of reads. The selection of a subset of fragments (reads) mapped to a selected subset of portions, and the removal of fragments (reads) in unselected portions, for a fetal genetic analysis can decrease the precision of the genetic analysis, due to the increase in variance , for example. In some embodiments, about 30% to about 70% (e.g., about 35%, 40%, 45%, 50%, 55%, 60%, or 65%) of sequencing reads obtained from a subject or sample map is removed from consideration under selection of a subset of portions for a genetic analysis of the fetus. In certain embodiments, about 30% to about 70% (e.g., about 35%, 40%, 45%, 50%, 55%, 60%, or 65%) of sequencing reads obtained from a subject or sample map to a subset of portions used for a fetal genetic analysis.

[206]Thus, without being bound by theory, there is generally a balance between increased accuracy associated with fetal reading enrichment and decreased accuracy associated with a decrease in the amount of readable data (e.g., removal of portions and/or fragments (reads)) for fetal genetic analysis. In some embodiments, a method comprises selecting an enriched subset of portions for fetal nucleic acid reads (e.g., fetal reads) that improves, or does not significantly decrease, the accuracy of a genetic analysis of the fetus. fetus. Considering this apparent trade-off, it was determined, as described here, that the use of a subset of portions, to which a significant amount of fragments (reads) of relatively short fragments are mapped, can improve the accuracy of fetal genetic analyses.

[207] In some embodiments a subset of chunks is selected according to fragments (reads) of CCF fragments, where the fragments (reads) mapped to a chunk have a length less than a length of the selected fragment. Sometimes a subset of servings is selected by filtering out servings that do not meet these filtering criteria. In certain embodiments, a subset of the portions is selected according to the amount of fragments (reads) derived from relatively short CCF fragments (e.g., about 200 base pairs or less) that map a portion. Any suitable method can be used to identify and/or select portions where a significant amount of fragments (reads) of CCF fragments with a length less than a selected fragment length (e.g. a length of the first selected fragment) are mapped . CCF fragments having a length less than a selected fragment length are often relatively short CCF fragments, and sometimes the chosen fragment length is about 200 base pairs or less (e.g., CCF fragments that are about 190, 180, 170, 160, 150, 140, 130, 120, 110, 100, 90, or 80 bases in length). The length of a CCF fragment can be determined (eg, deduced or inferred) by mapping two or more fragments (reads) derived from the fragment (eg, a paired end tip) to a reference genome. For end-paired reads derived from a CCF fragment, for example, reads can be mapped to a reference genome, the length of the genomic sequence between the mapped reads can be determined, and the total of the two readable lengths and the length of the genomic sequence between the fragments (reads) is equal to the length of the CCF fragment.

[208]The length of a CCF fragment template is sometimes determined directly from the length of a fragment (read) derived from the fragment (eg, fragment (read) from one end).

[209] In some embodiments, a subset of chunks, in which a significant amount of fragments (reads) of CCF fragments with a length less than a selected fragment length, is selected and/or identified according to whether the readable amount mapped of CCF fragments with a length less than a length of the first selected fragment is greater than the mapped readable amount of CCF fragments having a length less than a length of the second selected fragment. In certain embodiments, a subset of chunks, wherein a significant amount of fragments (reads) of CCF fragments having a length less than a selected fragment length, is selected and/or identified according to whether the mapped readable amount of fragments CCF having a length less than a selected first fragment length of a chunk is greater than the low, medium, or median amount of mapped fragments (reads) of fragments CCF having a length less than a selected second fragment length for chunks analyzed. In some embodiments, a subset of chunks, wherein a significant amount of fragments (reads) of CCF fragments having a length less than a selected fragment length, is selected and/or identified based on a ratio of fragment length (FLR) determined for each portion. A "fragment length ratio" is also referred to herein as a fetal ratio statistic (FRS).

[210] In certain embodiments, a FLR is determined, in part, according to the amount of fragments (reads) mapped to a portion of CCF fragments having a length less than a selected fragment length. In some embodiments, a FLR value is often a ratio of X to Y, where X is the number of fragments (reads) derived from CCF fragments having a length less than a length of the first selected fragment, and Y is the number of fragments (reads) derived from CCF fragments having a length smaller than a length of the second selected fragment. A length of the first selected fragment is often selected independently of the length of a second selected fragment, and vice versa, and the length of the second selected fragment is typically greater than the length of the first selected fragment. A length of the first fragment can be selected from about 200 bases or less to about 30 bases or less. In some embodiments, a chosen first fragment length is about 200, 190, 180, 170, 160, 155, 150, 145, 140, 135, 130, 125, 120, 115, 110, 105, 100, 95, 90, 85, 80, 75, 70, 65, 60, 55 or 50 bases. In some embodiments, a chosen first fragment length is from about 170 to about 130 bases, and sometimes is from about 160 to about 140 bases. In some embodiments, a chosen second fragment length is from about 2000 bases to about 200 bases. In certain embodiments a selected second fragment length is about 1000, 950, 800, 850, 800, 750, 700, 650, 600, 550, 500, 450, 400, 350, 300, 250 bases. In some embodiments, the length of the first selected fragment is from about 140 to about 160 bases (e.g., about 150 bases) and the length of the second selected fragment is from about 500 to about 700 bases (e.g., about 600 bases). In some embodiments, the length of the first selected fragment is about 150 bases and the length of the second selected fragment is about 600 bases.

[211]In some embodiments a FLR is an average, arithmetic mean, or median of several FLR values. For example, sometimes a FLR for a given portion is an average, arithmetic mean, or median of the FLR values for (i) two or more test samples, (ii) two or more subjects, or (iii) two or more test samples. test and two or more subjects. In certain embodiments, a mean, arithmetic mean, or median FLR is derived from the FLR values for two or more portions of a genome, chromosome, or segment thereof. In some embodiments, an FLR mean, arithmetic mean, or median is associated with an uncertainty (eg, standard deviation, median absolute deviation).

[212]In some embodiments, a subset of portions is selected and/or identified according to one or more FLR values (eg, a comparison of one or more FLR values). In certain embodiments a subset of portions is selected and/or identified according to an FLR and a threshold (e.g., a comparison of a FLR and a threshold). In certain embodiments, a mean, arithmetic mean, or FLR median derived from a given portion is compared with a mean, arithmetic mean, or FLR median derived from two or more portions of a genome, chromosome, or segment thereof. For example, sometimes an average FLR for a given serving is compared to a median FLR for a given serving. In certain embodiments a portion is selected and/or identified according to a determined mean, arithmetic mean, or median FLR for a portion and a determined mean, arithmetic mean, or median FLR for a set of portions (e.g., portion of a genome, chromosome, or segment thereof). In some embodiments, an average FLR for a portion is below a certain threshold determined according to a median FLR and the portion is removed from consideration (eg, in a fetal genetic analysis). In some embodiments, a mean, arithmetic mean, or median FLR for a portion is above a certain threshold determined according to a mean, arithmetic mean, or median FLR for a genome, chromosome, or segment thereof, and the portion is selected and/or added to a subset of portions for consideration (eg, when determining the presence or absence of a genetic variation). In some embodiments, an FLR for a serving is equal to or greater than about 0.15 to about 0.30 (e.g., about 0.16, 0.17, 0.18, 0.19, 0 ,20, 0.21, 0.21, 0.22, 0.23, 0.24, 0.25, 0.26, 0.27, 0.28, 0.29) and the portion is selected for consideration (eg, added to or incorporated into a subset of portions for a fetal genetic analysis). In some embodiments, an FLR for a portion is equal to or less than about 0.20 to about 0.10 (e.g., about 0.19, 0.18, 0.17, 0.16, 0 .15, 0.14, 0.13, 0.12, 0.11) and the portion is removed from consideration (e.g. filtered).

[213]Portions in a subset are sometimes selected and/or identified according, in part, to whether a significant amount of fragments (reads) of CCF fragments having a length less than a selected fragment length are mapped to a portion (e.g. according to an FLR). In some embodiments, portions in a subset may be selected and/or identified according to one or more characteristics or criteria in addition to the amount of fragments (reads) of mapped sequences of fragment lengths less than a selected fragment length. In some embodiments, a subset of chunks is selected and/or identified according to whether a significant amount of fragments (reads) of CCF fragments having a length less than a selected fragment length are mapped to a chunk (e.g., from according to an FLR) and one or more other characteristics. Non-limiting examples of other characteristics include the number of exons in, and/or CG content, a genome, chromosome or segment thereof, and/or one or more of the portions. Therefore, sometimes portions selected and/or identified according to whether a significant amount of fragments (reads) of CCF fragments having a length less than a selected fragment length are mapped to a portion (e.g., according to a FLR) to a subset, are further selected or removed according to the GC content of the portion and/or the number of exons in the portion. In some embodiments, a moiety is not selected or removed from consideration (e.g., filtered), if the GC content and/or the number of exons in the moiety do not correlate with a FLR for the moiety.

[214] In some embodiments, a subset of portions consists of, essentially consists of, or comprises portions that satisfy one or more particular criteria described herein (e.g., portions are characterized by a FLR equal to or greater than a certain value). In certain embodiments, portions that do not meet a criterion are included in a subset of portions that meet the criteria, for example, to increase the accuracy of a fetal genetic analysis. In certain embodiments, in a subset of portions that "consists essentially of" portions selected according to a criterion (e.g., an FLR equal to or greater than a certain value), of about 90% or more (e.g., about 91%, 91%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or more) of the servings meet the criterion, and about 10% or less (for example, about 9 %, 8%, 7%, 6%, 5%, 4%, 3%, 2%, about 1% or less) of the servings do not meet the criteria.

[215]Portions may be selected and/or filtered by any suitable method. In some embodiments portions are selected according to visual inspection of data, graphs, diagrams and/or tables. In certain embodiments portions are selected and/or filtered (e.g., in part) by a system or a machine comprising one or more microprocessors and memory. In some embodiments portions are selected and/or filtered (e.g., in part) by a non-transient computer-readable storage medium with an executable program stored thereon, where the program instructs a microprocessor to perform the selection and/or filtering. .

[216] A subset of portions selected by methods described here can be used for a fetal genetic analysis in different ways. In certain embodiments, fragments (reads) derived from a sample are used in a mapping process using a preselected subset of portions described herein, and not using all or most portions in a reference genome. Those reads that map to the preselected subset of portions are often used in further steps of a fetal genetic analysis, and fragments (reads) that do not map to the preselected subset of portions are not often used in further steps of a fetal genetic analysis. fetal genetic analysis (eg, reads that do not map are removed or filtered).

[217] In some embodiments, fragments (reads) of sequences derived from a sample are mapped to all or most portions of a reference genome, and a preselected subset of portions described herein are further selected. Fragments (reads) of a selected subset of portions are often used in additional steps of a fetal genetic analysis. In the latter embodiments, reads from unselected portions are often not used in later steps of a fetal genetic analysis (eg, reads from unselected portions are removed or filtered).

counts

[218]Sequence reads that are mapped or partitioned based on a selected feature or variable can be quantified to determine the number of reads that are mapped to one or more portions (e.g., portion of a genome of reference), in some modalities. In certain embodiments, the number of fragments (reads) of sequences that are mapped to a portion are count links (e.g., counts). Often, a count is associated with a portion. In certain embodiments counts for two or more portions (eg, a set of portions) are mathematically manipulated (eg, averaged, summed, normalized, similar, or a combination thereof). In some embodiments a count is determined from a portion or all of the reads of sequences mapped to (i.e., associated with) a portion. In certain embodiments, a count is determined from a predefined subset of mapped sequence reads. Predefined subsets of mapped sequence reads can be defined or selected using any suitable resource or variable. In some embodiments, predefined subsets of mapped sequence reads may include from 1 to n sequence reads, where n represents a number equal to the sum of all sequence reads generated from a subject. sample or test sample from the reference subject.

[219] In certain embodiments a count is derived from sequence reads that are processed or manipulated by a suitable method, operation, or mathematical process known in the art. A count (eg counts) may be determined by a suitable mathematical method, operation or process. In certain embodiments a count is derived from sequence reads associated with a portion in which part or all of the sequence reads are weighed, removed, filtered, normalized, adjusted, averaged, derived as an average , added or subtracted or processed by a combination thereof. In some embodiments, a count is derived from raw sequence reads and/or filtered sequence reads. In certain embodiments a counts value is determined by a mathematical process. In certain embodiments a counts value is an average, arithmetic mean, or sum of sequence reads mapped to a portion. Often, a count is an average number of counts. In some embodiments, a count is associated with an uncertainty value.

[220]In some embodiments, counts may be manipulated or transformed (eg, normalized, combined, summed, filtered, selected, averaged, derived as an average, similar, or a combination thereof). In some embodiments, the counts can be transformed to produce normalized counts. Counts can be processed (e.g., normalized) by a method known in the art and/or as described herein (e.g., normalization by portions, normalization by GC content, and linear and nonlinear least squares regression, GC LOESS , LOWESS, PERUN, RM, GCRM, cQn and/or combinations thereof).

[221]Counts (eg, raw, filtered, and/or normalized counts) can be processed and normalized to one or more levels. Levels and profiles are described in more detail later on. In certain embodiments the counts may be processed and/or normalized to a reference level. Reference levels are discussed later. Counts processed according to a level (eg, counts processed) can be associated with an uncertainty value (eg, a calculated variance, an error, standard deviation, Z-score, p-value, mean absolute deviation, etc.). In some embodiments an uncertainty value defines a range above and below a level. A deviation value may be used in place of an uncertainty value, and non-limiting examples of measures of deviation include standard deviation, mean absolute deviation, median absolute deviation, standard score (e.g., Z score, Z score, normal score, standardized variable) and the like.

[222]Counts are often obtained from a nucleic acid sample from a pregnant woman carrying a fetus. Counts of fragments (reads) of nucleic acid sequences mapped to one or more portions are often representative counts of both the fetus and the fetus's mother (eg, a pregnant female subject). In certain embodiments some of the counts mapped to a portion are from a fetal genome and some of the counts mapped to the same portion are from a maternal genome.

Data Processing and Normalization

[223]Fragments (reads) of mapped sequences that have been counted are referred to herein as raw data, as the data represent unmanipulated counts (eg, raw counts). In some embodiments, sequence readable data in a dataset may be further processed (e.g., mathematically and/or statistically manipulated) and/or displayed to facilitate providing a result. In certain embodiments, datasets, including larger datasets, may benefit from pre-processing to facilitate further analysis. Pre-processing of datasets sometimes involves removing redundant and/or non-informative portions or portions of a reference genome (e.g. portions of a reference genome with non-informative data, redundant mapped reads, portions with zero median counts, over-represented or over-represented sequences). Without being bound by theory, data processing and/or pre-processing can (i) remove noisy data, (ii) remove non-informative data, (iii) remove redundant data, (iv) reduce complexity of data sets. larger data, and/or (v) facilitate transformation of data from one form into one or more other forms. The terms "pre-treatment" and "treatment" when used with respect to data or datasets are collectively referred to herein as "processing". Processing can make data more conducive to further analysis, and can generate a result in some modalities. In some embodiments one or more or all of the processing methods (e.g., normalization methods, portion filtering, mapping, validation, the like, or combinations thereof) are performed by a processor, a microprocessor, a computer, in conjunction with memory and/or a microprocessor-controlled machine.

[224]The term "noisy data", as used herein, refers to (a) data that has significant variation between data points when analyzed or plotted, (b) data that has a significant standard deviation (for example, greater than 3 standard deviations), (c) data that have a significant standard error of the mean, similar, and combinations of the above. Noisy data sometimes occurs due to the quantity and/or quality of the starting material (e.g., nucleic acid sample), and sometimes occurs as part of processes for preparing or replicating DNA used to generate fragments (reads) of sequences. In certain embodiments, noise results from certain sequences being over-represented when prepared using PCR-based methods. Methods described here can reduce or eliminate the contribution of noisy data and therefore reduce the effect of noisy data on the given result.

[225] The terms "non-informative data", "non-informative portions of a reference genome", and "non-informative portions" as used herein refer to portions, or data derived therefrom, having a numerical value that is significantly different of a predetermined threshold value or that falls outside the predetermined cut-off range of values. The terms "threshold" and "threshold value" here refer to any number that is calculated using a qualifying dataset and serves as a diagnostic threshold of a genetic variation (e.g., a copy number variation, an aneuploidy , a chromosomal aberration, and the like). In certain embodiments a threshold is exceeded by the results obtained by methods described herein and a subject is diagnosed with a genetic variation (eg, trisomy 21). A threshold value or range of frequency values is calculated mathematically and/or statistically by manipulating readable sequence data (e.g., from a reference and/or subject), in some embodiments, and in certain embodiments, readable sequence data manipulated to generate a threshold value or range of values is string readable data (eg from a reference and/or subject). In some embodiments, an uncertainty value is determined. An uncertainty value is usually a measure of variance or error and can be any suitable measure of variance or error. In some embodiments an uncertainty value is a standard deviation, standard error, calculated variance, p-value, or mean absolute deviation (MAD). In some embodiments an uncertainty value can be calculated according to a formula in Example 4.

[226]Any suitable process may be used for processing the datasets described herein. Non-limiting examples of procedures suitable for use in processing data sets include filtering, normalizing, weighting, monitoring peak heights, monitoring peak areas, monitoring peak boundaries, determining area ratios, mathematically processing the data , statistical processing of data, application of statistical algorithms, analysis with fixed variables, analysis with optimized variables, plotting data to identify patterns or trends for further processing, the similar and combinations of the above. In some embodiments, datasets are processed based on various characteristics (e.g., GC content, redundant mapped reads, centromere regions, telomere regions, the like, and combinations thereof) and/or variables (e.g., (e.g., fetal gender, maternal age, maternal ploidy, percent fetal nucleic acid contribution, peer or combinations thereof). In certain embodiments, processing datasets as described herein can reduce the complexity and/or dimensionality of large and/or complex datasets. A non-limiting example of a complex dataset includes readable sequence data generated from one or more test subjects and a plurality of reference subjects of different ages and ethnic backgrounds. In some embodiments, the data sets may include from thousands to millions of sequence reads for each test and/or reference subject.

[227]Data processing can be performed in any number of steps, in certain embodiments. For example, data may be processed using only a single processing procedure in some embodiments, and in certain embodiments data may be processed using 1 or more, 5 or more, 10 or more, or 20 or more processing steps (e.g. , 1 or more processing steps, 2 or more processing steps, 3 or more processing steps, 4 or more processing steps, 5 or more processing steps, 6 or more processing steps, 7 or more processing steps, 8 or more processing steps, 9 or more processing steps, 10 or more processing steps, 11 or more processing steps, 12 or more processing steps, 13 or more processing steps, 14 or more processing steps, 15 or more processing steps, 16 or more processing steps, 17 or more processing steps, 18 or more processing steps, 19 or more processing steps, or 20 or more processing steps). In some embodiments, the processing steps may be the same step repeated two or more times (e.g., filtering two or more times, normalizing two or more times), and in certain embodiments, the processing steps may be two or more different processing steps (eg, filtering, normalizing; monitoring the height of peaks and edges; filtering, normalizing, normalizing to statistical manipulation, reference to determine p-values, and the like), performed simultaneously or sequentially. In some embodiments, any suitable number and/or combination of the same or different processing steps can be used to process readable sequence data to facilitate providing a result. In certain embodiments, processing data sets by the criteria described herein can reduce the complexity and/or dimensionality of a data set.

[228] In some embodiments, one or more processing steps may comprise one or more filtering steps. The term "filtering", as used herein, refers to removing portions or portions of a reference genome from consideration. Portions of a reference genome may be selected for removal according to any suitable criteria, including, but not limited to, redundant data (e.g., redundant or overlapping mapped fragments (reads), non-informative data (e.g., portions of a reference genome with zero median scores), portions of a reference genome with sequences overrepresented or underrepresented, noisy data, similar, or combinations of the foregoing. A filtering process often involves removing one or more portions of a reference genome from consideration and subtracting the counts in one or more portions of a selected reference genome to remove the counts counted or summed for the portions of a reference genome. reference genome, chromosome or chromosomes or genome under consideration. In some embodiments, portions of a reference genome may be deleted successively (e.g., one at a time to allow assessment of the effect of removing each individual portion), and in certain embodiments all portions of a reference genome are labeled for removal can be removed at the same time. In some embodiments, portions of a reference genome characterized by variation above or below a certain level are removed, which is sometimes referred to herein as filtering out "noisy" portions of a reference genome. In certain embodiments, a filtering process comprises obtaining data points from a dataset that deviate from the average profile level of a portion, a chromosome, or a segment of a chromosome by a predetermined multiple of the profile variance. , and in certain embodiments, a filtering process comprises removing data points from a data set that do not deviate from the average profile level of a portion, a chromosome, or segment of a chromosome by a predetermined multiple of the profile. variance. In some embodiments, a filtering process is used to reduce the number of candidate portions of a reference genome analyzed for the presence or absence of a genetic variation. Reducing the number of candidate portions of a reference genome analyzed for the presence or absence of genetic variation (e.g., microdeletion, micro-duplication) often reduces the complexity and/or dimensionality of a dataset, and sometimes increases the speed of looking for and/or identifying genetic variations and/or genetic aberrations by two or more orders of magnitude.

[229] In some embodiments one or more processing steps may comprise one or more normalization steps. Normalization can be performed by a suitable method described herein or known in the art. In certain modalities normalization comprises adjusting the values measured on different scales to a fictitious common scale. In certain embodiments normalization comprises a sophisticated mathematical adjustment to bring probability distributions of fitted values into alignment. In some embodiments normalization comprises aligning distributions on a normal distribution. In certain embodiments normalization comprises mathematical adjustments that allow the comparison of corresponding normalized values for different data sets in a way that eliminates the effects of certain gross influences (eg errors and anomalies). In certain embodiments normalization comprises scaling. Normalization sometimes involves dividing one or more sets of data by a predetermined variable or formula. Examples of normalization methods include portion normalization, GC content normalization, linear and nonlinear least squares regression, LOESS, GC LOESS, LOWESS (locally weighted scatterplot), PERUN, repeated masking (RM), GC-normalization and repeated masking (GCRM), conditional quantile normalization (cQn) and/or combinations thereof. In some embodiments, determining the presence or absence of a genetic variation (e.g., an aneuploidy) utilizes a normalization method (e.g., portion normalization, GC content normalization, linear and nonlinear least squares regression, LOESS , GC LOESS, LOWESS (locally weighted scatterplot), PERUN, repeated masking (RM), GC-normalization and repeated masking (GCRM), cQn, an art-known normalization method, and/or a combination thereof). In some modalities counts are normalized.

[230]For example, LOESS is an art-known regression modeling method that combines several regression models into a meta-model based on nearest neighbor k. LOESS is sometimes referred to as locally weighted polynomial regression. GC LOESS, in some embodiments, applies a LOESS model to relationships between fragment counts (eg, sequence reads, counts) and GC composition for portions of a reference genome. Plotting a smooth curve through a set of data points using LOESS is sometimes LOESS curve binding, particularly when each smoothed value is given by a weighted least squares regression on the range of values of the scatterplot scatter plot variable criterion. X axis. For each point in a dataset, the LOESS method fits a low-degree polynomial to a subset of the data, with explanatory variable values close to the point whose response is being evaluated. The polynomial is fitted using weighted least squares, giving more weight to points close to the point whose response is being estimated and less weight to points farther away. The value of the regression function for a point is then obtained by evaluating the local polynomial using the explanatory variable values that that data point has. The LOESS fit is sometimes considered complete after the regression function values have been computed for each of the data points. Many of the details of this method, such as the degree of the polynomial model and the weights, are flexible.

[231]Any suitable number of normalizations may be used. In some embodiments, data sets may be normalized 1 or more, 5 or more, 10 or more, or even 20 or more times. Data sets can be normalized to values (eg, normalized value) representative of any suitable characteristic or variable (eg, sample data, reference data, or both). Non-limiting examples of the types of data normalizations that can be used include normalizing raw count data for one or more selected test or reference portions to the total number of counts mapped to the entire chromosome or genome in which the portion or selected sections are mapped; normalize data from raw counts for one or more portions to average reference counts for one or more portions or the chromosome to which a selected portion or segment is mapped; normalize raw count data to previously normalized data or derivatives thereof; and normalizing previously normalized data to one or more other predetermined normalization variables. Normalizing a dataset sometimes has the effect of isolating statistical error, depending on the characteristic or property selected as the predetermined normalization variable. Normalizing a dataset sometimes also allows the comparison of data characteristics of data with different scales, bringing the data to a common scale (eg, predetermined normalization variable). In some embodiments, one or more normalizations to a statistically derived value can be used to minimize data differences and de-emphasize remote data. Normalizing portions or portions of a reference genome with respect to a normalization value is sometimes referred to as "portioning normalization".

[232] In certain embodiments, a processing step comprising normalization includes normalizing to a static window, and in some embodiments, a processing step comprising normalization includes normalizing to a moving or sliding window. The term "window" as used herein refers to one or more portions chosen for analysis and sometimes used as a reference for comparison (eg, used for normalization and/or other mathematical or statistical manipulation). The term "normalize to a static window" as used herein refers to a normalization process using one or more portions selected for comparison purposes between a test subject and reference subject data set. In some embodiments the selected portions are used to generate a profile. A static window usually includes a predetermined set of portions that do not change during manipulations and/or analysis. The terms "normalize a sliding window" and "normalize a sliding window", as used herein, refer to normalizations performed to localized portions of the genome region (e.g., immediate genetic surroundings, adjacent portions or sections, and the like) of a selected test portion, where one or more selected test portions are normalized into portions immediately surrounding the selected test portion. In certain embodiments, selected portions are used to generate a profile. A moving or sliding window normalization often includes repeatedly moving or sliding an adjacent test portion, and normalizing the newly selected test portion to portions immediately surrounding or adjacent to the newly selected test portion, where adjacent windows have one or more common portions. In certain embodiments, a plurality of selected test pieces and/or chromosomes can be analyzed by a sliding window process.

[233] In some embodiments, normalizing a sliding or moving window can generate one or more values, where each value represents the normalization of a different set of reference portion selections from different regions of a genome (e.g., chromosome) . In certain embodiments, one or more values generated are cumulative sums (eg, a numerically integral estimate of the profile of normalized counts over the selected portion, domain (eg, a portion of the chromosome), or chromosome). The values generated by the sliding or moving window process can be used to generate a profile and make it easier to arrive at a result. In some embodiments, cumulative sums of one or more moieties may be presented as a function of genomic position. Sliding or moving window analysis is sometimes used to analyze a genome for the presence or absence of microdeletions and/or microinserts. In certain embodiments, display of cumulative sums of one or more portions is used to identify the presence or absence of regions of genetic variation (eg, microdeletions, microduplications). In some embodiments, sliding or sliding window analysis is used to identify regions of the genome that contain microdeletions and in certain embodiments, sliding or sliding window analysis is used to identify regions of the genome that contain microduplications.

[234] A particularly useful normalization methodology for reducing the error associated with nucleic acid flags is referred to herein as Parameterized Error Removal and Unbiased Normalization (PERUN) described herein and, for example, in US patent application no. 13/669,136 and international patent application n. PCT/US12/59123 (W02013/052913) all content of which is incorporated herein by reference, including all text, tables, equations and drawings. PERUN's methodology can be applied to a variety of nucleic acid indicators (eg, fragments (reads) of nucleic acid sequences) for the purpose of reducing the error effects that confound predictions based on such indicators.

[235]For example, PERUN's methodology can be applied to fragments (reads) of nucleic acid sequences from a sample and reduces the effects of error that can impair genomic section-level determinations. Such an application is useful for using fragments (reads) of nucleic acid sequences to determine the presence or absence of a genetic variation in a subject manifested as a variable level of a nucleotide sequence (e.g., a portion, a level genomic section). Non-limiting examples of portion variations are chromosomal aneuploidies (eg, trisomy 21, trisomy 18, trisomy 13) and presence or absence of a sex chromosome (eg, XX in females versus XY in males). A trisomy of an autosome (eg, a chromosome other than the sex chromosome) may be referred to as an affected autosome. Other non-limiting examples of variations in genomic section levels include microdeletions, microinsertions, duplications, and mosaicism.

[236] In certain applications, the PERUN methodology can reduce systematic and/or experimental bias by normalizing nucleic acid indicators to particular genomic groups, the latter of which are referred to as moieties. Portions include a suitable set of nucleic acid indicators, a non-limiting example of which includes a length of contiguous nucleotides, which is referred to herein as a genomic section or portion of a reference genome. Portions (bins) may include other nucleic acid indicators as described herein. In such applications, the PERUN methodology generally normalizes nucleic acid indicators in particular bins across a number of samples in three dimensions.

[237] In certain applications, PERUN methodology can reduce experimental and/or systematic bias by normalizing nucleic acid indicators (eg, reads, counts) mapped to specific segments (eg, portions) of a genome of reference. In such applications, the PERUN methodology generally normalizes counts of nucleic acid reads in particular portions of a reference genome across a number of samples in three dimensions. A detailed description of PERUN and its applications is provided in the examples section here, in international patent application n. PCT/US12/59123 (W02013/052913) and in patent application publication no. US20130085681, all content of which is incorporated herein by reference, including all text, tables, equations and drawings.

[238] In certain embodiments, PERUN's methodology includes calculating a genomic section level for portions of a reference genome from (a) readable counts of sequence mapped to a portion of a reference genome for a test sample, (b) the experimental bias (eg, the GC bias) for the test sample, and (c) one or more fit parameters (eg, the fit estimates) for an adjusted relationship between (i) the bias experimental for a portion of a reference genome to which sequence reads are mapped and (ii) counts of sequence reads mapped to the portion. The experimental bias for each of the portions of a reference genome can be determined across several samples according to a ratio adjusted for each sample between (i) the counts of fragments (reads) of mapped sequences for each of the portions of a genome reference genome, and (ii) a mapping feature for each of the portions of a reference genome. This adjusted relationship for each sample can be pieced together from multiple samples in three dimensions. The set can be ordered according to experimental bias in certain embodiments, although the PERUN methodology can be practiced without ordering the set according to experimental bias. The fitted relationship for each sample and The fitted relationship for each portion of the reference genome can be independently fitted to a linear function or non-linear function by a suitable fitting method (e.g., a fit model) known in the art. Non-limiting examples of a suitable model that can be used to fit a relationship include a linear regression model, simple regression model, ordinary least squares regression model, multiple regression model, general multiple regression model, polynomial regression model , general linear model, generalized linear model, discrete choice regression model, logistic regression model, multinomial logit model, mixed logit model, probit model, multinomial probit model, ordered logit model, ordered probit model, Poisson model, multivariate response regression model, multilevel model, fixed effects model, random effects model, mixed model, nonlinear regression model, nonparametric model, semiparametric model, robust model, quantile model, isotonic model , principal components model, minimum angle model, local model, segmented model, and variable error models.

[239]In some embodiments, a relationship is a geometric and/or graphic relationship. The terms "relationship" and "relationship" as used herein are synonymous. In some embodiments a relationship is a mathematical relationship. In some embodiments, a relationship is plotted. In some embodiments a relationship is a linear relationship. In certain embodiments a relationship is a non-linear relationship. In certain embodiments a relationship is a regression (for example, a regression line). A regression can be a linear regression or a non-linear regression. A relationship can be expressed by a mathematical equation. Often, a relationship is defined, in part, by one or more constants and/or one or more variables. The relationship can be generated by a method known in the art. A relationship in two dimensions can be generated by one or more samples, in certain embodiments, and an error-proving, or possibly error-proving, variable can be selected for one or more dimensions. A relationship can be generated, for example, using graphing software known in the art which plots a graph using values of two or more variables supplied by a user. A relationship can be fitted using a method known in the art (e.g. by performing a regression, a regression analysis, eg by a suitable regression program, eg software). Certain relationships can be fitted by linear regression, and linear regression can generate a slope value and intercept value. Certain relationships are sometimes non-linear and can be fitted using a non-linear function such as a parabolic, hyperbolic or exponential function (eg a quadratic function) for example.

[240]In PERUN's methodology, one or more of the fitted relationships can be linear. For an analysis of circulating cell-free nucleic acid from pregnant women, where the experimental bias is the GC bias and the mapping characteristic is the GC content, a one-sample adjusted relationship between (i) the fragment counts (reads ) of mapped sequences for each cell-free portion, and (ii) the GC content for each of the portions of a reference genome, can be linear. For this last adjusted relationship, the slope pertains to the GC trend and a coefficient of the GC trend can be determined for each sample when the adjusted relationships are pooled across multiple samples. In such embodiments, the fit relationship for multiple samples and a portion between (i) the GC bias coefficient for the portion, and (ii) counts of fragments (reads) of mapped sequences to portion, may also be linear. An intercept and slope can be obtained from the last fitted relationship. In such applications, the slope addresses the sample-specific bias based on GC content and the intercept addresses a part-specific attenuation pattern common to all samples. PERUN methodology can significantly reduce this sample-specific bias and portion-specific attenuation when calculating genomic section levels to provide a result (eg, presence or absence of genetic variation, determination of fetal sex).

[241]In some embodiments normalization by PERUN makes use of fit to a linear function and is described by Equation A, Equation B or a derivation thereof. Equation A: M = LI + GS (A) Equation B: L = (M-GS)/I (B)

[242] In some embodiments L is a normalized PERUN level or profile. In some embodiments L is the intended yield of the PERUN normalization procedure. In certain embodiments L is portion specific. In some embodiments L is determined according to various portions of a reference genome and represents a normalized PERUN level of a genome, chromosome, or portions of the same segment. The L level is often used for further analysis (eg, to determine Z values, maternal deletions/duplications, fetal deletions/duplications, fetal gender, sexual aneuploidies, and so on). The normalization method according to Equation B is named Parameterized Error Removal and Unbiased Normalization (PERUN).

[243] In some embodiments G is a coefficient of the CG trend measured using a linear model, LOESS, or any equivalent approach. In some embodiments G is a slope. In some embodiments the GC G trend coefficient is evaluated as the regression slope for M counts (eg, raw counts) for portion i and the GC content of portion i determined from a reference genome. In some embodiments G represents secondary information, extracted from M and determined according to a relationship. In some embodiments G represents a relationship for a set of portion-specific counts and a set of portion-specific GC content values for a sample (eg, a test sample). In some embodiments the moiety-specific GC content is derived from a reference genome. In some embodiments, portion-specific GC content is derived from observed or measured (e.g., measured from sample) GC content. A GC trend coefficient is often determined for each sample in a sample group and is usually determined for a test sample. A GC trend coefficient is often sample-specific. In some embodiments a CG trend coefficient is a constant. In certain embodiments a GC trend coefficient, once derived from a sample, does not change.

[244]In some embodiments I is an intercept and S is a slope derived from a linear relationship. In some embodiments the relationship from which I and S are derived is different from the relationship from which G is derived. In some embodiments the relationship from which I and S are derived is fixed for a given experimental setup. In some embodiments I and S are derived from linear relationship according to counts (eg, raw counts) and a CG trend coefficient according to several samples. In some embodiments I and S are derived independently of the test sample. In some embodiments I and S are derived from multiple samples. I and S are often portion specific. In some embodiments, I and S are determined with the assumption that L = 1 for all portions of a reference genome in euploid samples. In some embodiments a linear relationship is determined for the euploid samples and specific I and S values for a selected portion (assuming L = 1) are determined. In certain embodiments, the same procedure is applied to all portions of a reference genome in a human genome and a set of I intercepts and S slopes is determined for each portion.

[245] In some embodiments a cross-validation approach is applied. Cross-validation is sometimes referred to as rotation estimation. In some embodiments a cross-validation approach is applied to accurately assess how a predictive model (eg such as PERUN) will perform in practice using a test sample. In some embodiments a cross-validation cycle comprises partitioning a data sample into complementary subsets, performing a cross-validation analysis on one subset (e.g., sometimes referred to as a training set), and validating the analysis using another subset. (e.g. sometimes linking a validation set or test set). In certain embodiments, multiple cross-validation cycles are performed using different partitions and/or different subsets. Non-limiting examples of cross-validation approaches include leaving one out, sliding edges, by K times, by 2 times, repeated random subsampling, like, or combinations thereof. In some embodiments a cross-validation randomly selects from a working set containing 90% of a set of samples comprising known euploid fetuses and uses that subset to train a model. In certain embodiments, the random selection is repeated 100 times, producing a set of 100 slopes and 100 intercepts for each portion.

[246] In some embodiments the value of M is a measured value derived from a test sample. In some embodiments M is raw count measured for a serving. In some embodiments, where I and S values are available for a portion, measurement M is determined from a test sample and is used to determine the normalized level of PERUN L for a genome, chromosome, segment or portion of the even according to Equation B.

[247] Thus, applying the PERUN methodology to sequence reads across multiple samples in parallel can significantly reduce the error caused by (i) sample-specific experimental bias (e.g., GC bias) and (ii) specific attenuation of the common portion for samples. Other methods in which each of these two error sources are addressed separately or in series are often not able to reduce these as effectively as the PERUN methodology. Without being bound by theory, PERUN's methodology is expected to reduce error more effectively in part because its additive processes generally do not spread as much as the multiplicative processes generally used in other normalization approaches (e.g., GC- LOESS).

[248]Additional normalization and statistical techniques can be used in combination with the PERUN methodology. An additional process can be applied before, after and/or during the use of the PERUN methodology. Non-limiting examples of processes that can be used in combination with the PERUN methodology are described below.

[249] In some embodiments, a secondary normalization or adjustment of a genomic section level for GC content can be used in conjunction with the PERUN methodology. A suitable GC content adjustment or normalization procedure can be used (eg, GC-LOESS, GCRM). In certain embodiments, a particular sample may be selected and/or identified for application of an additional GC normalization process. For example, applying the PERUN methodology can determine the GC trend for each sample, and a sample associated with a GC trend above a certain threshold can be selected for an additional GC normalization process. In such embodiments, a predetermined threshold level may be used to select such samples for further GC normalization.

[250] In certain embodiments, a portion filtering or weighting process may be used in conjunction with the PERUN methodology. A suitable portion weighting or filtering process can be used, non-limiting examples which are described herein, in International Patent Application No. PCT/US12/59123 (W02013/052913) and US patent application publication no. US20130085681, all contents of which are incorporated herein by reference, including all text, tables, equations and drawings. In some embodiments, a normalization technique that reduces the error associated with maternal insertions, duplications, and/or deletions (eg, variations in fetal and/or maternal copy number), is used in conjunction with the PERUN methodology.

[251]Genomic section levels calculated by the PERUN methodology can be used directly to provide a result. In some embodiments, genomic section levels can be used directly to provide a result for samples where the fetal fraction is from about 2% to about 6% or greater (e.g., fetal fraction of about 4% or bigger). Genomic section levels calculated by the PERUN methodology are sometimes further processed to provide a result. In some embodiments, genomic section levels are standardized. In certain embodiments, the sum, arithmetic mean, or median of genomic cross-section levels calculated for a test portion (e.g., chromosome 21) may be divided by the sum, arithmetic mean, or median of genomic cross-section levels calculated for portions other than the test portion (eg, autosomes other than chromosome 21), to generate an experimental genomic section level. An experimental genomic section level or a raw genomic section section can be used as part of a normalization analysis, such as calculating a Z score or Z score. A Z score can be generated for a sample by subtracting a level from expected genomic section from an experimental genomic section level or raw genomic section level and the resulting value can be divided by a standard deviation for the samples. Resulting Z scores can be distributed to different samples and analyzed, or can be related to other variables, such as fetal fraction and others, and analyzed to provide a result, in certain modalities.

[252]As noted here, PERUN's methodology is not limited to normalization according to GC trend and GC content per se, and can be used to reduce the error associated with other sources of error. A non-limiting example of a source of non-GC content trending is mapping capability. When normalization parameters other than grade and GC bias are addressed, one or more of the fitted relationships may be nonlinear (eg, hyperbolic, exponential). Where the experimental bias is determined from a non-linear relationship, for example, an estimate of the experimental bias curvature can be analyzed in some embodiments.

[253] The PERUN methodology can be applied to a variety of nucleic acid indicators. Non-limiting examples of nucleic acid markers are fragments (reads) of nucleic acid sequences and levels of nucleic acid at a particular location on a microarray. Non-limiting examples of sequence reads include those obtained from circulating cell-free DNA, circulating cell-free RNA, cellular DNA and cellular RNA. The PERUN methodology can be applied to fragments (reads) of sequences mapped to suitable reference sequences, such as genomic reference DNA, cellular reference RNA (e.g., transcriptome), and portions thereof (e.g., part(s) ) of a genomic complement of DNA or RNA transcriptome, part(s) of a chromosome).

[254]Thus, in certain embodiments, cellular nucleic acid (e.g., DNA or RNA) can serve as a nucleic acid indicator. Cellular nucleic acid reads mapped to portions of the reference genome can be normalized using PERUN methodology. Cellular nucleic acid bound to a particular protein are sometimes referred to as chromatin immunoprecipitation (ChIP) processes. ChIP-enriched nucleic acid is nucleic acid in association with cellular protein, such as DNA or RNA, for example. ChIP-enriched nucleic acid reads can be obtained using technology known in the art. ChIP-enriched nucleic acid reads can be mapped to one or more portions of a reference genome, and the results can be normalized using PERUN methodology to provide a result.

[255] In certain embodiments, cellular RNA reads can serve as nucleic acid markers. Cellular RNA reads can be mapped to reference RNA snippets and normalized using PERUN methodology to provide results. Known sequences of cellular RNA, referred to as a transcriptome, or its segment, can be used as a reference to which RNA reads from a sample can be mapped. Sample RNA reads can be obtained using technology known in the art. Results from RNA reads mapped to a reference can be normalized using PERUN methodology to provide a result.

[256] In some embodiments, microarray nucleic acid levels can serve as indicators of nucleic acids. Nucleic acid levels across samples for a particular address, or nucleic acid hybridization, in an array can be analyzed using PERUN methodology, thereby normalizing nucleic acid indicators provided by microarray analysis. In this way, a particular address or hybridization of nucleic acid in a microarray is analogous to a portion of fragments (reads) of mapped nucleic acid sequences, and PERUN methodology can be used to normalize the microarray data to provide a better result.

[257] In some embodiments, a processing step comprises a weighting coefficient. The terms "weighing", "weighting" or "weighting function" or grammatical derivatives or equivalents thereof, as used herein, refer to a mathematical manipulation of a portion or the entirety of a data set, sometimes used to change the influence of certain characteristics or variables of the dataset with respect to other characteristics or variables of the dataset (for example, increasing or decreasing the importance and/or contribution of data contained in one or more portions or portions of a reference genome, based on the quality and usefulness of the data on the portion or portions of a selected reference genome). The weighting function can be used to increase the influence of data with a relatively small measurement variance, and/or to decrease the influence of data with a relatively large measurement variance, in some embodiments. For example, portions of a reference genome with low quality or underrepresented sequence data can be "weighted down" to minimize the influence of a dataset, while selected portions of a reference genome can be "weighted down" to minimize the influence of a dataset. increased weight" to increase influence on a dataset. A non-limiting example of a weighting function is [1/(standard deviation) 2]. A weighting step is sometimes performed in substantially the same way as a normalization step. In some embodiments, a data set is divided by a predetermined variable (eg weight variable). A predetermined variable (e.g., the minimized target function, Phi) is often selected to weight different parts of a dataset differently (e.g., increasing the influence of certain types of data, while decreasing the influence of other types). of data).

[258] In certain embodiments, a processing step may comprise one or more mathematical and/or statistical manipulations. Any suitable mathematical and/or statistical manipulation, alone or in combination, may be used to analyze and/or manipulate a dataset described herein. Any number of suitable mathematical and/or statistical manipulations can be used. In some embodiments, a data set can be mathematically and/or statistically manipulated 1 or more, 5 or more, or 10 or more 20 or more times. Non-limiting examples of mathematical and statistical manipulations that can be used include addition, subtraction, multiplication, division, algebraic functions, least squares estimators, curve fitting, differential equations, rational polynomials, couple polynomials, orthogonal polynomials, Z scores, values p values, chi values, phi values, peak level analysis, determining peak margin locations, calculating peak area proportions, mean chromosome level analysis, calculating mean absolute deviation, sum of squared residuals, mean, standard deviation, standard error, like, or combinations thereof. Mathematical and/or statistical manipulation can be performed on all or a portion of the readable data of the sequences, or their processed products. Non-limiting examples of dataset variables or characteristics that can be statistically manipulated include raw counts, filtered counts, normalized counts, peak heights, peak widths, peak areas, peak margins, side tolerances, p values, levels medians, mean levels, distribution of counts within a genomic region, relative representation of nucleic acid species, like or combinations thereof.

[259]In some embodiments, a processing step may comprise the use of one or more statistical algorithms. Any suitable statistical algorithm, alone or in combination, can be used to analyze and/or manipulate a dataset described herein. Any suitable number of statistical algorithms can be used. In some embodiments, a dataset can be analyzed using 1 or more, 5 or more, 10 or more, or 20 or more statistical algorithms. Examples of statistical algorithms suitable for use with the methods described here include decision trees, null counts, multiple comparisons, comprehensive testing, Behrens-Fisher problem, bootstrapping, Fisher's method for combining independent tests of significance, null hypothesis, error type I error, type II error, exact test, one-sample Z-test, two-sample Z-test, one-sample t-test, paired t-test, combined two-sample t-test having equal variances, unpaired t-test two-samples having unequal variances, one-proportion z-test, two-proportions pooled z-test, two-proportions uncomposite z-test, one-sample chi-square test, two-sample F-test for equality of variances, confidence interval , credibility interval, significance, meta-analysis, simple linear regression, robust linear regression, the similar or combinations of the above. Non-limiting examples of dataset variables or characteristics that can be analyzed by statistical algorithms include raw counts, filtered counts, normalized counts, peak heights, peak widths, peak margins, lateral tolerances, p values, median levels , average levels, distribution of counts within a genomic region, relative representation of nucleic acid species, like or combinations thereof.

[260]In certain embodiments, a dataset may be analyzed using multiple statistical algorithms (e.g., 2 or more) (e.g., least squares regression, principal component analysis, linear discriminant analysis, quadratic discriminant, bagging, neural networks, support vector machine models, random forests, classification tree models, K nearest neighbors, logistic regression and/or loss of smoothing) and/or mathematical and/or statistical manipulations (e.g. example, herein referred to as manipulations). Using various manipulations can generate an N-dimensional space that can be used to provide a result, in some embodiments. In certain embodiments, analyzing a dataset through the use of various manipulations can reduce the complexity and/or dimensionality of the dataset. For example, using various manipulations on a reference dataset can generate an N-dimensional space (e.g., probability plot) that can be used to represent the presence or absence of genetic variation, depending on the genetic state. from the reference samples (eg, positive or negative for a selected genetic variation). Analysis of test samples using a substantially similar set of manipulations can be used to generate an N-dimensional space for each of the test samples. The complexity and/or dimensionality of a test subject dataset is sometimes reduced to a single N-dimensional value or point that can be readily compared to the N-dimensional space generated from the reference data. Test sample data that fall within the N-dimensional space populated by the reference subject data is indicative of a genetic state substantially similar to that of the reference subjects. Test sample data that fall outside the N-dimensional space populated by the reference data is indicative of a genetic state substantially different from that of the reference subject. In some embodiments, the references are euploid or otherwise do not have a genetic variation or medical condition.

[261]After datasets have been counted, optionally filtered, and normalized, the processed datasets may be further manipulated by one or more filtering and/or normalization procedures, in some embodiments. A dataset that has been further manipulated by one or more filtering and/or normalizing procedures can be used to generate a profile, in certain embodiments. One or more filtering and/or normalization procedures can sometimes reduce the complexity and/or dimensionality of the data set, in some embodiments. A result can be provided based on a dataset of reduced complexity and/or dimensionality.

[262] In some embodiments portions may be filtered according to a measure of error (e.g., standard deviation, standard error, calculated variance, p-value, mean absolute error (MAE), absolute median deviation, and/or mean deviation absolute (MAD). In certain embodiments an error measure refers to count variability. In some embodiments portions are filtered according to the variability of counts. In certain embodiments count variability is a measure of error determined for the counts mapped to a portion (i.e. portion) of a reference genome for multiple samples (e.g. multiple samples obtained from multiple subjects, e.g. 50 or more, 100 or more, 500 or more 1000 or more, 5000 or more or 10,000 or more subjects.) In some embodiments portions with a variability of counts above a predetermined upper range are filtered (eg excluded from consideration). greater than about 50, about 52, about 54, about 56, about 58, about 60, about 62, about 64, about 66, about 68, about 70, about 72, about 74 or equal to or greater than about 76. In some embodiments portions with a variability of counts below a predetermined lower range are filtered (eg, excluded from consideration). In some embodiments a predetermined lower range is a MAD value equal to or less than about 40, about 35, about 30, about 25, about 20, about 15, about 10, about5, about of 1, or equal to or less than about 0. In some embodiments portions with a variability of counts outside a predetermined range are filtered out (eg excluded from consideration). In some embodiments a predetermined range is a MAD value greater than zero and less than about 76, less than about 74, less than about 73, less than about 72, less than about 71, less than about 70, less than about 69, less than about 68, less than about 67, less than about 66, less than about 65, less than about 64 , less than about 62, less than about 60, less than about 58, less than about 56, less than about 54, less than about 52, or less than about 50. In some embodiments a predetermined range is a MAD value greater than zero and less than about 67.7. In some embodiments portions with a variability of counts within a predetermined range are selected (eg used to determine the presence or absence of a genetic variation).

[263]In some embodiments the variability of portion counts represents a distribution (eg, a normal distribution). In some embodiments portions are selected within a distribution quantile. In some embodiments portions with a quantile equal to or less than about 99.9%, 99.8%, 99.7%, 99.6%, 99.5%, 99.4%, 99.3%, 99 .2%, 99.1%, 99.0%, 98.9%, 98.8%, 98.7%, 98.6%, 98.5%, 98.4%, 98.3%, 98 .2%, 98.1%, 98.0%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 85%, 80%, or equal to or less than a quantile of about 75% for the distribution are selected. In some embodiments portions with a 99% quantile of the distribution of count variability are selected. In some embodiments portions with a MAD > 0 and a MAD < 67,725 with a 99% quantile e are selected, resulting in the identification of a set of stable portions of a reference genome.

[264] Non-limiting examples of portion filtering in relation to PERUN are provided herein and in International Patent Application No. PCT/US12/59123 (W02013/052913) all content of which is incorporated herein by reference, including all texts, tables, equations and drawings. Portions can be filtered based on, or based in part on, an error measure. An error measure comprising absolute values of deviation, such as an R-factor, may be used for portion or weight removal in certain embodiments. An R-factor, in some embodiments, is defined as the sum of absolute deviations from the predicted count values of the actual measurements divided by the predicted count values of the actual measurements (eg, Equation B here). While an error measure comprising absolute values of deviation can be used, a suitable measure of error can alternatively be employed. In certain embodiments, an error measure not comprising absolute values of deviation, such as a square-based spread, can be used. In some embodiments, portions are filtered or weighted according to a measure of mappability (eg, a mappability score). A portion is sometimes filtered or weighed according to a relatively low number of sequence reads mapped to the portion (e.g. 0, 1, 2, 3, 4, 5 reads mapped to the portion). portion). Portions can be filtered or weighed according to the type of analysis being performed. For example, for aneuploidy analysis of chromosome 13, 18 and/or 21, sex chromosomes can be filtered out, and only autosomes, or a subset of autosomes, can be analyzed. For determination of fetal sex, autosomes can be filtered, and only sex chromosomes (X and Y), or one of the sex chromosomes (X or Y), can be analyzed.

[265]In particular embodiments, the following filtering process may be employed. The same set of portions (eg portions of a reference genome) on a given chromosome (eg chromosome 21) are selected and the number of fragments (reads) in affected and unaffected samples are compared. The difference pertains to trisomy 21 and euploid samples and involves a set of portions that cover the majority of chromosome 21. The set of portions is the same between euploid and T21 samples. The distinction between a set of portions and a single section is not crucial, as a portion can be defined. The same genomic region is compared in different patients. This process can be used for a trisomy analysis, such as for T13 or T18, in addition to, or instead of, T21.

[266]After the datasets have been counted, optionally filtered and normalized, the processed datasets can be manipulated by weighting, in some embodiments. One or more portions may be selected for weighting to reduce the influence of data (e.g., noisy data, uninformative data) contained in the selected portions, in certain embodiments, and in some embodiments, one or more portions may be selected for weighting to improve or increase the influence of data (for example, data with small measured variation) contained in the selected portions. In some embodiments, a dataset is weighted using a single weighting function that decreases the influence of data with large variations and increases the influence of data with small variations. A weighting function is sometimes used to reduce the influence of data with large variations and increase the influence of data with small variations (for example, [1/(standard deviation)2]). In some embodiments, a profile graph of the further processed data manipulated by weighting is generated to facilitate sorting and/or providing a result. A result can be provided based on a profile graph of the weighted data.

[267]Filtering or weighting of portions can be performed at one or more suitable points in an analysis. For example, portions can be filtered or weighted before or after sequence reads have been mapped to portions of a reference genome. Portions may be filtered or weighted before or after an experimental bias for individual genome portions is determined in some embodiments. In certain embodiments, portions may be filtered or weighted before or after genomic cross-section levels are calculated.

[268]After datasets have been counted, optionally filtered, normalized, and optionally weighted, the processed datasets may be manipulated by one or more mathematical and/or statistical manipulations (e.g., statistical functions or statistical algorithm), in some modalities. In certain embodiments, the processed datasets can be further manipulated by calculating Z scores for one or more selected portions, chromosomes, or portions of chromosomes. In some embodiments, the processed datasets can be further manipulated by calculating P values. One embodiment of an equation for calculating a Z score and a p value is presented in Equation 1 (Example 2). In certain embodiments, mathematical and/or statistical manipulations include one or more assumptions concerning ploidy and/or fetal fraction. In some embodiments, a profile map of the processed data further manipulated by one or more statistical and/or mathematical manipulations is generated to facilitate classification and/or provide a result. A result can be provided based on a profile graph of statistically and/or mathematically manipulated data. A result provided based on the profile plot of statistically and/or mathematically manipulated data often includes one or more hypotheses concerning ploidy and/or fetal fraction.

[269]In certain embodiments, various manipulations are performed on processed datasets to generate an N-dimensional space and/or N-dimensional point, after the datasets have been counted, optionally filtered and normalized. A result can be provided based on a profile plot of analyzed datasets in N-dimensions.

[270]In some embodiments, datasets are processed using one or more peak level analyses, peak width analyses, peak margin location analyses, peak lateral tolerances, the like, derivations thereof, or combinations thereof. earlier, as part of or after datasets have been processed and/or manipulated. In some embodiments, a profile graph of the processed data using one or more peak level analyses, peak width analyses, peak edge location analyses, peak lateral tolerances, the like, derivations thereof, or combinations of the foregoing is generated to facilitate sorting and/or provide a result. A result can be provided based on a profile plot of data that has been processed using one or more peak level analyses, peak width analyses, peak margin location analyses, peak lateral tolerances, similar, themselves, or combinations of the above.

[271] In some embodiments, the use of one or more reference samples that are substantially free of a genetic variation in question can be used to generate a profile of median reference count counts, which can result in a predetermined value representative of the absence of genetic variation, and often deviates from a predetermined value in areas that correspond to the genomic locus where the genetic variation is located in the test subject, if the test subject had the genetic variation. In test subjects at risk of, or suffering from, a medical condition associated with a genetic variation, the numerical value for the selected portion or sections is expected to vary significantly from the predetermined value of the unaffected genomic loci. In certain embodiments, the use of one or more reference samples known to carry the genetic variation in question can be used to generate a profile of median reference counts, which can result in a predetermined value representative of the presence of the genetic variation. , and often deviates from a predetermined value in areas corresponding to the genomic locus in which a test subject does not carry the genetic variation. In test subjects not at risk of, or suffering from, a medical condition associated with a genetic variation, the numerical value for the selected portion or sections is expected to vary significantly from the predetermined value of the affected genomic loci.

[272]In some embodiments, data analysis and processing may include the use of one or more hypotheses. An adequate number or type of hypotheses can be used to analyze or process a data set. Non-limiting examples of hypotheses that may be used for data processing and/or analysis include maternal ploidy, fetal contribution, prevalence of certain sequences in a reference population, ethnic origin, prevalence of a selected medical condition in related family members , parallelism between profiles of raw counts from different patients and/or runs after GC normalization and repeated masking (eg, GCRM), identical matches represent PCR artifacts (eg, identical baseline position), assumptions inherent in an assay of fetal quantification (e.g. FQA), twin hypotheses (e.g. if 2 twins and only one is affected the effective fetal fraction is only 50% of the total measured fetal fraction (similar for triplets, quadruplets and similar)) , fetal cell-free DNA (eg, cfDNA) uniformly covers the entire genome, peers and combinations thereof.

[273] In cases where the quality and/or depth of the mapped sequence reads do not allow a prediction of the outcome of the presence or absence of a genetic variation at a desired confidence level (e.g., a confidence level of 95% or greater), based on the profiles of the normalized counts, one or more mathematical manipulation algorithms and/or additional statistical prediction algorithms may be used to generate additional numerical values useful for data analysis and/or providing a result. The term "normalized count profile" as used herein refers to a profile generated using normalized counts. Examples of methods that can be used to generate normalized counts and profiles of the normalized counts are described here. As noted, fragments (reads) of mapped sequences that were counted can be normalized with respect to test sample counts or reference sample counts. In some embodiments, a normalized count profile can be displayed as a graph.

profiles

[274]In some embodiments, a processing step may comprise generating one or more profiles (e.g., profile graph) of various aspects of a dataset or deriving the same (e.g., the product of one or more processing steps). mathematical and/or statistical data processing known in the art and/or described herein). The term "profile", as used herein, refers to a product of mathematical and/or statistical manipulation of data that can facilitate the identification of patterns and/or correlations in large amounts of data. A "profile" typically includes values resulting from one or more manipulations of data or datasets, based on one or more criteria. A profile often includes multiple data points. Any suitable number of data points can be included in a profile depending on the nature and/or complexity of a dataset. In certain embodiments, profiles may include 2 or more data points, 3 or more data points, 5 or more data points, 10 or more data points, 24 or more data points, 25 or more data points, 50 or more data points, 100 or more data points, 500 or more data points, 1000 or more data points, 5000 or more data points, 10,000 or more data points, or 100,000 or more data points.

[275] In some embodiments, a profile is representative of the entire dataset, and in certain embodiments, a profile is representative of a portion or a subset of a dataset. That is, a profile sometimes includes or is generated from data points representative of data that have not been filtered to remove any data, and a profile sometimes includes or is generated from data points representative of data that have been filtered to remove unwanted data. In some embodiments, a data point on a profile represents the results of data manipulation for a portion. In certain embodiments, a data point in a profile includes data manipulation results for groups of portions. In some embodiments, groups of portions can be adjacent to each other, and in certain embodiments, groups of portions can be from different parts of a chromosome or genome.

[276]Data points in a profile derived from a dataset can be representative of any suitable data categorization. Non-limiting examples of categories where data can be grouped to generate profile data points include: portions based on size, portions based on sequence characteristics (e.g., GC content, AT content, position in a chromosome (eg, short arm, long arm, centromere, telomeres), and similar), expression levels, chromosome, similar, or combinations thereof. In some embodiments, a profile may be generated from data points obtained from another profile (eg, profiling normalized data renormalized to a different normalization value to generate a profile of renormalized data). In certain embodiments, a profile generated from data points obtained from another profile reduces the number of data points and/or complexity of the data set. Reducing the number of data points and/or complexity of the dataset often makes it easier to interpret data and/or makes it easier to provide a result.

[277]A profile (eg, a genomic profile, a chromosome profile, a profile of a segment of a chromosome) is often a collection of normalized or non-normalized counts for two or more portions. A profile usually includes at least one level (eg, a genomic section level), and often comprises two or more levels (eg, a profile often has multiple levels). A tier is usually for a set of portions having about the same counts or normalized counts. Levels are described in greater detail here. In certain embodiments, a profile comprises one or more portions, which portions may be weighted, removed, filtered, normalized, adjusted, averaged, averaged, added, subtracted, processed, or transformed by any combination thereof. A profile often comprises normalized counts mapped to portions defining two or more levels, where the counts are further normalized to one of the levels by a suitable method. Often counts of a profile (eg a level profile) are associated with an uncertainty value.

[278]A profile comprising one or more levels is sometimes padded (eg padding space). Filling (eg, padding) refers to a process of identifying and adjusting the levels of a profile that are due to maternal micro-deletions or maternal duplications (eg, copy number variations). In some embodiments levels are filled in that are due to fetal micro-duplications or fetal micro-deletions. Micro-duplications or micro-deletions in a profile can, in some embodiments, artificially increase or decrease the overall level of a profile (e.g., a profile of a chromosome) leading to false positive or false negative determinations of a chromosome aneuploidy (for example, a trisomy). In some embodiments levels in a profile that are due to micro-duplications and/or deletions are identified and adjusted (eg filled in and/or removed) by a process sometimes referred to as padding or padding. In certain embodiments a profile comprises one or more first levels that are significantly different from a second level within the profile, each of the one or more first levels comprising a maternal number variation, fetal copy number variation, or a fetal copy number variation. maternal copy number and fetal copy number variation and one or more of the first levels are adjusted.

[279]A profile comprising one or more levels may include a first level and a second level. In some embodiments a first level is different (eg, significantly different) from a second level. In some embodiments, a first tier comprises a first set of portions, a second tier comprises a second set of portions, and the first set of portions is not a subset of the second set of portions. In certain embodiments, a first set of portions is different from a second set of portions from which a first and a second level are determined. In some embodiments a profile may have multiple first levels that are different (e.g., significantly different, e.g. have a significantly different value) from a second level within the profile. In some embodiments a profile comprises one or more first levels that are significantly different from a second level within the profile and one or more of the first levels are adjusted. In some embodiments a profile comprises one or more first levels that are significantly different from a second level within the profile, each of the one or more first levels comprising a maternal copy number variation, fetal copy number variation, or fetal copy number variation. of maternal copy number and a variation of fetal copy number, and one or more of the first levels are adjusted. In some embodiments a first level within a profile is removed from the profile or adjusted (eg filled in). A profile may comprise multiple levels that include one or more first levels that are significantly different from one or more second levels, and often most of the levels in a profile are second levels, where second levels are approximately equal to each other. In some modalities more than 50%, more than 60%, more than 70%, more than 80%, more than 90% or more than 95% of the levels in a profile are second levels.

[280]A profile is sometimes presented as a graph. For example, one or more levels representing counts (eg normalized counts) of servings can be plotted and visualized. Non-limiting examples of profile graphs that can be generated include raw count (e.g., raw count profile or raw profile), normalized count, weighted portion, z score, p value, ratio of area versus adjusted ploidy, level mean versus ratio between adjusted and measured fraction, principal components, like, or combinations thereof. Profile graphs allow the visualization of manipulated data, in some modalities. In certain embodiments, a profile plot may be used to provide a result (eg, area ratio versus adjusted ploidy, mean level versus ratio between adjusted and measured fetal fraction, principal components). The terms "raw count profile graph" or "raw profile graph", as used herein, refer to a graph of the counts in each portion of a region normalized to total counts in a region (e.g., genome, portion , chromosome, chromosome portions of a reference genome, or a segment of a chromosome). In some embodiments, a profile can be generated using a static window process, and in certain embodiments, a profile can be generated using a sliding window process.

[281]A profile generated by a test subject is sometimes compared with a profile generated for one or more reference subjects, in order to facilitate the interpretation of mathematical and/or statistical manipulations of a dataset and/or to provide a result. In some embodiments, a profile is generated based on one or more starting hypotheses (e.g., maternal nucleic acid contribution (e.g., maternal fraction), fetal nucleic acid contribution (e.g., fetal fraction), sample ploidy of reference, the similar or their combinations). In certain embodiments, a test profile often revolves around a predetermined value representative of the absence of a genetic variation, and often deviates from a predetermined value in areas that correspond to the genomic location in which the genetic variation is located. in the test subject, whether the test subject had the genetic variation. In test subjects at risk of, or suffering from, a medical condition associated with a genetic variation, the numerical value of a selected portion is expected to vary significantly from the predetermined value for unaffected genomic loci. Depending on the initial hypotheses (for example, fixed ploidy or optimized ploidy, fixed fetal fraction or optimized fetal fraction or combinations thereof) the predetermined limit or cut-off value or limit range of values indicating the presence or absence of a genetic variation may vary while still providing a useful result for determining the presence or absence of a genetic variation. In some embodiments, a profile is indicative of and/or representative of a phenotype.

[282]By way of non-limiting example, normalized sample and/or reference count profiles may be obtained from readable raw sequence data by (a) calculating the median reference counts of selected chromosomes, portions or their segments from a set of known references that do not carry genetic variation, (b) removing non-informative portions from the raw counts of the reference sample (eg, filtering); (c) normalize the reference counts for all remaining portions of a reference genome to the total residual number of counts (e.g., sum of counts after removing uninformative portions of a reference genome) for the selected chromosome from the selected reference sample or genomic location, thereby generating a normalized reference subject profile; (d) removing corresponding portions of the sample from the test subject; and (e), normalizing the remaining test subject counts for one or more selected genomic loci to the sum of the median residual reference counts for the chromosome or chromosomes containing the selected genomic loci, thereby generating a profile of the test subject normalized. In certain embodiments, an additional normalization step with respect to the entire genome, reduced by the filtered portions in (b), may be included between (c) and (d).

[283]A dataset profile may be generated by one or more readable data manipulations of counted mapped sequences. Some modalities include the following. Fragments (reads) of sequences are mapped and the number of counts (ie, sequence tags) for the genomic mapping of each of the portions are determined (eg, counted). A raw count profile is generated from the mapped sequence reads that are counted. A result is provided by comparing a test subject's raw count profile to a reference median count profile for chromosomes, segments, or portions or segments thereof from a set of reference subjects known not to have a genetic variation, in certain embodiments.

[284]In some embodiments, the stream readable data is optionally filtered to remove noisy data or uninformative portions. After filtering, the remaining counts are typically summed to generate a filtered dataset. A filtered counts profile is generated from a filtered dataset, in certain embodiments.

[285]After the readable sequence data has been counted and optionally filtered, datasets can be normalized to generate levels or profiles. A data set can be normalized by normalizing one or more selected portions to an appropriate normalization reference value. In some embodiments, a normalization reference value is representative of the total counts for the chromosome or chromosomes from which portions are selected. In certain embodiments, a normalization reference value is representative of one or more corresponding portions, chromosome portions or chromosomes of a reference data set prepared from a reference subject set known not to have a genetic variation. In some embodiments, a normalization reference value is representative of one or more corresponding portions, chromosome portions, or chromosomes of a test subject dataset prepared from a test subject being analyzed for the presence or absence of a variation. genetics. In certain embodiments, the normalization process is performed using a static window approach, and in some embodiments the normalization process is performed using a moving or sliding window approach. In certain embodiments, a profile comprising normalized counts is generated to facilitate classification and/or provide a result. A result may be provided based on a profile graph comprising normalized counts (eg using such a profile graph).

Levels

[286]In some embodiments, a value (eg, a number, a quantitative value) is assigned to a level. A level can be determined by a suitable method, operational or mathematical process (eg a processed level). A level often is, or is derived from, counts (eg, normalized counts) for a set of portions. In some embodiments a level of a portion is substantially equal to the total number of counts mapped to a portion (e.g., counts, normalized counts). Often, a level is determined from counts which are processed, transformed or manipulated by a suitable method, operational or mathematical process known in the art. In some embodiments a level is derived from counts that are processed, and non-limiting examples of processed counts include weighted, removed, filtered, normalized, adjusted, averaged, derived as an average (e.g., mean level), added, subtracted, or transformed, or a combination thereof. In some embodiments a tier comprises counts that are normalized (eg, normalized portion counts). A level may separate counts normalized by a suitable process, non-limiting examples of which include normalization by portions, normalization by GC content, linear or non-linear least squares regression, GC LOESS, LOWESS, PERUN, RM, GCRM, cQn , the similar and/or their combinations. A level can comprise normalized counts or relative amounts of counts. In some embodiments a level is for counts or normalized counts of two or more portions that are calculated and the level is referred to as an average level. In some embodiments a tier is for a set of servings having an average count or average of normalized counts which is referred to as an average tier. In some embodiments a level is derived from portions comprising raw and/or filtered counts. In some embodiments, a level is based on scores that are raw. In some embodiments a level is associated with an uncertainty value (eg a standard deviation, a MAD). In some embodiments a level is represented by a Z-score or p-value.

[287]A level for one or more portions is synonymous with a "genomic section level" here. The term "level" as used herein is sometimes synonymous with the term "elevation". Determining the meaning of the term "level" can be determined from the context in which it is used. For example, the term "level", when used in the context of genomic sections, profiles, reads and/or counts, often means an elevation. The term "level", when used in the context of a substance or composition (eg, RNA level, plexus level) often refers to an amount. The term "level", when used in the context of uncertainty (eg, error level, confidence level, deviation level, uncertainty level) often refers to a quantity.

[288]Normalized or non-normalized counts for two or more levels (e.g. two or more levels of a profile) can sometimes be mathematically manipulated (e.g. added, multiplied, averaged, normalized, like or combination thereof ) according to levels. For example, normalized or non-normalized counts for two or more levels can be normalized according to one, some or all levels of a profile. In some embodiments normalized or non-normalized counts of all levels are from a profile normalized according to a level in the profile. In some embodiments normalized or non-normalized counts of a first level in a profile are normalized according to normalized or non-normalized counts of a second level in the profile.

[289]Non-limiting examples of a tier (e.g., a 1st tier, a 2nd tier) are a tier for a set of portions comprising processed counts, a tier for a set of portions comprising a mean, median, or arithmetic mean of counts, a level for a set of portions comprising normalized counts, the like, or any combination thereof. In some embodiments, a first level and a second level in a profile are derived from counts of portions mapped to the same chromosome. In some embodiments, a first level and a second level in a profile are derived from counts of portions mapped to different chromosomes.

[290]In some embodiments a level is determined from the normalized or non-normalized counts mapped to one or more portions. In some embodiments, the level is determined from the normalized or non-normalized counts for two or more portions, where the normalized counts for each portion are often the same. There can be variation in counts (eg normalized counts) within a set of servings for a level. In a set of slices for a level there may be one or more slices having counts that are significantly different than other slices in the set (eg peaks and/or troughs). Any suitable number of normalized or non-normalized counts associated with any suitable number of servings can define a level.

[291] In some embodiments one or more levels can be determined from the normalized or non-normalized counts of all or some of the portions of a genome. Often, a level can be determined from all or some of the normalized or non-normalized counts of a chromosome, or its segment. In some embodiments, two or more scores derived from two or more servings (eg, a set of servings) determine a level. In some arrangements two or more counts (for example, counts of two or more parts) determine a level. In some embodiments, counts from 2 to about 100,000 servings determine a level. In some embodiments, counts of 2 to about 50,000, 2 to about 40,000, 2 to about 30,000, 2 to about 20,000, 2 to about 10,000, 2 to about 5000, 2 to about 2500, 2 to 1250, 2 to about 1000, 2 to about 500, 2 to about 250, 2 to about 100, or 2 to about 60 servings determine a level. In some embodiments counts from about 10 to about 50 servings determine a level. In some embodiments counts from about 20 to about 40 or more servings determine a level. In some embodiments, a level comprises counts of about 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26,27, 28, 29, 30, 31, 32, 33,34,35, 36, 37, 38, 39, 40, 45, 50, 55, 60 or more servings. In some embodiments, a level corresponds to a set of portions (eg, a set of portions of a reference genome, a set of portions of a chromosome, or a set of portions of a segment of a chromosome).

[292]In some embodiments, a level is determined for normalized or non-normalized counts of portions that are contiguous. In some embodiments, portions (eg, a set of portions) that are contiguous represent neighboring segments of a genome or neighboring segments of a chromosome or gene. For example, two or more contiguous portions, when aligned by merging the portions end to end, can represent a sequence set of DNA sequence greater than each of the portions. For example, two or more contiguous portions may represent an intact genome, chromosome, gene, intron, exon, or segment thereof. In some embodiments a tier is determined from a collection (eg, a set) of contiguous portions and/or non-contiguous portions.

different levels

[293] In some embodiments, a profile of normalized counts comprises a level (eg, a first tier) significantly different from another tier (eg, a second tier) within the profile. A first level can be higher or lower than a second level. In some embodiments, a first tier is for a set of parts comprising one or more fragments (reads) comprising a copy number variation (e.g., a maternal copy number variation, fetal copy number variation, or a copy number variation). of maternal copy number and one fetal copy number variation) and the second tier is for a set of portions comprising fragments (reads) having substantially no copy number variation. In some embodiments, significantly different refers to an observable difference. In some embodiments significantly different refers to statistically different or a statistically significant difference. A statistically significant difference is sometimes a statistical evaluation of an observed difference. A statistically significant difference can be assessed using a suitable method in the art. Any suitable threshold or range can be used to determine that two levels are significantly different. In certain embodiments, two levels (for example, mean levels) that differ by about 0.01 percent or more (for example, 0.01 percent of one or both of the values for the level) are significantly different. In some embodiments, two levels (eg mean levels) that differ by about 0.1 percent or more are significantly different. In certain embodiments, two levels (eg mean levels) that differ by about 0.5 percent or more are significantly different. In some embodiments, two levels (e.g., average levels) that differ by about 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5 or more than about 10% are significantly different. In some embodiments, two levels (eg mean levels) are significantly different and there is no overlap at any level and/or no overlap within a range defined by a calculated uncertainty value for one or both levels. In certain embodiments, the uncertainty value is a standard deviation expressed as sigma. In some embodiments, two levels (eg mean levels) are significantly different and they differ by about 1 or more times the uncertainty value (eg 1 sigma). In some embodiments, two levels (e.g. mean levels) are significantly different and they differ by about 2 or more times the uncertainty value (e.g. 2 sigmas), about 3 or more, about 4 or more, about of 5 or more, about 6 or more, about 7 or more, about 8 or more, about 9 or more, or about 10 or more times the uncertainty value. In some embodiments, two levels (e.g., average levels) are significantly different when they differ by about 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3, 0, 3.1, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, or 4.0 times the uncertainty value or more. In some modalities, the confidence level increases as the difference between the two levels increases. In certain modalities the confidence level decreases as the difference between the two levels decreases and/or as the uncertainty value increases. For example, sometimes the confidence level increases with the proportion of the difference between levels and the standard deviation (e.g. MADs)

[294]One or more prediction algorithms can be used to determine the significance or make sense of detection data collected under varying conditions that can be weighted independently of or dependent on one another. The term "variable" as used herein refers to a factor, quantity, or function of an algorithm that has a value or set of values.

[295] In some embodiments, a first set of portions often includes portions that are different from (eg, non-overlapping with) a second set of portions. For example, sometimes a first level of normalized counts is significantly different from a second level of normalized counts in a profile, and the first level is for a first set of portions, the second level is for a second set of portions, and the portions do not overlap with the first and second set of portions. In certain embodiments, a first set of portions is not a subset of a second set of portions from which a first tier and second tier are determined, respectively. In some embodiments a first set of portions is different and/or distinct from a second set of portions from which a first tier and second tier are determined, respectively.

[296]In some embodiments a first set of portions is a subset of a second set of portions in a profile. For example, sometimes a second level of normalized counts for a second set of portions in a profile comprises counts normalized from a first set of portions to a first level in the profile, and the first set of portions is a subset of the second set of portions in the profile. In some embodiments, a level mean, median or arithmetic mean is derived from a second level where the second level comprises a first level. In some embodiments, a second tier comprises a second set of portions representing an entire chromosome and a first tier comprises a first set of portions wherein the first set is a subset of the second set of portions and the first tier represents a variation in the number of maternal copy, fetal copy number variation, or a maternal copy number variation and a fetal copy number variation that is present on the chromosome.

[297]In some embodiments, a second-level value is closer to the mean, arithmetic mean, or median value of a profile count of a chromosome, or its segment, than the first level. In some embodiments, a second level is a middle level of a chromosome, a portion of a chromosome or its segment. In some embodiments, a first level is significantly different from a predominant level (eg, a second level), representing a chromosome, or segment thereof. A profile can include multiple first levels that differ significantly from a second level, and each first level can independently be higher or lower than the second level. In some embodiments, a first level and a second level are derived from the same chromosome, and the first level is higher or lower than the second level, and the second level is the predominant level of the chromosome. In some embodiments, a first level and a second level are derived from the same chromosome, a first level is indicative of a copy number variation (e.g., a maternal and/or fetal copy number variation, deletion, insertion, duplication ) and a second level is a significant level or predominant level of portions of a chromosome, or its segment.

[298]In certain embodiments, a fragment (read) in a second set of portions for a second level substantially does not include a genetic variation (e.g., a copy number variation, a maternal and/or fetal copy number variation ). Often, a second set of servings for a second tier includes some variability (eg variability in tier, variability in counts for servings). In some embodiments, one or more portions of a set of portions for a level associated with substantially no copy number variation includes one or more fragments (reads) having a copy number variation present in a maternal and/or fetal genome. For example, sometimes a portion set includes a copy number variation that is present in a small segment of a chromosome (e.g., less than 10 portions) and the portion set is to a level associated with substantially no copy number variation. Thus, a set of portions that include substantially no copy number variation can still include a copy number variation that is present at less than about 10, 9, 8, 7, 6, 5, 4, 3, 2 or 1 servings of a level.

[299] In some embodiments a first tier is for a first set of portions and a second tier is for a second set of portions and the first set of portions and second set of portions are contiguous (e.g., adjacent with respect to the sequence of portions). nucleic acid of a chromosome or its segment). In some embodiments the first set of portions and second set of portions are not contiguous.

[300]Relatively short sequence reads from a mixture of maternal and fetal nucleic acid can be used to provide counts that can be turned into a level and/or a profile. Counts, levels and profiles can be represented in electronic or tangible form and can be visualized. Counts mapped to chunks (e.g. represented as levels and/or profiles) can provide a visual representation of a genome, chromosome, or a portion or a segment of a maternal and/or fetal chromosome that is present in the fetus and/or woman pregnant.

Reference Level and Normalized Reference Value

[301]In some embodiments a profile comprises a reference level (eg, a level used as a reference). Often, a profile of normalized counts provides a reference level from which expected levels and expected ranges are determined (see discussion below on expected levels and ranges). A reference level is often for normalized counts of portions comprising mapped reads from both a mother and a fetus. A reference level is often the sum of normalized counts of mapped fragments (reads) from a fetus and a mother (eg, a pregnant woman). In some embodiments a reference level is for portions comprising mapped fragments (reads) of an euploid mother and/or an euploid fetus. In some embodiments a reference level is for portions comprising mapped fragments (reads) having a maternal and/or fetal genetic variation (e.g. an aneuploidy (e.g. a trisomy), a copy number variation, a micro-duplication , a micro-deletion, an insertion). In some embodiments a reference level is for portions that include substantially no maternal and/or fetal genetic variation (e.g., an aneuploidy (e.g., a trisomy), a copy number variation, a micro-duplication, a micro- deletion, an insertion). In some embodiments a second level is used as a reference level. In certain embodiments a profile comprises a first level of normalized counts and a second level of normalized counts, the first level is significantly different from the second level and the second level is the reference level. In certain embodiments a profile comprises a first level of counts normalized to a first set of chunks, a second level of counts normalized to a second set of chunks, the first set of chunks including mapped reads having a copy number variation maternal and/or fetal, the second set of portions comprises mapped reads having substantially no maternal copy number variation and/or fetal copy number variation, and the second level is a reference level.

[302] In some embodiments counts mapped to portions of one or more levels of a profile are normalized according to counts of a reference level. In some embodiments, normalizing one-level counts to one-level counts comprises dividing one-level counts by one-level counts or a multiple or fraction thereof. Counts normalized according to counts from a reference level have often been normalized according to another process (e.g. PERUN) and counts from a reference level have often also been normalized (e.g. by PERUN) . In some embodiments the counts of a level are normalized according to the counts of a reference level and the counts of the reference level can adapt to a suitable value either before or after normalizing. The process of adjusting the counts of a reference level can comprise any suitable constant (ie number) and any suitable mathematical manipulation can be used for the counts of a reference level.

[303]The normalized reference value (NRV) is often determined according to the normalized counts of a reference level. Determining an NRV can comprise any suitable normalization process (eg mathematical manipulation) applied to counts from a reference level, when the same normalization process is used to normalize counts from other levels within the same profile. Determining an NRV often involves dividing a reference level by itself. Determining an NRV often involves dividing a reference level by a multiple of si. Determining an NRV often involves dividing a reference level by the sum or difference of the reference level and a constant (eg any number).

[304]An NRV is sometimes referred to as a null value. An NRV can be any suitable value. In some embodiments, an NRV is any value other than zero. In some embodiments an NRV is an integer. In some embodiments an NRV is a positive integer. In some embodiments, an NRV equals 1, 10, 100, or 1000. Often, an NRV equals 1. In some embodiments, an NRV equals zero. Counts from a reference level can be normalized to any suitable NRV. In some embodiments, counts from a reference level are normalized to an NRV of zero. Often, counts from a reference level are normalized to an NRV of 1.

expected levels

[305]An expected level is sometimes a predefined level (eg theoretical level, predicted level). An "expected level" is sometimes referred to herein as a "predetermined level value". In some embodiments, an expected level is a predicted value for a level of normalized counts for a set of portions that include the copy number variation. In certain embodiments, an expected level is determined by a set of portions that include virtually no copy number variation. An expected level can be determined in terms of chromosome ploidy (e.g. 0, 1, 2 (i.e. diploid), 3 or 4 chromosomes) or a microploidy (homozygous or heterozygous deletion, duplication, insertion or absence thereof) . Often, an expected level is determined for maternal microploidy (eg, a maternal and/or fetal copy number variation).

[306]An expected level for a genetic variation or a copy number variation may be determined in any suitable way. Often, an expected level is determined by a proper mathematical manipulation of a level (for example, counts mapped to a set of servings for a level). In some embodiments an expected level is determined through the use of a constant sometimes referred to as an expected level constant. An expected level for copy number variation is sometimes calculated by multiplying a reference level, normalized counts of a reference level, or an NRV by an expected level constant, adding an expected level constant, subtracting an expected level constant , dividing by a constant of the expected level, or by a combination thereof. Often, an expected level (eg, an expected level of a maternal and/or fetal copy number variation) determined for the same subject, sample, or test group is determined according to the same reference level or NRV.

[307]Often, an expected level is determined by multiplying a reference level, normalized counts of a reference level, or an NRV by an expected level constant where the reference level, normalized counts of a reference level, or NRV is not equal to zero. In some embodiments an expected level is determined by adding an expected level constant to the reference level, normalized counts from a reference level, or an NRV that equals zero. In some embodiments, the expected level, normalized counts of a reference level, NRV, and expected level constant are adjustable. The fitting process can comprise any suitable constant (ie number) and any suitable mathematical manipulation, where the same fitting process is applied to all values under consideration.

Expected level constant

[308]A constant of the expected level can be determined by a suitable method. In some embodiments a constant of the expected level is determined arbitrarily. Often, a constant of the expected level is determined empirically. In some embodiments a constant of the expected level is determined according to a mathematical manipulation. In some embodiments an expected level constant is determined against a reference (e.g., a reference genome, a reference sample, reference test data). In some embodiments, an expected level constant is predetermined to a level representative of the presence or absence of a genetic variation or copy number variation (e.g., a duplication, insertion, or deletion). In some embodiments, an expected level constant is predetermined for a level representative of the presence or absence of a maternal copy number change, fetal copy number change, or a maternal copy number change and a fetal copy number change. fetal copy. An expected level constant for copy number variance can be any constant or a suitable set of constants.

[309] In some embodiments, the expected level constant for a homozygous duplication (e.g., a homozygous duplication) may be from about 1.6 to about 2.4, from about 1.7 to about 2, 3, from about 1.8 to about 2.2, or from about 1.9 to about 2.1. In some embodiments the expected level constant for a homozygous duplication is about 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2, 2.3 or about 2,4. Often, the expected level constant for a homozygous duplication is about 1.90, 1.92, 1.94, 1.96, 1.98, 2.0, 2.02, 2.04, 2.06 , 2.08 or about 2.10. Often, the expected level constant for a homozygous duplication is around 2.

[310] In some embodiments, the expected level constant for a heterozygous duplication (e.g., a homozygous duplication) is from about 1.2 to about 1.8, from about 1.3 to about 1.7 , or from about 1.4 to about 1.6. In some embodiments the expected level constant for a heterozygous duplication is about 1.2, 1.3, 1.4, 1.5, 1.6, 1.7 or about 1.8. Often, the expected level constant for a heterozygous duplication is about 1.40, 1.42, 1.44, 1.46, 1.48, 1.5, 1.52, 1.54, 1.56 , 1.58 or about 1.60. In some embodiments, the expected level constant for a heterozygous duplication is about 1.5.

[311]In some embodiments, the expected level constant for the absence of a copy number change (e.g., the absence of a maternal copy number change and/or fetal copy number change) is about 1.3 to about 0.7, from about 1.2 to about 0.8, or from about 1.1 to about 0.9. In some embodiments the expected level constant for the absence of a copy number change is about 1.3, 1.2, 1.1, 1.0, 0.9, 0.8 or about 0. 7. Often the expected level constant for the absence of a copy number change is about 1.09, 1.08, 1.06, 1.04, 1.02, 1.0, 0.98, 0, 96, 0.94, or about 0.92. In some embodiments, the expected level constant for the absence of a copy number change is about 1.

[312]In some embodiments, the expected level constant for a heterozygous deletion (eg, a maternal, fetal, or maternal and fetal heterozygous deletion) is from about 0.2 to about 0.8, from about 0, 3 to about 0.7, or from about 0.4 to about 0.6. In some embodiments the expected level constant for a heterozygous deletion is about 0.2, 0.3, 0.4, 0.5, 0.6, 0.7 or about 0.8. Often the expected level constant for a heterozygous deletion is about 0.40, 0.42, 0.44, 0.46, 0.48, 0.5, 0.52, 0.54, 0.56, 0.58 or about 0.60. In some embodiments, the expected level constant for a heterozygous deletion is about 0.5.

[313]In some embodiments, the expected level constant for a homozygous deletion (e.g., a homozygous deletion) may be from about -0.4 to about 0.4, from about -0.3 to about 0.3, from about -0.2 to about 0.2, or from about -0.1 to about 0.1. In some embodiments the expected level constant for a homozygous deletion is about -0.4, -0.3, -0.2, -0.1, 0.0, 0.1, 0.2, 0, 3 or about 0.4. Often, the expected level constant for a homozygous deletion is around -0.1, -0.08, -0.06, -0.04, -0.02, 0.0, 0.02, 0, 04, 0.06, 0.08 or about 0.10. Often, the expected level constant for a homozygous deletion is around 0.

Expected level range

[314]In some embodiments, the presence or absence of a genetic variation or copy number variation (e.g., a maternal copy number variation, fetal copy number variation, or a maternal copy number variation and a fetal copy number variation) is determined by a level that falls within or outside a range of the expected level. An expected level range is often determined according to an expected level. In some embodiments a range of the expected level is determined to be a level comprising substantially no genetic variation or substantially no copy number variation. A suitable method can be used to determine a range of the expected level.

[315]In some embodiments, a range of the expected level is defined according to a suitable uncertainty value calculated for a level. Non-limiting examples of an uncertainty value are a standard deviation, standard error, calculated variance, p-value, and mean absolute deviation (MAD). In some embodiments, a range of the expected level for a genetic variation or a copy number variation is determined, in part, by calculating the uncertainty value for a level (e.g., a first level, a second level, a first level). level and a second level). In some embodiments an expected level range is defined according to an uncertainty value calculated for a profile (eg a profile of normalized counts for a chromosome or its segment). In some embodiments, an uncertainty value is calculated for a level comprising substantially no genetic variation or substantially no copy number variation. In some embodiments, an uncertainty value is calculated for a first level, a second level, or a first level and a second level. In some embodiments an uncertainty value is determined by a first level, a second level or a second level comprises a first level.

[316]A range of the expected level is sometimes calculated, in part, by multiplying, adding, subtracting, or dividing an uncertainty value by a constant (eg, a predetermined constant) n. A suitable mathematical procedure or combination of processes can be used. The constant n (eg predetermined constant n) is sometimes referred to as a confidence interval. A selected confidence interval is determined according to the constant n that is selected. The constant n (eg the predetermined constant n, the confidence interval) can be determined in a suitable way. The constant n can be a number or the fraction of a number greater than zero. The constant n can be an integer. Often the constant n is a number less than 10. In some embodiments the constant n is a number less than about 10, less than about 9, less than about 8, less than about 7, less than about 6, less than about 5, less than about 4, less than about 3, or less than about 2. In some embodiments the constant n is about 10, 9.5 , 9, 8.5, 8, 7.5, 7, 6.5, 6, 5.5, 5, 4.5, 4, 3.5, 3, 2.5, 2 or 1. The constant n can be empirically determined from data derived from subjects (a pregnant woman and/or a fetus) with a known genetic disposition.

[317]Often, an uncertainty value and constant n define a range (for example, an uncertainty cutoff). For example, sometimes an uncertainty value is a standard deviation (e.g. +/- 5) and is multiplied by a constant n (e.g. a confidence interval), thereby defining a cut-off or uncertainty range (e.g. example, 5n to -5n).

[318]In some embodiments, a range of the expected level of genetic variation (for example, one maternal copy number variation, fetal copy number variation, or one maternal copy number variation and one fetal copy number variation ) is the sum of an expected level plus a constant n times the uncertainty (e.g. n x sigma (e.g. 6 sigma)). In some embodiments the expected level range for a genetic variation or copy number variation designated by k can be defined by the following formula: Formula R: (Expected level range}k = (expected level}k + no where o is a uncertainty value, n is a constant (e.g., a predetermined constant) and the expected level range and expected level are for the genetic variation k (e.g., k = a heterozygous deletion, e.g., k = the absence of a genetic variation). For example, for an expected level equal to 1 (for example, absence of a copy number variation), an uncertainty value (for example, o) equal to +/-0.05, and n = 3, the expected level range is defined as 1.15 to 0.85. In some embodiments, the expected level range for a heterozygous duplication is defined as 1.35 to 1.65 when the expected level for a heterozygous duplication is 1.5, n = 3, and the uncertainty value o is +/- 0.05. In some embodiments the expected level range for a heterozygous deletion is determined to be 0.65 to 0.35, when the expected level for a heterozygous doubling is 0.5, n = 3, and the uncertainty value o is -/+ 0.05. In some embodiments the expected level range for a homozygous duplication is determined to be 2.15 to 1.85 when the expected level for a heterozygous duplication is 2.0, n = 3 and the uncertainty value o is +/- 0 .05. In some embodiments the expected level range for a homozygous deletion is determined to be 0.15 to -0.15 when the expected level for a heterozygous duplication is 0.0, n = 3 and the uncertainty value o is +/- 0 .05.

[319]In some embodiments a range of the expected level for a homozygous copy number variation (for example, a maternal, fetal, or maternal and fetal homozygous copy number variation) is determined, in part, according to a range of the expected level of variation in the corresponding heterozygous copy number. For example, sometimes a range of the expected level for a homozygous duplication comprises all values greater than an upper bound of a range of the expected level for a heterozygous duplication. In some embodiments a range of the expected level for a homozygous duplication comprises all values greater than or equal to an upper limit of a range of the expected level for a heterozygous duplication. In some embodiments a range of the expected level for a homozygous duplication comprises all values greater than an upper limit of a range of the expected level for a heterozygous duplication and less than the upper limit defined by the formula R where a is a value of uncertainty e is a positive value, n is a constant, and k is a homozygous doubling. In some embodiments a range of the expected level for a homozygous duplication comprises all values greater than or equal to an upper limit of a range of the expected level for a heterozygous duplication and less than or equal to the upper limit defined by the formula R where a is an uncertainty value, a is a positive value, n is a constant, and k is a homozygous doubling.

[320]In some embodiments, a range of the expected level for a homozygous deletion comprises all values less than a lower bound of a range of the expected level for a heterozygous deletion. In some embodiments a range of the expected level for a homozygous deletion comprises all values less than or equal to a lower bound of a range of the expected level for a heterozygous deletion. In some embodiments, a range of the expected level for a homozygous deletion comprises all values less than a lower limit of a range of the expected level for a heterozygous deletion and greater than the lower limit defined by the formula R where a is a value of uncertainty, a is a negative value, n is a constant, and k is a homozygous deletion. In some embodiments a range of the expected level for a homozygous deletion comprises all values less than or equal to a lower limit of a range of the expected level for a heterozygous deletion and greater than or equal to the lower limit defined by the formula R where a is an uncertainty value, a is a negative value, n is a constant, and k is a homozygous deletion.

[321]An uncertainty value can be used to determine a threshold value. In some embodiments, a range (eg, a limiting range) is obtained by computing the uncertainty value determined from raw, filtered, and/or normalized counts. A range can be determined by multiplying the value for a level of uncertainty (e.g., normalized one-level counts) by a predetermined constant (e.g., 1, 2, 3, 4, 5, 6, etc.), which represents the uncertainty multiple (for example, number of standard deviations) chosen as the cut-off limit (for example, multiplying by 3 for 3 standard deviations), through which a band is generated, in some modalities. A range can be determined by adding and/or subtracting a value (e.g. a predetermined value, an uncertainty value, an uncertainty value multiplied by a predetermined constant) and/or from a level in which a track is generated, in some embodiments. For example, for a level equal to 1, a standard deviation of +/0.2, where a predetermined constant is 3, the range can be calculated as (1 + 3(0.2)) to (1 + 3 (-0.2)) or 1.6 to 0.4. A range can sometimes define an expected range or expected level range for copy number variation. In certain embodiments, some or all of the portions that exceed a threshold value, being outside a range or being within a range of values, are removed as part of, before, or after a normalization process. In some embodiments, some or all of the portions that exceed a calculated threshold value, being outside a range or being within a range, are weighted or adjusted as part of, or prior to, the normalization or classification process. Weighting examples are described here. The terms "redundant data", and "redundant mapped reads" as used herein refer to sample-derived sequence reads that are identified as having already been assigned a genomic location (e.g., base position). and/or counted per portion.

[322]In some embodiments an uncertainty value is determined according to the following formula: Where Z represents the standard deviation between two levels, L is the mean (or median) level, and sigma is the standard deviation (or MAD). The index O indicates a segment of a profile (for example, a second level, a chromosome, an NRV, an "euploid level", a level of absence of a copy number variation), and A indicates another segment of a profile (for example, a first level, a level representing a copy number variation, a level representing an aneuploidy (for example, a trisomy). The variable No represents the total number of portions in the segment of the profile indicated by the index O. NA represents the total number of servings in the profile segment indicated by the A index.

Categorizing a Copy Number Variation

[323]A level (e.g., a first level) that differs significantly from another level (e.g., a second level) can often be categorized as a copy number variation (e.g., a maternal copy number variation and /or fetal, a fetal copy number variation, a deletion, duplication, insertion) within a range of the expected level. In some embodiments, the presence of a copy number variation is categorized when a first level is significantly different from a second level and the first level is within the range of the expected level for a copy number variation. For example, a copy number variation (e.g., a maternal and/or fetal copy number variation, a fetal copy number variation) can be categorized where a first level is significantly different from a second level and the first level is within the range of the expected level for a copy number variation. In some embodiments a heterozygous duplication (e.g., a maternal or fetal or maternal and fetal heterozygous duplication) or heterozygous deletion (e.g., a maternal or fetal or maternal and fetal heterozygous deletion) is categorized when a first level is significantly different from a second level and first level are within the range of the expected level for a heterozygous duplication or heterozygous deletion, respectively. In some embodiments a homozygous duplication or homozygous deletion is categorized when a first level is significantly different from a second level and the first level is within the range of the expected level for a homozygous duplication or homozygous deletion, respectively.

Level adjustments

[324]In some embodiments, one or more levels are adjusted. A process for adjusting a level is often referred to as padding. In some embodiments, multiple levels in a profile (e.g., a profile of a genome, a profile of a chromosome, a profile of a portion or segment of a chromosome) are adjusted. In some embodiments, from about 1 to about 10,000 or more levels in a profile are fitted. In some embodiments about 1 to about 1000, about 1 to 900, 1 to about 800, 1 to about 700, 1 to about 600, 1 to about 500, 1 to about 400, 1 to about 300, 1 to about 200, 1 to about 100, 1 to about 50, 1 to about 25, 1 to about 20, 1 to about 15, 1 to about 10, or about 1 to 5 at a levels in a profile are adjusted. In some modes a level is set. In some embodiments, a level (e.g., a first level of a normalized count profile) that differs significantly from a second level is adjusted. In some embodiments a level categorized as a copy number variation is set. In some embodiments a level (e.g. a first level of a normalized count profile) that differs significantly from a second level is categorized as a copy number variation (e.g. a copy variation number, for example, a variation of the maternal copy number) and is adjusted. In some embodiments, a level (e.g., a first level) is within a range of the expected level for a maternal copy number change, fetal copy number change, or a maternal copy number change and a fetal copy number change. fetal copy number and the level is adjusted. In some embodiments, one or more levels (for example, in levels in a profile) are not adjusted. In some embodiments, a level (eg, a first level) is outside a range of the expected level for a copy number variation, and the level is not adjusted. Often, a level within a range of the expected level in the absence of a copy number variance is not adjusted. Any suitable number of adjustments can be made to one or more levels in a profile. In some embodiments, one or more levels are adjusted. In some modes 2 or more, 3 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more times, and 10 or more levels are set.

[325]In some embodiments, a value of a first level is adjusted according to a value of a second level. In some embodiments a first level, identified as representative of a copy number variation, is set to the value of a second level, wherein the second level is often associated with no copy number variation. In certain embodiments, a first level value identified as representative of a copy variation number is adjusted such that the first level value is approximately equal to a second level value.

[326]An adjustment may comprise a suitable mathematical operation. In some embodiments a fit comprises one or more mathematical operations. In some embodiments a level is adjusted by normalizing, filtering, averaging, multiplying, dividing, adding or subtracting or combination thereof. In some embodiments a level is set by a predetermined value or a constant. In some embodiments a level is adjusted by changing the value of the level to the value of another level. For example, a first level can be adjusted by modifying its value with the value of a second level. A value in such cases can be a processed value (eg mean value, normalized and the like).

[327]In some embodiments a level is categorized as a copy number variation (e.g., a maternal copy number variation) and is adjusted according to a predetermined value referred to herein as a predetermined adjustment value (VAP). Often, a PAV is determined by a particular copy number variation. Often, a PAV determined by a copy number variation (eg, homozygous duplication, homozygous deletion, heterozygous duplication, heterozygous deletion) is used to adjust a level categorized as a specific copy number variation (eg, homozygous duplication, homozygous deletion, heterozygous duplication, heterozygous deletion). In certain embodiments, a level is categorized as a copy number variation and then adjusted according to a specific PAV for the type of categorized copy number variation. In some embodiments a level (e.g., a first level) is categorized as a maternal copy number variation, fetal copy number variation, or a maternal copy number variation and a fetal copy number variation and is adjusted accordingly. by adding or subtracting a PAV from the level. Often, a level (for example, a first level) is categorized as a maternal copy number variation and is adjusted by adding a PAV to the level. For example, a level categorized as a duplication (e.g., a maternal, fetal, or maternal and fetal homozygous duplication) can be adjusted by adding a given PAV for a specific duplication (e.g., a homozygous duplication) thereby providing a level adjusted. Often, a PAV determined by a doubling of copy number is a negative value. In some embodiments providing an adjustment to a representative level of a duplication by using a PAV determined for a duplication results in a reduction of the level value. In some embodiments, a level (e.g., a first level) that differs significantly from a second level is categorized as a copy number deletion (e.g., a homozygous deletion, heterozygous deletion, homozygous duplication, homozygous duplication) and the first level is adjusted by adding a given PAV to a copy number deletion. Often, a given PAV for a copy number deletion is a positive value. In some embodiments, providing a fit to a representative level of a deletion using a PAV determined for a deletion results in an increase in the level value.

[328]A PAV can be any suitable value. Often, a PAV is determined by and is specific to a copy number variation (for example, a categorized copy number variation). In certain embodiments a PAV is determined according to an expected level for a copy number variation (eg, a categorized copy number variation) and/or a PAV factor. A PAV is sometimes determined by multiplying the expected level by a PAV factor. For example, a PAV for a copy number variation can be determined by multiplying the expected level determined for a copy number variation (for example, a heterozygous deletion) by a PAV factor determined for the same copy number variation ( for example, a heterozygous deletion). For example, VAP can be determined by the following formula: VAPk = (expected level}k x (PAV factor}k for copy number variation k (for example, k = a heterozygous deletion)

[329]A PAV factor can be any suitable value. In some embodiments a PAV factor for a homozygous duplication is between about -0.6 and about -0.4. In some embodiments a PAV factor for a homozygous duplication is about -0.60, -0.59, -0.58, -0.57, -0.56, -0.55, -0.54, - 0.53, -0.52, -0.51, -0.50, -0.49, -0.48, -0.47, -0.46, -0.45, -0.44, - 0.43, -0.42, -0.41 and -0.40. Often a PAV factor for a homozygous duplication is about -0.5.

[330]For example, for an NRV of about 1 and an expected level of a homozygous duplication equal to about 2, the PAV for the homozygous duplication is determined to be about -1 according to the formula above. In that case, a first level categorized as a homozygous duplication is adjusted by adding about -1 to the value of the first level, for example.

[331] In some embodiments a PAV factor for a heterozygous duplication is between about -0.4 and about -0.2. In some embodiments a PAV factor for a heterozygous duplication is about -0.40, -0.39, -0.38, -0.37, -0.36, -0.35, -0.34, - 0.33, -0.32, -0.31, -0.30, -0.29, -0.28, -0.27, -0.26, -0.25, -0.24, - 0.23, -0.22, -0.21 and -0.20. Often, a PAV factor for a heterozygous duplication is about -0.33.

[332]For example, for an NRV of about 1 and an expected level of a heterozygous duplication equal to about 1.5, the PAV for the homozygous duplication is determined to be about -0.495 according to the above formula. In this case, a first level categorized as a heterozygous duplication is adjusted by adding about -0.495 to the first level value, for example.

[333] In some embodiments a PAV factor for a heterozygous deletion is between about 0.4 and about 0.2. In some embodiments a PAV factor for a heterozygous deletion is about 0.40, 0.39, 0.38, 0.37, 0.36, 0.35, 0.34, 0.33, 0.32, 0.31, 0.30, 0.29, 0.28, 0.27, 0.26, 0.25, 0.24, 0.23, 0.22, 0.21 and 0.20. Often, a PAV factor for a heterozygous deletion is about 0.33.

[334]For example, for an NRV of about 1 and an expected level of a heterozygous deletion equal to about 0.5, the PAV for the heterozygous deletion is determined to be about 0.495 according to the above formula. In this case, a first level categorized as a heterozygous deletion is adjusted by adding about 0.495 to the first level value, for example.

[335] In some embodiments a PAV factor for a homozygous deletion is between about 0.6 and about 0.4. In some embodiments a PAV factor for a homozygous deletion is about 0.60, 0.59, 0.58, 0.57, 0.56, 0.55, 0.54, 0.53, 0.52 , 0.51, 0.50, 0.49, 0.48, 0.47, 0.46, 0.45, 0.44, 0.43, 0.42, 0.41 and 0.40. Often a PAV factor for a homozygous deletion is about 0.5.

[336]For example, for an NRV of about 1 and an expected level of a homozygous deletion equal to about 0, the PAV for the homozygous deletion is determined to be about 1 according to the above formula. In that case, a first level categorized as a homozygous deletion is adjusted by adding about 1 to the first level value, for example.

[337]In certain embodiments, a PAV is about equal to or equal to an expected level for a copy number change (eg, the expected level of a copy number change).

[338]In some embodiments, the counts of a level are normalized before making an adjustment. In certain embodiments, the counts of some or all of the levels in a profile are normalized before making an adjustment. For example, counts from a level can be normalized to counts from a reference level or an NRV. In certain embodiments, the counts of one level (e.g., a second level) are normalized to the counts of a reference level or an NRV and the counts of all other levels (e.g., a first level) in a profile are normalized against counts from the same reference level or NRV before making an adjustment.

[339]In some embodiments, a level of a profile results from one or more adjustments. In certain embodiments, a profile level is determined after one or more levels in the profile are adjusted. In some embodiments, a profile level is recalculated after one or more adjustments are made.

[340]In some embodiments, a copy number variation (e.g., a maternal copy number variation, fetal copy number variation, or a maternal copy number variation and a fetal copy number variation) is determined (for example, directly or indirectly determined) from an adjustment. For example, a level in a profile that has been adjusted (for example, an adjusted first level) may be identified as a variation of the maternal copy number. In some embodiments, the range of fit indicates the type of copy number variation (eg, heterozygous deletion, homozygous duplication, and the like). In certain embodiments, an adjusted level in a profile may be identified as representative of a copy number range in accordance with the value of a PAV for the copy number range. For example, for a given profile, PAV is about -1 for a homozygous duplication, about -0.5 for a heterozygous duplication, about 0.5 for a heterozygous deletion, and about 1 for a homozygous deletion. In the preceding example, an adjusted level of about -1 can be identified as a homozygous duplication, for example. In some embodiments, one or more copy number variations may be determined from a profile or a level comprising one or more adjustments.

[341]In certain embodiments, levels set in a profile are compared. In some modalities anomalies and errors are identified by comparing adjusted levels. For example, often one or more adjusted levels in a profile are compared and a certain level can be identified as an anomaly or error. In some embodiments an anomaly or error is identified within one or more portions that make up a level. An anomaly or error can be identified at the same level (eg in a profile) or at one or more levels that represent portions that are adjacent, contiguous, conjugate or abutting. In some embodiments one or more fitted levels are the levels of portions that are adjacent, contiguous, conjugate or abutting where the one or more fitted levels are compared and an anomaly or error is identified. An anomaly or error can be a peak or dip in a profile or level where a cause of the spike or dip is known or unknown. In certain embodiments adjusted levels are compared and an anomaly or error is identified where the anomaly or error is due to stochastic, systematic, random or user error. In some embodiments adjusted levels are compared and an anomaly or error is removed from a profile. In certain embodiments, adjusted levels are compared and an anomaly or error is adjusted.

Determination of fetal nucleic acid content

[342]The amount of fetal nucleic acid (e.g., concentration, relative amount, absolute amount, copy number, and the like) in nucleic acid is determined in some embodiments. In certain embodiments, the amount of fetal nucleic acid in a sample is referred to as the "fetal fraction". In some embodiments "fetal fraction" refers to the fraction of nucleic acid in circulating cell-free fetal nucleic acid in a sample (e.g., a blood sample, a serum sample, a plasma sample), obtained from a pregnant woman. In some embodiments, a method in which a genetic variation is determined may also comprise determining the fetal fraction. In some embodiments, the presence or absence of a genetic variation is determined according to a fetal fraction (eg, a fetal fraction for determining a sample). Fetal fraction determination can be performed in a suitable manner, non-limiting examples of which include the methods described below.

[343]Fetal fraction can be determined, in some embodiments, using methods described herein for determining fragment length. Fetal cell-free nucleic acid fragments are generally shorter than maternally derived nucleic acid fragments (see, for example, Chan et al., (2004) Clin. Chem. 50:88-92; Lo et al. , (2010) Med Sci Transl. 2:61ra91). Thus, fetal fraction can be determined, in some embodiments, by counting fragments under a particular length threshold and comparing the counts to the amount of total nucleic acid in the sample. Methods for counting nucleic acid fragments of a particular length are described in more detail below.

[344]In certain embodiments, the amount of fetal nucleic acid is determined according to markers specific to a male fetus (e.g., Y-chromosome STR markers (e.g., DYS 19, DYS 385, DYS 392 markers); marker of RhD in RhD-negative women), the allelic proportions of polymorphic sequences according to one or more specific markers for fetal nucleic acid and non-maternal nucleic acid (e.g., differential epigenetic biomarkers (e.g., methylation; described in more detail below) ) between mother and fetus, or markers of fetal RNA in maternal blood plasma (see e.g. Lo, 2005, Journal of Histochemistry and Cytochemistry 53(3): 293-296)).

[345]Determination of fetal nucleic acid content (eg, fetal fraction) is sometimes performed using a fetal quantifier assay (FQA), as described, for example, in US patent application publication no. 2010/0105049, which is incorporated herein by reference. This type of assay allows the detection and quantification of fetal nucleic acid in a maternal sample based on the methylation pattern of the nucleic acid in the sample. In certain embodiments, the amount of fetal nucleic acid in a maternal sample can be determined relative to the total amount of nucleic acid present, thereby providing the percentage of nucleic acid in the fetal sample. In certain embodiments, the fetal nucleic acid copy number can be determined in a maternal sample. In certain embodiments, the amount of fetal nucleic acid can be determined in a sequence-specific (or moiety-specific) manner and sometimes with sufficient sensitivity to allow accurate chromosomal dosage analysis (e.g., to detect the presence or absence of a fetal aneuploidy).

[346]A fetal quantifier assay (FQA) can be performed in conjunction with any of the methods described here. Such an assay can be performed by any method known in the art and/or described in US patent application publication no. 2010/0105049, such as, for example, by a method that can distinguish between maternal and fetal DNA based on differential methylation status, and quantify (i.e., determine the amount of) fetal DNA. Methods for differentiating nucleic acid based on methylation state include, but are not limited to, methylation-sensitive capture, for example, using an MBD2-Fc fragment in which the methyl binding domain of MBD2 is fused to the Fc fragment of an antibody (MBD-FC) ( Gebhard et al., (2006) Cancer Res 66(12): 6118-28 .); methylation-specific antibodies; bisulfite conversion methods, eg MSP (methylation sensitive PCR), COBRA, methylation sensitive nucleotide primer extension, or Sequenom MassCLEAVE™ technology (Ms-SnuPE); and the use of methylation-sensitive restriction enzymes (eg, digestion of maternal DNA in a maternal sample using one or more methylation-sensitive restriction enzymes thereby enriching the fetal DNA). Methyl-sensitive enzymes can also be used to differentiate nucleic acid on the basis of methylation status, which, for example, can preferentially or substantially cleave or digest its DNA recognition sequence, if the latter is unmethylated. Thus, an unmethylated DNA sample will be cut into smaller fragments than a methylated DNA sample, and a hypermethylated DNA sample will not be cleaved. Except where explicitly stated, any method for differentiating nucleic acid based on methylation state can be used with the compositions and methods of the technology herein. The amount of fetal DNA can be determined, for example, by introducing one or more competitors at known concentrations during an amplification reaction. Determination of the amount of fetal DNA can also be done, for example, by RT-PCR, primer extension, sequencing and/or counting. In certain cases, the amount of nucleic acid can be determined using BEAMing technology as described in US Patent Application Publication No. 2007/0065823. In certain embodiments, the restriction efficiency can be determined and the efficiency rate is used to further determine the amount of fetal DNA.

[347]In certain embodiments, a fetal quantifier assay (FQA) can be used to determine the concentration of fetal DNA in a maternal sample, for example, by the following method: a) determining the total amount of DNA present in a maternal sample; b) selectively digesting maternal DNA in a maternal sample using one or more methylation-sensitive restriction enzymes thereby enriching the fetal DNA; c) determine the amount of fetal DNA from step b); and d) comparing the amount of fetal DNA from step c) with the total amount of DNA from step a), thereby determining the concentration of fetal DNA in the maternal sample. In certain embodiments, the absolute copy number of fetal nucleic acid in a maternal sample can be determined, for example, using mass spectrometry and/or a system that uses a competitive PCR approach for absolute copy number measurements. See, for example, Ding and Cantor (2003) US PNAS 100: 30593064, and US Patent Application Publication No. 2004/0081993, both of which are incorporated herein by reference.

[348] In certain embodiments, fetal fraction can be determined based on the allelic ratios of polymorphic sequences (e.g., single nucleotide polymorphisms (SNP)), such as, for example, using a method described in the publication of the US patent application At the. 2011/0224087, which is incorporated herein by reference. In such a method, the fragments (reads) of nucleotide sequences are obtained for a maternal sample and fetal fraction are determined by comparing the total number of fragments (reads) of nucleotide sequences that map to a first allele and the total number of fragments ( reads) of nucleotide sequences that map a second allele to an informative polymorphic site (e.g., the SNP) in a reference genome. In certain embodiments, fetal alleles are identified, for example, by their relative minor contribution to the mixture of fetal and maternal nucleic acid in the sample as compared to the greater contribution to the mixture by maternal nucleic acids. Thus, the relative abundance of fetal nucleic acid in a maternal sample can be determined as a parameter of the total number of fragments (reads) of a sequence mapped to a target nucleic acid sequence in a reference genome for each of the two alleles. of a polymorphic site.

[349] The amount of fetal nucleic acid in extracellular nucleic acid can be quantified and used in conjunction with a method provided herein. Thus, in certain embodiments, the methods of technology described herein comprise an additional step of determining the amount of fetal nucleic acid. The amount of fetal nucleic acid can be determined in a nucleic acid sample from a subject, either before or after processing to prepare the nucleic acid from the sample. In certain embodiments, the amount of fetal nucleic acid is determined in a sample after the nucleic acid sample is processed and prepared, where the amount is used for further evaluation. In some embodiments, a result comprises adjusting the fraction of fetal nucleic acid in the sample nucleic acid (e.g., adjusting counts, removing samples, binding or not binding).

[350]The determination step may be performed before, during, at any point in a method described here, or after certain methods (eg, aneuploidy detection, fetal gender determination) described herein. For example, to achieve a fetal or aneuploidy gender determination method with a given sensitivity or specificity, a fetal nucleic acid quantification method can be performed before, during, or after the fetal or aneuploidy gender determination method to identify samples greater than about 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, 15% , 16%, 17%, 18%, 19%, 20%, 21%, 22%, 23%, 24%, 25% or more fetal nucleic acid. In some embodiments, samples determined to have a certain threshold amount of fetal nucleic acid (e.g., about 15% or more fetal nucleic acid; about 4% or more fetal nucleic acid) are further analyzed by determining fetal gender. or aneuploidy, or the presence or absence of aneuploidy or genetic variation, for example. In certain embodiments, determinations of, for example, fetal gender or the presence or absence of aneuploidy are screened (e.g., screened and reported to a patient) only for those samples that have a certain threshold amount of fetal nucleic acid (e.g., about 15% or more fetal nucleic acid; about 4% or more fetal nucleic acid).

[351] In some embodiments, determination of the fetal fraction or determination of the amount of acidic nucleic acid is not required or necessary to identify the presence or absence of an aneuploidy chromosome. In some embodiments, identifying the presence or absence of an aneuploidy chromosome does not require differentiation of fetal versus maternal DNA sequence. In certain embodiments this is because the summary contribution of both maternal and fetal sequences on a particular chromosome, chromosome portion or segment thereof is analyzed. In some embodiments, the identification of the presence or absence of an aneuploidy chromosome is not based on a priori sequence information that would allow distinguishing fetal DNA from maternal DNA.

Determination of the fetal fraction based on the level

[352]In some embodiments, a fetal fraction is determined according to a level categorized as representative of a range of maternal and/or fetal copy number. For example, determining fetal fraction often involves assessing an expected level for a maternal and/or fetal copy number variation used for fetal fraction determination. In some embodiments a fetal fraction is determined by a level (e.g., a first level) categorized as representative of a copy number variation within a range of expected level determined for the same type of copy number variation. Often, a fetal fraction is determined according to an observed level that is within a range of the expected level and is therefore categorized as a maternal and/or fetal copy number variation. In some embodiments a fetal fraction is determined when an observed level (e.g., a first level) categorized as a maternal and/or fetal copy number change is different from the expected level determined for the same maternal and/or fetal copy number change. or fetal.

[353]In some embodiments a level (e.g., a first level, an observed level), is significantly different from a second level, the first level is categorized as a maternal and/or fetal copy number variation, and a fraction fetus is determined according to the first level. In some embodiments a first level is an observed and/or experimentally obtained level that is significantly different from a second level in a profile and a fetal fraction is determined according to the first level. In some modalities the first level is an arithmetic mean, mean or summed level and a fetal fraction is determined according to the first level. In certain embodiments a first level and a second level are observed and/or experimentally obtained levels and a fetal fraction is determined according to the first level. In some cases, a first level comprises normalized counts for a first set of portions and a second level comprises normalized counts for a second set of portions and a fetal fraction is determined in accordance with the first level. In some embodiments a first set of portions of a first tier includes a copy number range (e.g., the first tier is representative of a copy number range), and a fetal fraction is determined in accordance with the first tier. In some embodiments the first set of portions of a first level includes a homozygous or heterozygous maternal copy number variation and a fetal fraction is determined according to the first level. In some embodiments a profile comprises a first tier for a first set of portions and a second tier for a second set of portions, the second set of portions including substantially any copy number variation (e.g., a maternal copy number variation , fetal copy number range, or a maternal copy number range and a fetal copy number range) and a fetal fraction is determined according to the first level.

[354]In some embodiments, a level (e.g., a first level, an observed level), is significantly different from a second level, the first level is categorized as a maternal and/or fetal copy number variation, and a Fetal fraction is determined according to the first level and/or an expected level of copy number variation. In some embodiments a first level is categorized as for a copy number variation according to an expected level for a copy number variation and a fetal fraction is determined according to a difference between the first level and the expected level. In certain embodiments a level (e.g., a first level, an observed level) is categorized as a maternal and/or fetal copy number variation, and a fetal fraction is determined as twice the difference between the first level and expected level of the copy number variation. In some embodiments a level (e.g. a first level, an observed level) is categorized as a maternal and/or fetal copy number variation, the first level is subtracted from the expected level thereby providing a difference, and a fetal fraction is determined as twice the difference. In some embodiments a level (e.g. a first level, an observed level) is categorized as a maternal and/or fetal copy number variation, the expected level is subtracted from a first level thereby providing a difference, and the fetal fraction is determined as twice the difference.

[355]Often, a fetal fraction is given as a percentage. For example, a fetal fraction can be divided by 100 thereby providing a percentage value. For example, for a first level representative of a homozygous maternal duplication and having a level of 155 and an expected level for a homozygous maternal duplication having a level of 150, a fetal fraction can be determined as 10% (e.g., (fetal fraction = 2X(155 - 150)).

[356]In some embodiments a fetal fraction is determined from two or more levels within a profile that are categorized as copy number variations. For example, sometimes two or more levels (e.g., two or more first levels) in a profile are identified as significantly different from a reference level (e.g., a second level, a level that does not substantially include number variation). of copy number), the two or more levels are categorized as representative of a range of maternal and/or fetal copy number, and a fetal fraction is determined from each of the two or more levels. In some embodiments a fetal fraction is determined to be about 3 or more, about 4 or more, about 5 or more, about 6 or more, about 7 or more, about 8 or more, or about 9 or more. more fetal fraction determinations within a profile. In some embodiments a fetal fraction is determined from about 10 or more, about 20 or more, about 30 or more, about 40 or more, about 50 or more, about 60 or more, about 70 or more, about 80 or more, or about 90 or more fetal fraction determinations within a profile. In some embodiments a fetal fraction is determined from about 100 or more, about 200 or more, about 300 or more, about 400 or more, about 500 or more, about 600 or more, about 700 or more, about 800 or more, about 900 or more, or about 1000 or more fetal fraction determinations within a profile. In some embodiments a fetal fraction is determined from about 10 to about 1000, about 20 to about 900, about 30 to about 700, about 40 to about 600, about 50 to about 500 , about 50 to about 400, about 50 to about 300, about 50 to about 200, or about 50 to about 100 fetal fraction determinations within one profile.

[357]In some modalities a fetal fraction is determined as the arithmetic mean or average of determinations of multiple fetal fractions within a profile. In certain embodiments, a fetal fraction determined from multiple fetal fraction determinations is an average (e.g., an arithmetic mean, mean, standard mean, median, or the like) of multiple fetal fraction determinations. Often, a fetal fraction determined from multiple fetal fraction determinations is an average value determined by a suitable method known in the art or described herein. In some embodiments an average value of a fetal fraction determination is a weighted average. In some embodiments an average value of a fetal fraction determination is an unweighted average. A mean, median, or arithmetic mean fetal fraction determination (i.e., a mean, median, or arithmetic mean fetal fraction determination value) generated from multiple fetal fraction determinations is sometimes associated with an uncertainty value (for example, , a deviation, standard deviation, MAD, or the like). Before determining a mean, median, or arithmetic mean fetal fraction value from multiple determinations, one or more deviant determinations are removed in some embodiments (described in more detail here).

[358]Some fetal fraction determinations within a profile are sometimes not included in the total determination of a fetal fraction (eg, determination of mean fetal fraction or arithmetic mean). In some embodiments a fetal fraction determination is derived from a first level (eg, a first level that is significantly different from a second level) in a profile and the first level is not indicative of a genetic variation. For example, some early levels (eg peaks or troughs) in a profile are generated from anomalies or unknown causes. Such values often generate fetal fraction determinations that differ significantly from other fetal fraction determinations obtained from true copy number variations. In some embodiments, fetal fraction determinations that differ significantly from other fetal fraction determinations in a profile are identified and removed from a fetal fraction determination. For example, some fetal fraction determinations obtained from anomalous peaks and troughs are identified by comparison to other fetal fraction determinations within a profile and are excluded from the total fetal fraction determination.

[359]In some embodiments, an independent fetal fraction determination that differs significantly from a mean, median, or arithmetic mean fetal fraction determination is an identified, recognized, and/or observable difference. In certain embodiments, the term "significantly differs" can mean statistically different and/or a statistically significant difference. An "independent" fetal fraction determination may be a determined fetal fraction (eg, in some embodiments a single determination) of a specific level categorized as a copy number variation. Any suitable threshold or range may be used to determine that a fetal fraction determination differs significantly from a mean, median, or arithmetic mean fetal fraction determination. In certain embodiments, a fetal fraction determination differs significantly from a mean, median, or arithmetic mean fetal fraction determination, and the determination may be expressed as a percent deviation from the mean or arithmetic mean value. In certain embodiments, a fetal fraction determination that differs significantly from a mean, median, or arithmetic mean fetal fraction determination differs by about 10 percent or more. In some embodiments, a fetal fraction determination that differs significantly from a mean, median, or arithmetic mean fetal fraction determination differs by about 15 percent or more. In some embodiments, a fetal fraction determination that differs significantly from a mean, median, or arithmetic mean fetal fraction determination differs by about 15% to about 100% or more.

[360]In certain embodiments a determination of fetal fraction differs significantly from a determination of mean, median, or arithmetic mean fetal fraction according to a multiple of an uncertainty value associated with the determination of mean fetal fraction or arithmetic mean. Often an uncertainty value and constant n (eg a confidence interval) defines a range (eg an uncertainty cutoff). For example, sometimes an uncertainty value is one standard deviation for fetal fraction determinations (e.g. +/- 5) and is multiplied by a constant n (e.g. a confidence interval), thereby defining a range or uncertainty cutoff (eg, 5n to -5n, sometimes referred to as 5 sigma). In some embodiments, an independent fetal fraction determination is outside a range defined by an uncertainty cutoff and is considered significantly different from a mean, median, or arithmetic mean fetal fraction determination. For example, for a mean value of 10 and an uncertainty cutoff of 3, an independent fetal fraction greater than 13 or less than 7 is significantly different. In some embodiments, a fetal fraction determination that differs significantly from a mean, median, or arithmetic mean fetal fraction determination differs by more than n times the uncertainty value (e.g., n x sigma) where n is approximately equal to or greater than 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10. In some embodiments, a fetal fraction determination that differs significantly from a fetal mean, median, or arithmetic mean determination differs by more than n times the uncertainty value (e.g. sigma n x) where n is approximately equal to or greater than 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8 , 1.9, 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3.0, 3 ,1' 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, or 4.0.

[361] In some embodiments, a level is representative of a fetal and/or maternal microploidy. In some embodiments a level (e.g., a first level, an observed level), is significantly different from a second level, the first level is categorized as a maternal and/or fetal copy number variation, and the first level and/or or at the second level is representative of fetal microploidy and/or maternal microploidy. In certain embodiments a first level is representative of a fetal microploidy, in some embodiments a first level is representative of a maternal microploidy. Often a first level is representative of a fetal microploidy and a maternal microploidy. In some embodiments a level (e.g., a first level, an observed level) is significantly different from a second level, the first level is categorized as a maternal and/or fetal copy number variation, the first level is representative of a maternal and/or microploidy and a fetal fraction is a fraction determined according to fetal and/or maternal microploidy. In some cases, a first level is categorized as a maternal and/or fetal copy number variation, the first level is representative of a fetal microploidy, and a fetal fraction is determined according to the fetal microploidy. In some embodiments a first level is categorized as a maternal and/or fetal copy number variation, the first level is representative of a maternal microploidy, and a fetal fraction is determined according to the maternal microploidy. In some embodiments a first level is categorized as a maternal and/or fetal copy number variation, the first level is representative of a maternal and fetal microploidy, and a fetal fraction is determined according to maternal and fetal microploidy.

[362]In some embodiments, determining a fetal fraction comprises determining a fetal and/or maternal microploidy. In some embodiments a level (e.g. a first level, an observed level) is significantly different from a second level, the first level is categorized as a maternal and/or fetal copy number variation, a maternal and/or fetal microploidy is determined according to the first level and/or second level and a fetal fraction is determined. In some embodiments a first level is categorized as a maternal and/or fetal copy number variation, a fetal microploidy is determined according to the first level and/or second level, and a fetal fraction is determined according to the fetal microploidy. In certain embodiments a first level is categorized as a maternal and/or fetal copy number variation, a maternal microploidy is determined according to the first level and/or second level, and a fetal fraction is determined according to the maternal microploidy. In some embodiments a first level is categorized as a maternal and/or fetal copy number variation, a maternal fetal microploidy is determined according to the first level and/or second level, and a fetal fraction is determined according to the maternal microploidy and fetal.

[363] A fetal fraction is often determined when the microploidy of the mother is different from (eg, not the same as) the microploidy of the fetus for a given level or for a level categorized as a copy number variation. In some embodiments a fetal fraction is determined when the mother is homozygous for one duplication (eg, a microploidy of 2) and the fetus is heterozygous for the same duplication (eg, a microploidy of 1.5). In some embodiments a fetal fraction is determined when the mother is heterozygous for a duplication (eg, a microploidy of 1.5) and the fetus is homozygous for the same duplication (eg, a microploidy of 2) or the duplication is absent in the fetus (eg, a microploidy of 1). In some embodiments a fetal fraction is determined when the mother is homozygous for a deletion (eg, a microploidy of 0) and the fetus is heterozygous for the same deletion (eg, a microploidy of 0.5). In some embodiments a fetal fraction is determined when the mother is heterozygous for a deletion (e.g., a microploidy of 0.5) and the fetus is homozygous for the same deletion (e.g., a microploidy of 0), or the deletion is absent in the fetus (eg, a microploidy of 1).

[364] In certain embodiments, a fetal fraction cannot be determined, when the microploidy of the mother is the same (eg, identified as the same) as the microploidy of the fetus for a given level identified as a copy number variation . For example, for a given level where both the mother and fetus carry the same number of copies of a copy number variation, a fetal fraction is not determined, in some embodiments. For example, a fetal fraction cannot be determined for a level categorized as a copy number variation when both the mother and the fetus are homozygous for the same deletion or homozygous for the same duplication. In certain embodiments, a fetal fraction cannot be determined for a level categorized as a copy number variation when both the mother and the fetus are heterozygous for the same deletion or heterozygous for the same duplication. In arrangements where multiple fetal fraction determinations are made for a sample, determinations that deviate significantly from a mean, median, or arithmetic mean value may result from copy number variation for which maternal ploidy equals fetal ploidy, and such determinations may be removed from consideration.

[365] In some embodiments the microploidy of a maternal copy number variation and fetal copy number variation is unknown. In some embodiments, in cases where there is no maternal and/or fetal microploidy determination for a copy number variation, a fetal fraction is generated and compared to a mean, median, or arithmetic mean fetal fraction determination. A fetal fraction determination for a copy number variation that differs significantly from a mean, median, or arithmetic mean fetal fraction determination is sometimes because the microploidy of the mother and fetus are the same for the copy number variation. . A fetal fraction determination that differs significantly from a mean, median, or arithmetic mean fetal fraction determination is often excluded from a total fetal fraction determination regardless of the source or cause of the difference. In some embodiments, the microploidy of the mother and/or the fetus is determined and/or verified by a method known in the art (e.g., by target sequencing methods).

Additional methods of determining the fetal fraction

[366]Fetal fraction (eg, for a sample) may be determined, in some embodiments, according to portion-specific fraction estimates. Without being limited by theory, it was determined here that the amount of fragments (reads) of fetal CCF fragments (e.g., fragments of a particular length, or range of lengths) maps with frequencies that vary for portions (e.g., within same sample, for example, within the same sequencing run). Furthermore, without being bound by theory, it was determined here that certain portions, when compared across multiple samples, tend to have a similar representation of fragments (reads) from fetal CCF fragments (e.g., fragments with a particular length, or range of lengths), and that the representation correlates with portion-specific fetal fraction (eg, the relative amount, percentage, or proportion of CCF fragments originating from a fetus).

[367] In some modalities estimates of the fetal portion-specific fraction are determined based, in part, on the portion-specific parameters and their relationship to the fetal fraction. Portion-specific parameters can be any suitable parameter that is reflective of (e.g., correlates to) the amount or proportion of fragments (reads) of CCF fragment lengths of a particular size (e.g., size range) in a portion. A specific portion parameter can be an average, arithmetic mean, or median of specific portion parameters determined for multiple samples. Any suitable portion specific parameter can be used. Non-limiting examples of specific portion parameters include FLR (e.g., FRS), a number of fragments (reads) having a length less than a selected fragment length, genomic coverage (i.e. coverage), mappability, counts ( (e.g., fragment counts (reads) of sequences mapped to the portion, e.g., normalized counts, normalized PERUN counts), DNasel sensitivity, methylation status, acetylation, histone distribution, guanine-cytosine (GC) content, chromatin structure, the similar or combinations thereof. A portion specific parameter can be any suitable parameter that correlates to FLR and/or FRS in a portion specific manner. In some embodiments, some or all of the specific portion parameters are a direct or indirect representation of a FLR for a portion. In some embodiments specific serving parameter is not guanine-cytosine (GC) content.

[368]In some embodiments a specific chunk parameter is any suitable value representing, correlated with or proportional to a number of fragments (reads) of CCF fragments, where the fragments (reads) mapped to a chunk have a length less than a length of the selected fragment. In certain embodiments, a specific portion parameter is a representation of the amount of fragments (reads) derived from relatively short fragments of CCF (e.g., about 200 base pairs or less) that map a portion. CCF fragments having a length less than a selected fragment length are often relatively short CCF fragments, and sometimes a selected fragment length is about 200 base pairs or less (e.g., CCF fragments that are about 190, 180, 170, 160, 150, 140, 130, 120, 110, 100, 90, or 80 bases in length). The length of a CCF fragment or a fragment (read) derived from a CCF fragment can be determined (e.g., deduced or inferred) by any suitable method (e.g., a sequencing method, a hybridization method). In some embodiments the length of a CCF fragment is determined (e.g., deduced or inferred) by a fragment (read) obtained from a paired-end sequencing method. In certain embodiments, the length of a CCF fragment template is determined directly from the length of a fragment (read) derived from the CCF fragment (e.g., one-end read).

[369]Specific portion parameters may be weighted or adjusted by one or more weighting factors. In some embodiments weighted or adjusted portion-specific parameters may provide estimates of portion-specific fetal fraction for a sample (eg, a test sample). In some embodiments, weighting or fitting generally converts a portion counts (e.g., reads mapped to a portion) or other portion-specific parameter into an estimate of the portion-specific fetal fraction, and such a conversion sometimes , is considered a transformation.

[370]In some embodiments a weighting factor is a coefficient or constant that, in part, describes and/or defines a relationship between a fetal fraction (eg, a fetal fraction determined from multiple samples) and a specific portion parameter for multiple samples (e.g. a training set). In some embodiments a weighting factor is determined according to a ratio for various fetal fraction determinations and various portion-specific parameters. A relationship can be defined by one or more weighting factors, and one or more weighting factors can be determined from a relationship. In some embodiments a weighting factor (e.g., one or more weighting factors) is determined from an adjusted ratio for a portion according to (i) a determined fetal nucleic acid fraction for each of several samples, and (ii) a specific portion parameter for multiple samples.

[371]A weighting factor can be any coefficient, estimated coefficient, or suitable constant derived from a suitable relationship (for example, a mathematical relationship, an algebraic relationship, a fitted relationship, a regression, a regression analysis, a model suitable regression tests). A weighting factor can be determined according to, derived from, or estimated from a suitable ratio. In some embodiments weighting factors are estimated coefficients of an adjusted relationship. Fitting a relation to multiple samples is sometimes referred to here as training a model. Any suitable model and/or method of fitting a relationship (eg training a model to a training set) can be used. Non-limiting examples of a suitable model that may be used include a regression model, linear regression model, simple regression model, ordinary least squares regression model, multiple regression model, general multiple regression model, regression model polynomial, general linear model, generalized linear model, discrete choice regression model, logistic regression model, multinomial logit model, mixed logit model, probit model, multinomial probit model, ordered logit model, ordered probit model , Poisson model, multivariate response regression model, multilevel model, fixed effects model, random effects model, mixed model, nonlinear regression model, nonparametric model, semiparametric model, robust model, quantile model, isotonic model , principal components model, minimum angle model, local model, segmented model, and errors-in-variables model. In some embodiments a fitted relationship is not a regression model. In some embodiments fitted relations are chosen from a decision tree model, support vector machine model and neural network model. The training result of a model (eg a regression model, a relation) is often a relation that can be described mathematically, where the relation comprises one or more coefficients (eg weighting factors). For example, for a linear least squares model, a general multiple regression model can be trained using fetal fraction values and a specific portion parameter (e.g. a coverage, for example, see Example 7) resulting in a relationship described by equation (30), where the weighting factor 13 is further defined in equations (31), (32) and (33). More complex multivariate models can determine one, two, three or more weighting factors. In some embodiments a model is trained on fetal fraction and two or more portion-specific parameters (e.g. coefficients) obtained from multiple samples (e.g. relationships fitted to multiple samples, e.g. by a matrix) .

[372]A weighting factor may be derived from a suitable relationship (e.g., a suitable mathematical relationship, an algebraic relationship, a fitted relationship, a regression, a regression analysis, a regression model) by a suitable method . In some embodiments fitted relationships are fitted by an estimate, non-limiting examples of which include least squares, ordinary least squares, linear, partial, total, generalized, weighted, nonlinear, interactively reweighted, Ridge regression, minimum absolute deviations, Bayesian , multivariate, low-rank Bayesian, LASSO, Weighted Point Selection Criteria (WRSC), Point Selection Criteria (RSC), an elastic network estimator (eg, an elastic network regression), and combinations thereof.

[373]A weighting factor can be any suitable value. In some embodiments a weighting factor is between about -1 x 10-2 and about 1 x 10-2, between about -1 x 10-3 and about 1 x 10-3, between about -5 x 10-4 and about 5 x 10-4, or between about -1 x 10-4 and about 1 x 10-4. In some embodiments, the distribution of weighting factors for various samples is substantially symmetric. A distribution of weighting factors for multiple samples is sometimes a normal distribution. A distribution of weighting factors for multiple samples is sometimes not a normal distribution. In some embodiments the width of a distribution of weighting factors is dependent on the amount of fragments (reads) of fetal CCF nucleic acid fragments. In some embodiments portions comprising greater fetal nucleic acid content generate greater coefficients (e.g. positive or negative e.g. see Figure 31). A weighting factor can be zero or a weighting factor can be greater than zero. In some embodiments about 70% or more, about 75% or more, about 80% or more, about 85% or more, about 90% or more, about 95% or more, or about 98% or more of the weighting factors for a serving are greater than zero.

[374]A weighting factor may be determined by or associated with any suitable portion of a genome. A weighting factor can be determined by or associated with any suitable portion of any suitable chromosome. In some embodiments a weighting factor is determined by or associated with some or all portions in a genome. In some embodiments a weighting factor is determined by or associated with portions of some or all chromosomes of a genome. A weighting factor is sometimes determined by or associated with portions of selected chromosomes. A weighting factor can be determined by or associated with portions of one or more autosomes. A weighting factor can be determined by or associated with the portions of a plurality of portions that include portions of autosomes or a subset thereof. In some embodiments a weighting factor is determined by or associated with portions of a sex chromosome (eg, ChrX and/or ChrY). A weighting factor can be determined by or associated with portions of one or more autosomes and one or more sex chromosomes. In certain embodiments a weighting factor is determined by or associated with the portions of a plurality of portions on all autosomes and X and Y chromosomes. A weighting factor may be determined by or associated with the portions of a plurality of portions that do not include portions on an X and/or Y chromosome. In certain embodiments a weighting factor is determined by or associated with portions of a chromosome where the chromosome comprises an aneuploidy (eg, a whole chromosome aneuploidy). In certain embodiments a weighting factor is determined by or associated only with portions of a chromosome in which the chromosome is not aneuploidy (eg, an euploid chromosome). A weighting factor can be determined by or associated with portions on a plurality of portions that do not include portions on chromosomes 13, 18 and/or 21.

[375]In some embodiments a weighting factor is determined by a portion according to one or more samples (eg, a training set of samples). The weighting factors are often portion-specific. In some embodiments one or more weighting factors are independently assigned to a portion. In some embodiments a weighting factor is determined according to a ratio of a fetal fraction determination (e.g., a sample-specific fetal fraction determination) for multiple samples and a specific portion parameter determined across multiple samples. Weighting factors are often determined from multiple samples, for example, from about 20 to about 100,000 or more, from about 100 to about 100,000 or more, from about 500 to about 100,000 or more, from from about 1000 to about 100,000 or more, or from about 10,000 to about 100,000 or more samples. Weighting factors can be determined from samples that are euploid (eg samples from subjects comprising an euploid fetus, eg samples where no aneuploid chromosome is present). In some embodiments weighting factors are obtained from samples comprising an aneuploid chromosome (eg, samples from subjects comprising an euploid fetus). In some embodiments weighting factors are determined from multiple samples of subjects having a euploid fetus and subjects having a fetus with trisomy. Weighting factors can be derived from multiple samples where the samples are from subjects with a male fetus and/or a female fetus.

[376]A fetal fraction is often determined by one or more samples from a training set from which a weighting factor is derived. A fetal fraction from which a weighting factor is determined, sometimes a sample-specific fetal fraction determination. A fetal fraction from which a weighting factor is determined can be determined by any suitable method described herein or known in the art. In some embodiments a determination of fetal nucleic acid content (e.g., fetal fraction) is performed using a suitable fetal quantifier assay (FQA) described herein or known in the art, non-limiting examples of which include determinations of fetal fraction according to with markers specific for a male fetus, based on allelic ratios of polymorphic sequences, according to one or more markers specific for fetal nucleic acid and non-maternal nucleic acid, by use of methylation-based DNA discrimination (e.g., A Nygren, et al., (2010) Clinical Chemistry 56(10):1627-1635), by a mass spectrometry method and/or a system utilizing a competitive PCR approach, by a method described in the application publication of US patent No. 2010/0105049, which is incorporated herein by reference, the like or combinations thereof. Often, a fetal fraction is determined, in part, according to a level (eg, one or more levels of a genomic section, a level of a profile) of a Y chromosome. In some embodiments, a fetal fraction is determined according to a suitable test of a Y chromosome (for example, by comparing the amount of a fetus-specific locus (such as the SRY locus on the Y chromosome in a male pregnancy) with that of a locus on any autosome that is common to both mother and fetus through the use of quantitative real-time PCR (eg, Lo YM, et al., (1998) Am J Hum Genet 62: 768-775)).

[377]Portion-specific parameters (eg, for a test sample) may be weighted or adjusted by one or more weighting factors (eg, weighting factors derived from a training set). For example, a weighting factor can be derived for a portion according to a relationship of a specific portion parameter and fetal fraction determination for a multi-sample training set. A specific portion parameter of a test sample can be adjusted and/or weighted according to the weighting factor derived from the training set. In some embodiments, a portion-specific parameter from which a weighting factor is derived is the same as the portion-specific parameter (e.g., of a test sample) that is adjusted or weighted (e.g., both parameters are a FLR). In one embodiment, a specific portion parameter, from which a weighting factor is derived, is different from the specific portion parameter (e.g., of a test sample) that is adjusted or weighted. For example, a weighting factor can be determined from a relationship between coverage (i.e., a specific portion parameter) and fetal fraction for a training set of samples, and a FLR (i.e., another portion parameter). specific) for a portion of a test sample can be adjusted according to the coverage-derived weighting factor. Without being bound by theory, a specific portion parameter (e.g. for a test sample) may sometimes be adjusted and/or weighted by a weighting factor derived from a different portion specific parameter (e.g. a training set), due to a relationship and/or correlation between each portion-specific parameter and a portion-specific FLR.

[378]An estimate of the portion-specific fetal fraction can be determined for a sample (eg, a test sample) by weighting a portion-specific parameter by a given weighting factor for that portion. Weighting may comprise adjusting, converting and/or transforming a portion-specific parameter according to a weighting factor through the application of any suitable mathematical manipulation, non-limiting examples of which include multiplication, division, addition, subtraction, integration, symbolic computation, algebraic computation, algorithm, trigonometric or geometric function, transformation (for example, a Fourier transform), the like, or combinations thereof. Weighting may comprise adjusting, converting and/or transforming a portion-specific parameter according to a weighting factor of a suitable mathematical model (eg, the model presented in Example 7).

[379] In some embodiments a fetal fraction is determined for a sample according to one or more portion-specific fetal fraction estimates. In some embodiments, a fetal fraction is determined (eg, estimated) from a sample (eg, a test sample) by weighting or adjusting a portion-specific parameter for one or more portions. In certain embodiments a fraction of the fetal nucleic acid of a test sample is estimated based on adjusted counts or an adjusted subset of counts. In certain embodiments a fetal nucleic acid fraction of a test sample is estimated based on an adjusted FLR, an adjusted FRS, adjusted coverage, and/or adjusted mappability for a portion. In some embodiments, about 1 to about 500,000, about 100 to about 300,000, about 500 to about 200,000, about 1000 to about 200,000, about 1500 to about 200,000, or about 1500 to about 200,000 50,000 portion-specific parameters are weighted or adjusted.

[380]A fetal fraction (eg, for a test sample) may be determined according to various estimates of the portion-specific fetal fraction (eg, for the same test sample) by any suitable method. In some embodiments, a method for increasing the accuracy of estimating a fetal nucleic acid fraction in a test sample from a pregnant woman comprises determining one or more portion-specific fetal fraction estimates where the fetal fraction estimate for the sample is determined according to one or more portion-specific fetal fraction estimates. In some embodiments estimating or determining a fetal nucleic acid fraction of a sample (e.g., a test sample) comprises summing one or more estimates of the portion-specific fetal fraction. Sum may comprise determining an arithmetic mean, mean, median, AUC, or integral value according to portion-specific fetal fraction estimates.

[381] In some embodiments, a method for increasing the accuracy of estimating a fetal nucleic acid fraction in a test sample from a pregnant woman, comprises obtaining counts of fragments (reads) of sequences mapped to portions of a genome of reference, whose sequence reads are circulating insect cell nucleic acid fragments (reads) of a test sample from a pregnant woman, where at least a subset of the counts obtained are derived from a region of the genome that contributes with a greater number of counts derived from fetal nucleic acid relative to total counts from the region than fetal nucleic acid counts relative to the total number of counts from another region of the genome. In some embodiments an estimate of the fetal nucleic acid fraction is determined according to a subset of the portions, where the subset of the portions is selected according to the portions that map to a greater number of fetal nucleic acid-derived counts than counts of fetal nucleic acid from another moiety. In some embodiments the subset of moieties is selected according to moieties to which a greater number of counts derived from fetal nucleic acid, relative to non-fetal nucleic acid, than fetal nucleic acid counts, relative to non-fetal nucleic acid, from another portion. The counts mapped to all or a subset of the servings can be weighted, thereby providing the weighted counts. The weighted counts can be used to estimate the fetal nucleic acid fraction, and the counts can be weighted according to the portions to which a greater number of counts derived from fetal nucleic acid than fetal nucleic acid counts from another portion are mapped . In some embodiments the counts are weighted according to the portions to which a greater number of counts derived from fetal nucleic acid, relative to non-fetal nucleic acid, than counts of fetal nucleic acid, relative to acid, are mapped. non-fetal nucleic acid, from another portion.

[382]A fetal fraction may be determined for a sample (e.g., a test sample) according to various estimates of the portion-specific fetal fraction for the sample, where the portion-specific estimates are from portions of any suitable region or segment of a genome. Estimates of the portion-specific fetal fraction can be determined from one or more portions of a suitable chromosome (e.g., one or more selected chromosomes, one or more autosomes, a sex chromosome (e.g., ChrX and/or ChrY), a aneuploid chromosome, an euploid chromosome, the peer or combinations thereof).

[383] Portion-specific parameters, weighting factors, portion-specific fetal fraction estimates (e.g., weighting), and/or fetal fraction determinations may be determined by a system, machine, apparatus, human-readable storage medium non-transient computer (for example, with an executable program stored on it), the like or their combination. In certain embodiments, portion-specific parameters, weighting factors, portion-specific fetal fraction estimates (e.g., weighting), and/or fetal fraction determinations are determined (e.g., in part) by a system or machine comprising one or more microprocessors and memory. In some embodiments, portion-specific parameters, weighting factors, portion-specific fetal fraction estimates (e.g., weighting), and/or fetal fraction determinations are determined (e.g., in part) by a computer-readable storage medium. non-transient with an executable program stored therein, where the program instructs a microprocessor to perform the determination.

fetal ploidy

[384]Determining fetal ploidy, in some embodiments, is used, in part, to make a determination of the presence or absence of a genetic variation (eg, a chromosomal aneuploidy, a trisomy). A fetal ploidy can be determined, in part, from a measurement of the fetal fraction determined using a suitable fetal fraction determination method, including the methods described herein. A fetal ploidy and/or the presence of a genetic variation (eg an aneuploidy) can be determined according to a fetal fraction. In some embodiments fetal ploidy is determined according to a fetal fraction determination and equation (8), (20), (21) or a variation or derivation thereof (Example 2). In some embodiments, fetal ploidy is determined using a method described below. In some embodiments, each method described below requires a calculated reference count Fi (sometimes represented as fi) determined by a portion (i.e., a portion, i) of a multi-sample genome where the ploidy of the fetus for the portion i of the genome is euploid. In some embodiments an uncertainty value (eg, a standard deviation, o) is determined for the phi reference count. In some embodiments a reference phi count, an uncertainty value, a test sample count and/or a measured fetal fraction (F) are used to determine fetal ploidy according to a method described below. In some embodiments, a reference count (e.g., a mean, median, or arithmetic mean reference count) is normalized by a method described herein (e.g., portion normalization, GC content normalization, linear and non-linear least squares regression). -linear, LOESS, GC LOESS, LOWESS, PERUN, RM, GCRM and/or combinations thereof). In some embodiments a reference count of a segment of a genome that is euploid is equal to 1 when the reference count is normalized by PERUN. In some embodiments, both the reference count (e.g., for a fetus known to be euploid) and counts of a test sample of a portion or segment of a genome are normalized by PERUN and the reference count is equal to 1. Likewise, in some embodiments, a reference count of a portion or segment of a genome that is euploid is equal to 1 when the counts are normalized (i.e., divided by) a median of the reference count. For example, in some embodiments both the reference count (e.g., for a fetus that is euploid) and the counts of a test sample for a portion or segment of a genome are normalized by a median reference count, the count of normalized reference equals 1, and the test sample count is normalized (eg, divided by) by the median reference count. In some embodiments, both the baseline count (e.g., for a fetus that is euploid) and the test sample counts for a portion or segment of a genome are normalized by GCRM, GC, MR, or a suitable method. In some embodiments a reference count is a mean, median, or arithmetic mean reference count. A reference count is usually a normalized count for a portion (for example, a normalized genomic section level). In some embodiments a reference count and the counts for a test sample are the raw counts. A baseline count, in some embodiments, is determined from a mean, median, or arithmetic mean count profile. In some embodiments, the reference count is a calculated genomic section level. In some embodiments, a reference count of a reference sample and a count of a test sample (eg, a patient sample, eg, Yi) are normalized by the same method or process.

[385]In some embodiments a measurement of the fetal fraction (F) is determined. This fetal fraction value can then be used to determine fetal ploidy according to equation (8), a derivation or a variation thereof. In some embodiments, a negative value is returned if the fetus is euploid and a positive value is returned if the fetus is not euploid. In some embodiments a negative value indicates that the fetus is euploid for the genome segment considered. In certain embodiments, a value that is not negative indicates that the fetus comprises an aneuploidy (eg, a duplication). In certain embodiments, a value that is not negative indicates that the fetus comprises a trisomy. In certain embodiments, any positive value indicates that the fetus comprises an aneuploidy (eg, a trisomy, a duplication).

[386]In some embodiments a sum of squared residuals is determined. For example, an equation representing the sum of squared residuals derived from equation (8) is illustrated in equation (18). In some embodiments a residual sum of squares is determined from equation (8) for a ploidy value X set to a value of 1 (see equation (9)) and for a ploidy value set to a value of 3/2 ( see equation (13)). In some embodiments the sum of squared residues (equations (9) and (13)) are determined for a segment of a chromosome or genome (eg, for all portions of a reference genome i in a genome segment). For example, the sum of squared residuals (eg equations (9) and (13)) can be determined for chromosomes 21, 13, 18 or a portion thereof. In some embodiments, to determine a ploidy status of a fetus, the result of equation (13) is subtracted from equation (9) to arrive at a value, phi (for example, see equation (14)). In certain embodiments, the sign (ie, positive or negative) of the phi value determines the presence or absence of a fetal aneuploidy. In certain embodiments, a phi value (e.g., from equation (14)) that is negative indicates the absence of an aneuploidy (e.g., the fetus is euploid for portions of a reference genome i) and a phi value that is not negative indicates the presence of an aneuploidy (for example, a trisomy).

[387]In some embodiments the reference count fi, the uncertainty value for the reference count a and/or the measured fetal fraction (F) are used in equations (9) and (13) to determine the sum of squared residuals for the sum of all portions of a reference genome i. In some embodiments the reference count, the uncertainty value for reference count a and/or the measured fetal fraction (F) are used in equations (9) and (13) to determine fetal ploidy. In some embodiments the counts (e.g., normalized counts, e.g., calculated genomic section level) represented by yi for portion i, for a test sample are used to determine the ploidy status of a fetus for portion i. For example, in certain embodiments, the ploidy status for a segment of a genome is determined according to a reference count phi, an uncertainty value (e.g., from the reference count), a fetal fraction (F) determined for a test sample and yi scores determined for the test sample in which ploidy status is determined according to equation (14) or a derivation or variation thereof. In some embodiments, yi counts and/or reference counts are normalized by a method described herein (e.g., portion normalization, normalization by GC content, linear and nonlinear least squares regression, LOESS, GC LOESS, LOWESS, PERUN, RM, GCRM and their combinations). In some embodiments a fetal ploidy status (e.g., euploid, aneuploid, trisomy) for a portion or segment of a genome or chromosome is determined by the non-limiting example described above and in the Examples section.

[388]In some embodiments a fetal fraction is determined from a test sample, y counts are determined for a test sample, and both are used to determine a ploidy of a fetus from a test sample. In certain embodiments of the method described herein, the value of fetal ploidy represented by X is not fixed or assumed. In certain embodiments of the method described herein, fetal fraction F is fixed. In some embodiments, a ploidy (e.g., a ploidy value) is determined for a portion or segment of a genome according to equation (20) or (21) (Example 2). In some embodiments of this method, a ploidy value is determined, where the value is close to 1, 3/2, 5/4 or . In some embodiments a ploidy value of about 1 indicates a euploid fetus, a value of about 3/2 indicates a fetal trisomy, and in the case of twins, a value of about 5/4 indicates a fetus comprising a trisomy and the other is euploid for the portion or segment of the genome considered. Additional information regarding the determination of the presence or absence of a fetal aneuploidy from a fetal ploidy determination is discussed in another section below.

[389] In some embodiments, fetal fraction is determined, fixed as its determined value, and fetal ploidy is determined from a regression. Any suitable regression can be used, non-limiting examples of which include a linear regression, non-linear regression (e.g., a polynomial regression), and the like. In some embodiments, a linear regression is used according to equation (8), (20), (21) and/or a derivation or variation thereof. In some embodiments, linear regression is used according to a sum of squared residuals derived from equation (8), (20), (21) and/or a derivation or variation thereof. In some embodiments, fetal ploidy is determined according to equation (8), (20), (21) and/or a derivation or variation thereof and a regression is not used. In some embodiments, fetal ploidy is determined according to a sum of squared residues derived from equation (8), (20), (21) and/or a derivation or variation thereof for various portions of a reference genome i and a regression it is not used. A derivation of an equation is any variation of the equation obtained from a mathematical proof of an equation.

[390] In some embodiments a reference count phi (described earlier here), an uncertainty value a, and/or a measured fetal fraction (F) are used in equations (20) and (21) to determine a fetal ploidy. In some embodiments a reference count fi, an uncertainty value a and/or a measured fetal fraction (F) are used in equations (20) or (21) to determine a fetal ploidy X for portion i or for a sum of several portions of a reference genome i (for example, for the sum of all portions of a reference genome i for a chromosome or its segment). In some embodiments the counts (e.g., normalized counts, calculated genomic section level), represented by yi for portion i, for a test sample are used to determine the ploidy of a fetus for a segment of a genome represented by several portions of a reference genome i. For example, in certain embodiments, the X ploidy for a segment of a genome is determined according to a fi reference count, an uncertainty value, a determined fetal fraction (F) for a test sample, and the determined yi counts for the test sample in which ploidy is determined according to equation (20), (21) or a derivation or variation thereof. In some embodiments, yi counts and/or reference counts are normalized by a method described herein (e.g., portion normalization, GC content normalization, linear and nonlinear least squares regression, LOESS, GC LOESS, LOWESS, PERUN, RM, GCRM and their combinations). In some embodiments the yi counts and/or reference counts are normalized and/or processed by the same method (e.g., portion normalization, GC content normalization, linear and nonlinear least squares regression, LOESS, GC LOESS, LOWESS, PERUN, RM, GCRM, a method described herein, or combinations thereof). In some embodiments yi and fi counts are counts mapped to the same portion or segment of a chromosome or genome.

[391]The u uncertainty value can be a suitable measure of error, non-limiting examples of which include standard deviation, standard error, calculated variance, p-value, and/or mean absolute deviation (MAD). The uncertainty value can be determined for any suitable measurement, non-limiting examples of which include Z scores, Z values, t values, p values, cross validation error, genomic section level, calculated genomic section levels , levels, scores, similar, or combinations thereof. In some embodiments u is set to a value of 1. In some embodiments u is not set to a value of 1. In some embodiments the value of u is estimated and sometimes measured and/or calculated.

[392] In some embodiments Mi is the maternal ploidy (ie, maternal ploidy) for a portion of the i genome. In some embodiments Mi is determined for the same patient (eg the same test sample) from which Yi is determined. In some embodiments maternal Mi ploidy is known or determined according to a method described herein. In some modalities maternal ploidy is determined before or after filling in (for example, after making level adjustments). In certain modalities Mi is estimated or determined from a visualization profile. In some embodiments maternal Mi ploidy is not known. In some embodiments maternal Mi ploidy is assumed. For example, in some embodiments, it is assumed or known that the mother has no deletions and/or duplications in the segment of the genome being evaluated. In some embodiments, maternal ploidy is assumed or known to be 1. In some embodiments, maternal ploidy is set to a value of 1 after padding (eg, after making level adjustments). In some embodiments maternal ploidy is ignored and is set to a value of 1. In some embodiments equation (21) is derived from equation (20) with the assumption that the mother has no deletions and/or duplications in the genome segment being rated.

[393] In some embodiments a method for determining fetal ploidy is according to nucleic acid sequence reads for a test sample obtained from a pregnant woman. In some embodiments, the sequence reads are circulating cell-free nucleic acid reads from a sample (e.g., a test sample). In some embodiments, a method for determining fetal ploidy comprises obtaining counts of fragments (reads) of sequences mapped to portions of a reference genome. In some embodiments, sequence reads are mapped to a subset of portions of the reference genome. In certain embodiments determining fetal ploidy comprises determining a fetal fraction. In some embodiments determining fetal pod comprises calculating or determining genomic section levels. In certain embodiments determining fetal ploidy comprises determining a fetal fraction and calculating or determining genomic section levels. A fetal fraction and calculated genomic section levels can be determined from the same test sample (eg, the same part of the test sample). In some embodiments the fetal fraction and calculated genomic section levels are determined from the same fragments (reads) obtained from the same test sample (eg, the same part of the test sample). In certain embodiments, the fetal fraction and calculated genomic section levels are determined from the same fragments (reads) obtained from the same sequencing run and/or from the same flow cell. In some embodiments, the fetal fraction and calculated genomic section levels are determined from the same equipment and/or machine (eg, a sequencing apparatus, flow cell, or the like).

[394] In some embodiments a method for determining fetal ploidy is determined according to fetal fraction determination and normalized counts (e.g., calculated genomic section levels), where fetal fraction determination and normalized counts (for example, (e.g., calculated genomic section levels) are determined from different parts of a test sample (e.g., different aliquots, or, for example, different test samples taken at about the same time from the same subject or patient). For example, sometimes a fetal fraction is determined from a first part of a test sample and normalized counts and/or genomic section levels are determined from a second part of the test sample. In some embodiments, fetal fraction and calculated genomic section levels are determined from different test samples (eg, different parts of a test sample) taken from the same subject (eg, patient). In some modalities the fetal fraction and calculated genomic section levels are determined from fragments (reads) obtained at different times. In some embodiments, fetal fraction determination and normalized counts (e.g., calculated genomic section levels) are determined from different equipment and/or different machines (e.g., sequencing apparatus, flow cell, or similar).

Result

[395]Methods described herein can provide a determination of the presence or absence of a genetic variation (eg, fetal aneuploidy) for a sample, thereby providing a result (eg, thereby providing a result determining the presence or absence of a genetic variation (e.g., fetal aneuploidy)). A genetic variation often includes a gain, loss, and/or alteration (e.g., duplication, deletion, fusion, insertion, mutation, rearrangement, substitution, or aberrant methylation) of genetic information (e.g., chromosomes, chromosome segments, polymorphic regions, translocated regions, altered nucleotide sequence, the like or combinations thereof) that results in a detectable change in the genome or genetic information of a test subject relative to a reference. Presence or absence of a genetic variation can be determined by transforming, analyzing and/or manipulating fragments (reads) of sequences that have been mapped to portions (eg, counts, counts of genomic portions of a reference genome). Determining an outcome, in some embodiments, comprises analyzing nucleic acid from a pregnant woman. In certain embodiments, a result is determined according to counts (e.g., normalized counts) obtained from a pregnant woman, where the counts are from nucleic acid obtained from the pregnant woman.

[396]Methods described here sometimes determine the presence or absence of a fetal aneuploidy (eg, total chromosome aneuploidy, partial chromosome aneuploidy, or segmental chromosomal aberration (eg, mosaicism, deletion and/or insertion)) for a test sample from a pregnant woman carrying a fetus. In certain embodiments methods described herein detect euploidy or the lack of euploidy (non-euploidy) for a sample from a pregnant woman carrying a fetus. Methods described here sometimes detect trisomy for one or more chromosomes (eg, chromosome 13, chromosome 18, chromosome 21, or combination of both) or segments thereof.

[397] In some embodiments, the presence or absence of a genetic variation (eg, a fetal aneuploidy) is determined by a method described herein, by a method known in the art, or by a combination thereof. Presence or absence of a genetic variation is usually determined from sequence read counts mapped to portions of a reference genome. Sequence read counts used to determine the presence or absence of genetic variation are sometimes raw counts and/or filtered counts, and often are normalized counts. A suitable normalization process or processes can be used to generate normalized counts, non-limiting examples of which include normalization on portions, normalization by GC content, linear and non-linear least squares regression, LOESS, GC LOESS, LOWESS, PERUN, RM, GCRM and their combinations. Normalized counts are sometimes expressed as one or more levels or in levels in a profile for a particular set or sets of servings. Normalized counts are sometimes adjusted or padded before determining the presence or absence of a genetic variation.

[398]In some embodiments an outcome is determined according to one or more levels. In some embodiments, the determination of the presence or absence of a genetic variation (eg, a chromosome aneuploidy) is determined according to one or more adjusted levels. In some embodiments a determination of the presence or absence of a genetic variation (e.g., a chromosome aneuploidy) is determined according to a profile comprising from 1 to about 10,000 adjusted levels. Often, the determination of the presence or absence of a genetic variation (for example, a chromosome aneuploidy) is determined according to a profile comprising about 1 to about 1000, 1 to about 900, 1 to about 800, 1 to about 700, 1 to about 600, 1 to about 500, 1 to about 400, 1 to about 300, 1 to about 200, 1 to about 100, 1 to about 50, 1 to about 25, 1 and about 20, about 1 to about 10, or 1 to about 5 settings. In some embodiments a determination of the presence or absence of a genetic variation (e.g., a chromosome aneuploidy) is determined according to a profile comprising about 1 fit (e.g., a fit level). In some embodiments a result is determined according to one or more profiles (e.g. a profile of a chromosome or its segment) comprising one or more, two or more, 3 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more times, or 10 or more adjustments. In some embodiments, determining the presence or absence of a genetic variation (eg, a chromosome aneuploidy) is determined according to a profile where some levels in a profile are not adjusted. In some embodiments, the determination of the presence or absence of a genetic variation (eg, a chromosome aneuploidy) is determined according to a profile in which adjustments are not made.

[399]In some embodiments, an adjustment of one level (eg, a first level) in a profile reduces a false determination or false result. In some embodiments, an adjustment of a level (e.g., a first level) in a profile reduces the frequency and/or the probability (e.g., the statistical probability, probability) of a false determination or false result. A false determination or result could be a determination or result that is not accurate. A false determination or result can be a determination or result that is not a reflection of the real or true genetic makeup or the real or true genetic predisposition (e.g., the presence or absence of a genetic variation) of a subject (e.g., a pregnant woman, a fetus and/or a combination thereof). In some embodiments a false determination or result is a false negative determination. In some embodiments a negative determination or negative result is the absence of a genetic variation (eg, aneuploidy, copy number variation). In some embodiments a false determination or false result is a false positive or false positive determination. In some embodiments a positive determination or positive result is the presence of a genetic variation (eg, aneuploidy, copy number variation). In some embodiments, a determination or result is used in diagnosis. In some embodiments, a determination or result is for a fetus.

[400] Presence or absence of a genetic variation (eg fetal aneuploidy) is sometimes determined by comparing counts from a set of portions to a reference. Counts measured for a test sample and which fall within a test region (eg, a set of portions of interest) are referred to as "test counts" here. Test counts are sometimes processed counts, summed counts or averages, a representation, normalized counts, or one or more levels or levels as described herein. In certain embodiments, test scores are calculated and summed (eg, an arithmetic mean, mean, median, mode, or sum is calculated) for a set of portions, and the averaged or summed counts are compared to a threshold or range. Test counts are sometimes expressed as a representation, which can be expressed as a ratio or percentage of counts for a first set of servings to counts for a second set of servings. In certain embodiments a first set of portions is for one or more test chromosomes (e.g. chromosome 13, chromosome 18, chromosome 21, or combination thereof) and sometimes a second set of portions is for a genome or a part of a genome (for example, autosomes, or autosomes and sex chromosomes). In some embodiments, a first set of portions is for one or more sex chromosomes (eg, X chromosome, Y chromosome, or combination thereof), and sometimes a second set of portions is for one or more autosomes. In some embodiments, a first set of portions is for one or more first regions of test chromosomes (e.g., X chromosome, Y chromosome, or combination thereof), and sometimes a second set of portions is for one or more second regions of a test chromosome (eg, X chromosome, Y chromosome, or a combination thereof), or the entire test chromosome. In certain embodiments a representation is compared to a threshold or range. In certain embodiments test scores are expressed as one or more levels or from levels to normalized counts over a set of portions, and the one or more levels or levels are compared to a threshold or range. Test counts (e.g. averaged or summed counts, representation, normalized counts, one or more levels or levels) above or below a particular threshold, in a particular range or outside a particular range are sometimes determinants of presence a genetic variation or lack of euploidy (eg, non-euploidy). Test counts (e.g. averaged or summed counts, representation, normalized counts, one or more levels or levels) below or above a particular threshold, within a particular range, or outside a particular range are sometimes determinants of absence of a genetic variation or euploidy.

[401] Presence or absence of a genetic variation (eg fetal aneuploidy) is sometimes determined by comparing counts, non-limiting examples of which include test counts, reference counts, raw counts, filtered counts, or counts averages or sums, representations (e.g., chromosome representations), normalized counts, one or more levels or levels (e.g., for a set of portions, e.g., profiles, genomic section levels), Z scores, the similar or their combinations. In some embodiments test counts are compared to a reference (eg, reference counts). A reference (e.g., a reference count) can be a suitable determination of counts, non-limiting examples of which include raw counts, filtered counts, averaged or summed counts, representations (e.g., chromosome representations), normalized counts, one or more levels or tiers (e.g., for a set of portions, e.g., profiles, genomic section levels), Z-scores, the like, or combinations thereof. Reference counts are often counts for an euploid test region or for a segment of a genome or chromosome that is euploid. In some embodiments, reference counts and test counts are obtained from the same sample and/or the same subject. In some embodiments reference counts are from different samples and/or from different subjects. In some embodiments, reference counts are determined from and/or compared to a corresponding segment of the genome from which test counts are derived and/or determined. A matched segment refers to a segment, portion, or set of portions that map to the same location in a reference genome. In some embodiments, reference counts are determined from and/or compared to a different segment of the genome from which test counts are derived and/or determined.

[402] In certain embodiments, test counts are sometimes for a first set of portions and a reference includes counts for a second set of portions different than the first set of portions. Reference counts are sometimes for a nucleic acid sample from the same pregnant woman from which the test sample is obtained. In certain embodiments, reference counts are for a nucleic acid sample from one or more different pregnant women than the woman from which the test sample was obtained. In some embodiments, a first set of portions is on chromosome 13, chromosome 18, chromosome 21, a segment thereof or combination thereof, and the second set of portions is on another chromosome or chromosomes or segment thereof. In a non-limiting example, where a first set of portions is on chromosome 21 or a segment thereof, a second set of portions often is on another chromosome (e.g., chromosome 1, chromosome 13, chromosome 14, chromosome 18, chromosome 19, segment of the same or combination of the previous ones). A reference is often located on a chromosome or segment thereof which is typically euploid. For example, chromosome 1 and chromosome 19 are often euploid in fetuses due to a high rate of early fetal mortality associated with chromosome 1 and chromosome 19 aneuploidies. A measure of the deviation between test counts and reference counts can be generated.

[403]In certain embodiments a reference comprises counts for the same set of portions as for test counts, where the counts for the reference are from one or more reference samples (e.g., often multiple reference samples from multiple subjects from reference). A reference sample is often from one or more different pregnant women than the woman from whom a test sample is obtained. A measure of deviation (eg an uncertainty measure, an uncertainty value) between the test counts and the reference counts can be generated. In some embodiments a measure of deviation is determined from test scores. In some embodiments a measure of deviation is determined from the reference counts. In some embodiments a measure of deviation is determined from an entire profile or a subset of portions within a profile.

[404]A suitable measure of deviation can be selected, non-limiting examples of which include standard deviation, mean absolute deviation, median absolute deviation, maximum absolute deviation, standard score (e.g. z value, z, normal score, standardized variable) and similar. In some embodiments, the reference samples are euploid for a test region and the deviation between the test counts and the reference counts is evaluated. In some embodiments, a determination of the presence or absence of a genetic variation is based on the number of deviations (e.g., measures of deviations, MAD) between test counts and reference counts for a segment or portion of a genome. or chromosome. In some embodiments the presence of a genetic variation is determined when the number of deviations between test counts and reference counts is greater than about 1, greater than about 1.5, greater than about 2, greater than about 2.5, greater than about 2.6, greater than about 2.7, greater than about 2.8, greater than about 2.9, greater than about 3, greater than about 3.1, greater than about 3.2, greater than about 3.3, greater than about 3.4, greater than about 3.5, greater than greater than about 4, greater than about 5, or greater than about 6. For example, sometimes a test count differs from the reference count by more than 3 measures of deviation (eg, 3 Sigma, 3 MAD) and the presence of a genetic variation is determined. In some embodiments a test count obtained from a pregnant woman is greater than a reference count by more than 3 measures of deviation (e.g. 3 Sigma, 3 MAD) and the presence of a fetal chromosome aneuploidy (e.g. a fetal trisomy) is determined. A deviation of more than three between test counts and reference counts is often indicative of a non-euploid test region (eg, the presence of a genetic variation). Test counts significantly above reference counts whose reference counts are indicative of euploidy are sometimes determinants of a trisomy. In some embodiments a test count obtained from a pregnant woman is lower than a reference count by more than 3 deviation measures (e.g. 3 Sigma, 3 MAD) and the presence of a fetal chromosome aneuploidy (e.g. a fetal monosomy) is determined. Test counts significantly below reference counts, which reference counts are indicative of euploidy, are sometimes determinants of a monosomy.

[405]In some embodiments the absence of a genetic variation is determined when the number of deviations between the test counts and the reference counts is less than about 3.5, less than about 3.4, less than than about 3.3, less than about 3.2, less than about 3.1, less than about 3.0, less than about 2.9, less than about 2, 8, less than about 2.7, less than about 2.6, less than about 2.5, less than about 2.0, less than about 1.5, or less than that about 1.0. For example, test counts sometimes differ from a reference count by at least 3 deviation measures (eg 3 Sigma, 3 MAD) and the absence of a genetic variation is determined. In some embodiments of a test score obtained from a pregnant woman that differs from a reference count by at least 3 measures of deviation (e.g. 3 Sigma, 3 MAD) and the absence of a fetal chromosome aneuploidy (e.g. , a fetal euploid) is determined. In some embodiments (e.g., deviation of less than three between test counts and reference counts (e.g., 3-sigma for standard deviation), it is often indicative of a euploid test region (e.g., absence of a genetic variation.) The measure of the deviation between test counts for a test sample and reference counts for one or more reference subjects can be plotted and visualized (eg z-score plot).

[406]Any other suitable reference can be taken with test counts to determine the presence or absence of a genetic variation (or determination of euploidy or non-euploidy) for a test region of a test sample. For example, a fetal fraction determination can be taken on test counts to determine the presence or absence of a genetic variation. A suitable process for the quantification of the fetal fraction can be used, non-limiting examples of which include a mass spectrometry process, sequencing process or combination thereof.

[407] In some embodiments, the presence or absence of a fetal chromosome aneuploidy (eg, a trisomy) is determined, in part, from a determination of fetal ploidy. In some embodiments a fetal ploidy is determined by a suitable method described herein. In some embodiments a fetal ploidy determination of about 1.20 or greater, 1.25 or greater, 1.30 or greater, about 1.35 or greater, of about 1.4 or greater, or about 1 .45 or greater indicates the presence of a fetal chromosome aneuploidy (eg, the presence of a fetal trisomy). In some embodiments a fetal ploidy determination of about 1.20 to about 2.0, about 1.20 to about 1.9, about 1.20 to about 1.85, about 1.20 about 1.8, about 1.25 to about 2.0, about 1.25 to about 1.9, about 1.25 to about 1.85, about 1.25 to about from 1.8, about 1.3 to about 2.0, about 1.3 to about 1.9, about 1.3 to about 1.85, about 1.3 to about 1 .8, about 1.35 to about 2.0, about 1.35 to about 1.9, about 1.35 to about 1.8, about 1.4 to about 2.0 , about 1.4 to about 1.85 or about 1.4 to about 1.8 indicates the presence of a fetal chromosome aneuploidy (eg, the presence of a fetal trisomy). In some embodiments fetal aneuploidy is a trisomy. In some forms of fetal aneuploidy it is a trisomy of chromosome 13, 18 and/or 21.

[408] In some embodiments a fetal ploidy of less than about 1.35, less than about 1.30, less than about 1.25, less than about 1.20, or less than about 1 .15 indicates the absence of a fetal aneuploidy (for example, the absence of a fetal trisomy, for example, euploid). In some embodiments a fetal ploidy determination of about 0.7 to about 1.35, about 0.7 to about 1.30, about 0.7 to about 1.25, about 0.7 about 1.20, about 0.7 to about 1.15, about 0.75 to about 1.35, about 0.75 to about 1.30, about 0.75 to about from 1.25, about 0.75 to about 1.20, about 0.75 to about 1.15, about 0.8 to about 1.35, about 0.8 to about 1 .30, about 0.8 to about 1.25, about 0.8 to about 1.20, or about 0.8 to about 1.15 indicates the absence of a fetal chromosome aneuploidy (for (e.g., the absence of a fetal trisomy, e.g., euploid).

[409] In some embodiments a fetal ploidy of less than about 0.8, less than about 0.75, less than about 0.70, or less than about 0.6 indicates the presence of an aneuploidy fetus (for example, the presence of a chromosome deletion). In some embodiments, a fetal ploidy determination of about 0 to about 0.8, about 0 to about 0.75, about 0 to about 0.70, about 0 to about 0.65, about from 0 to about 0.60, about 0.1 to about 0.8, about 0.1 to about 0.75, about 0.1 to about 0.70, about 0.1 about 0.65, about 0.1 to about 0.60, about 0.2 to about 0.8, about 0.2 to about 0.75, about 0.2 to about from 0.70, about 0.2 to about 0.65, about 0.2 to about 0.60, about 0.25 to about 0.8, about 0.25 to about 0 .75, about 0.25 to about 0.70, about 0.25 to about 0.65, about 0.25 to about 0.60, about 0.3 to about 0.8 , about 0.3 to about 0.75, about 0.3 to about 0.70, about 0.3 to about 0.65, about 0.3 to about 0.60 indicates the presence of a fetal chromosome aneuploidy (eg, the presence of a chromosome deletion). In some arrangements determined fetal aneuploidy is a deletion of the entire chromosome.

[410]In some embodiments a determination of the presence or absence of a fetal aneuploidy (for example, according to one or more of the ranges of a ploidy determination above) is determined according to the zone of attachment. In certain embodiments a link is made (for example, a link determining the presence or absence of a genetic variation, for example, a result) when a value (for example, a ploidy value, a fetal fraction value, a degree of uncertainty) or collection of values is within a predefined range (e.g. a zone, a binding zone). In some embodiments a binding zone is defined according to a collection of values that are obtained from the same patient sample. In certain embodiments a linkage zone is defined according to a collection of values that are derived from the same chromosome or its segment. In some embodiments a linkage zone based on ploidy determination is defined according to a confidence level (eg high confidence level, eg low uncertainty level) and/or a fetal fraction. In some embodiments a binding zone is defined according to a ploidy determination and a fetal fraction of about 2.0% or greater, about 2.5% or greater, about 3% or greater, about 3, 25% or more, about 3.5% or more, about 3.75% or more, or about 4.0% or more. For example, in some embodiments a link is made where a fetus comprises a trisomy 21, based on a ploidy determination of greater than 1.25 with a fetal fraction determination of 2% or greater or 4% or greater for a sample. obtained from a pregnant woman carrying a fetus. In certain embodiments, for example, a link is made that a fetus is euploid based on a ploidy determination of less than 1.25 with a fetal fraction determination of 2% or more or 4% or more of a sample obtained from a pregnant woman bearing a fetus. In some embodiments a binding zone is defined by a confidence level of about 99% or greater, about 99.1% or greater, about 99.2% or greater, about 99.3% or greater, about of 99.4% or more, about 99.5% or greater, about 99.6% or greater, about 99.7% or greater, about 99.8% or greater, or about 99.9% or more. In some embodiments a connection is made without the use of a connection zone. In some embodiments a link is made using a link zone and additional data or information. In some embodiments a link is made based on a ploidy value without the use of a link zone. In some embodiments a link is made without calculating a ploidy value. In some embodiments a link is made based on visual inspection of a profile (eg, visual inspection of genomic section levels). A link can be made by any suitable method, based in whole or in part on the determinations, values and/or data obtained by the methods described herein, non-limiting examples of which include a determination of fetal ploidy, a determination of the fraction fetal, maternal ploidy, uncertainty and/or confidence determinations, portion levels, levels, profiles, Z scores, expected chromosome representations, measured chromosome representations, counts (e.g., normalized counts, raw counts), maternal or fetal copy (eg, categorized copy number variations), significantly different levels, adjusted levels (eg, padding), the similar, or combinations thereof.

[411]In some embodiments a binding zone is where a binding is not made. In some embodiments a binding zone is defined by a value or a set of values that indicate low accuracy, high risk, high error, low confidence, high uncertainty, the like, or a combination thereof. In some embodiments a binding zone is defined, in part, by a fetal fraction of about 5% or less, about 4% or less, about 3% or less, about 2.5% or less, about 2.0% or less, about 1.5% or less, or about 1.0% or less.

[412] In some embodiments, a method of determining the presence or absence of a genetic variation (eg, fetal aneuploidy) is performed with an accuracy of at least about 90% to about 100%. For example, the presence or absence of a genetic variation can be determined with an accuracy of at least about 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.1%, 99.2%, 99.3%, 99.4%, 99.5%, 99.6%, 99.7%, 99.8% or 99.9%. In some embodiments, the presence or absence of a genetic variation is determined with an accuracy that is about the same or greater than the accuracy using other methods of determining genetic variation (e.g., karyotype analysis). In some embodiments, the presence or absence of a genetic variation is determined with an accuracy having a confidence interval (CI) of about 80% to about 100%. For example, the confidence interval (CI) might be around 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92% , 93%, 94%, 95%, 96%, 97%, 98%, or 99%.

[413]Outcome can sometimes be determined in terms of sequence labeling density. "Sequence marker density" refers to the normalized value of fragment markers (reads) or sequence tags for a defined genomic section, where the sequence marker density is used to compare different samples and for further analysis. The sequence marker density value is often normalized within a sample. In some embodiments, normalization can be performed by counting the number of markers that are within each genomic section; obtaining an average sequence marker count value for each chromosome; obtaining an average value of all autosomal values; and using this value as a normalization constant to account for differences in the total number of sequence markers obtained for the different samples. Sequence marker density is sometimes about 1 per one disomic chromosome. Sequence marker densities can vary according to sequencing artifacts, most notably G/C bias, which can be corrected for by using an external or internal reference standard (e.g., derived from substantially all sequence markers ( genomic sequences), which can be, for example, a single chromosome or a calculated value of all autosomes, in some embodiments). Thus, the dosage imbalance of a chromosome or chromosomal regions can be inferred from the percentage representation of the locus among other mappable markers of the species. Dosage imbalance of a particular chromosome or chromosomal regions can therefore be quantitatively determined and normalized. Methods for normalizing and quantifying sequence marker density are discussed in more detail below.

[414] In some embodiments, a proportion of all sequence reads are from a sex chromosome (eg, X chromosome, Y chromosome) or a chromosome involved in aneuploidy (eg, chromosome 13, chromosome 18, chromosome 21), and other sequence reads are from other chromosomes. By taking into account the relative size of the sex chromosome or chromosome involved in aneuploidy (eg "target chromosome": chromosome 21) compared to other chromosomes, a normalized frequency, within a reference range, of specific sequences can be obtained target chromosome, in some embodiments. If the fetus has an aneuploidy, for example, on a target chromosome, then the normalized frequency of sequences derived from the target chromosome is statistically higher than the normalized frequency of sequences derived from the non-target chromosome, thereby allowing detection of the aneuploidy. The degree of alteration of the normalized frequency will be dependent on the fractional concentration of fetal nucleic acids in the analyzed sample, in some embodiments.

[415] A genetic variation is sometimes associated with the medical condition. A determinant result of a genetic variation is sometimes a determinant result of the presence or absence of a condition (e.g., a medical condition), disease, syndrome, or abnormality, or includes the detection of a condition, disease, syndrome, or abnormality (eg, non-limiting examples listed in Table 1). In certain embodiments a diagnosis comprises the evaluation of an outcome. A result determining the presence or absence of a condition (e.g., a medical condition), disease, syndrome, or abnormality by methods described herein can sometimes be independently verified by other tests (e.g., by karyotyping and/or amniocentesis) . Analyzing and processing data can provide one or more results. The term "result", as used herein, can refer to a data processing result that facilitates the determination of the presence or absence of a genetic variation (eg, an aneuploidy, a copy number variation). In certain embodiments, the term "result", as used herein, refers to a finding that predicts and/or determines the presence or absence of a genetic variation (e.g., an aneuploidy, a copy number variation). In certain embodiments, the term "result", as used herein, refers to a finding that predicts and/or determines a risk or likelihood of the presence or absence of a genetic variation (e.g., an aneuploidy, a gene variation, copy number) on a subject (for example, a fetus). A diagnosis sometimes includes the use of a result. For example, a healthcare professional can review a result and provide a diagnosis based on, or based in part on, the result. In some embodiments, determining, detecting, or diagnosing a condition, syndrome, or abnormality (eg, listed in Table 1) comprises using a result to determine the presence or absence of a genetic variation. In some embodiments, a result based on counted mapped sequence reads or transformations thereof is determinant of the presence or absence of a genetic variation. In certain modalities, a result generated using one or more methods (eg, data processing methods described herein) is determinative of the presence or absence of one or more conditions, syndromes, or abnormalities listed in Table 1. In certain modalities a diagnosis comprises a determination of the presence or absence of a condition, syndrome, or abnormality. Diagnosis often involves determining a genetic variation as the nature and/or cause of a disease, syndrome, or abnormality. In certain modalities a result is not a diagnosis. A result often comprises one or more numeric values generated using a processing method described here in the context of one or more probability considerations. A risk or probability consideration may include, but is not limited to: an uncertainty value, a measure of variability, confidence level, sensitivity, specificity, standard deviation, coefficient of variation (CV) and/or confidence level, scores Z values, Chi values, Phi values, ploidy values, adjusted fetal fraction, area proportions, mean level, the like, or combinations thereof. A probability consideration can facilitate the determination of whether a subject is at risk of having, or has, a genetic variation, and a finding determining a presence or absence of a genetic disease often includes such a consideration.

[416]An outcome is sometimes a phenotype. An outcome is sometimes a phenotype associated with a confidence level (e.g., an uncertainty value, e.g., a fetus is positive for trisomy 21 at a 99% confidence level; a pregnant woman is carrying a male fetus at a 95% confidence level; a test subject is negative for a cancer associated with a genetic variation at a 95% confidence level. Different methods of generating result values can sometimes produce different types of results. In general, there are four types of possible scores or links that can be made based on result values generated using methods described here: true positive, false positive, true negative, and false negative. The terms "score", "scores", "linkage" and "linkages" as used herein refer to calculating the probability that a particular genetic variation is present or absent in a subject/sample. The value of a score can be used to determine, for example, a variation, difference, or the proportion of fragments (reads) of mapped sequences that can correspond to a genetic variation. For example, calculating a positive score for a selected genetic variation or portion of a dataset, with respect to a reference genome can lead to an identification of the presence or absence of a genetic variation, the genetic variation that is sometimes associated with a medical condition (eg, cancer, pre-eclampsia, trisomy, monosomy, and the like). In some embodiments, a result comprises a level, a profile, and/or a graph (e.g., a profile graph). In those embodiments where a result comprises a profile, a suitable profile or combination of profiles can be used for a result. Non-limiting examples of profiles that can be used for a result include Z-score profiles, p-value profiles, chi-value profiles, phi-value profiles, peer, and combinations thereof.

[417]A result generated to determine the presence or absence of genetic variation sometimes includes a null result (for example, a data point between two sets, a numerical value with a standard deviation, encompassing the values for both the presence and absence of a genetic variation, a dataset with a profile plot that is not similar to the profile plots for subjects having or free of the genetic variation being investigated). In some embodiments, a result indicative of a null result is still a determinant result, and the determination may include the need for additional information and/or a repeat generation of and/or data analysis to determine the presence or absence of a variation. genetics.

[418] A result may be generated after performing one or more of the processing steps described herein, in some embodiments. In certain embodiments, a result is generated as a result of one of the processing steps described herein, and in some embodiments, a result may be generated after each statistical and/or mathematical manipulation of a data set is performed. A result relating to the determination of the presence or absence of a genetic variation may be expressed in an appropriate form, which form comprises, without limitation, a probability (e.g., odds ratio, p-value), likely, in or out value of a set, value above or below a threshold value, value within a range (for example, a threshold range), the value of a measure of variance or confidence, or risk factor, associated with the presence or absence of a genetic variation for a subject or sample. In certain embodiments, comparison between samples allows confirmation of sample identity (e.g., allows identification of samples and/or repeat samples that have been mixed (e.g., mislabeled, pooled, and the like)).

[419]In some embodiments, a result comprises a value above or below a predetermined threshold or cutoff value (e.g., greater than 1, less than 1), and a level of confidence or uncertainty associated with the value. In certain embodiments a predetermined threshold or cutoff value is an expected level or range of the expected level. A result can also describe an assumption used in data processing. In certain embodiments, a result comprises a value that is within or outside a predetermined range of values (e.g., a threshold range) and the associated uncertainty or confidence level for that value being within or outside the range. In some embodiments, a result comprises a value that is equal to a predetermined value (e.g., equal to 1, equal to zero), or is equal to a value within a predetermined value range, and its uncertainty. associated or confidence level for that value being equal to or within or outside a range. A result is sometimes graphically represented as a graph (for example, the profile graph).

[420]As noted above, a result can be characterized as a true positive, true negative, false positive, or false negative. The term "true positive", as used herein, refers to a subject correctly diagnosed as having a genetic variation. The term "false positive" as used herein refers to a subject misidentified as having a genetic variation. The term "true negative" as used herein refers to a subject correctly identified as not having a genetic variation. The term "false negative" as used herein refers to a subject misidentified as not having a genetic variation. Two measures of performance for any given method can be calculated based on the proportions of these occurrences: (i) a sensitivity value, which is generally the fraction of predicted positives that are correctly identified as being positive; and (ii) a specificity value, which is usually the fraction of predicted negatives correctly identified as being negative.

[421]In certain embodiments, one or more of sensitivity, specificity, and/or confidence level are expressed as a percentage. In some embodiments, the percentage, regardless of each variable, is greater than about 90% (for example, about 90, 91, 92, 93, 94, 95, 96, 97, 98, or 99%, or greater than than 99% (e.g., about 99.5% or greater, about 99.9% or greater, about 99.95% or greater, about 99.99% or greater)). Coefficient of variation (CV) in some arrangements is expressed as a percentage, and sometimes the percentage is around 10% or less (e.g. around 10, 9, 8, 7, 6, 5, 4, 3 , 2, or 1%, or less than 1% (e.g., about 0.5% or less, about 0.1% or less, about 0.05% or less, about 0.01% or less )). A probability (eg, that a particular result is not due to chance) in certain embodiments is expressed as a Z score, a p-value, or t-test results. In some embodiments, a measured variance, confidence interval, sensitivity, specificity, and the like (e.g., collectively referred to as confidence parameters) for a result can be generated using one or more data processing manipulations described herein. Specific examples for generating results and associated confidence levels are described in the Examples section and in international patent application no. PCT/US12/59123 (W02013/052913) all content of which is incorporated herein by reference, including all text, tables, equations and drawings.

[422]The term "sensitivity" as used herein refers to the number of true positives divided by the number of true positives, plus the number of false negatives, where sensitivity (sens) can be within the range of 0 < sens < 1. The term "specificity" as used herein refers to the number of true negatives divided by the number of true negatives plus the number of false positives, where the sensitivity (spec) can be within the range of 0 < spec < 1. In some embodiments, a method that has sensitivity and specificity equal to one, or 100%, or close to one (e.g., between about 90% to about 99%), is sometimes selected. In some embodiments, a method that has a sensitivity of about 1, or 100% is selected, and in certain embodiments, a method that has a sensitivity of about 1 is selected (e.g., a sensitivity of about 90%, a sensitivity of about 91%, a sensitivity of about 92%, a sensitivity of about 93%, a sensitivity of about 94%, a sensitivity of about 95%, a sensitivity of about 96%, a sensitivity of about 97%, a sensitivity of about 98%, or a sensitivity of about 99%). In some embodiments, a method that has a specificity equal to 1, or 100% is selected, and in certain embodiments, a method that has a specificity close to 1 is selected (e.g., a specificity of about 90%, a specificity about 91%, a specificity of about 92%, a specificity of about 93%, a specificity of about 94%, a specificity of about 95%, a specificity of about 96%, a specificity of about about 97%, a specificity of about 98%, or a specificity of about 99%).

[423]Ideally, the number of false negatives equals zero or close to zero, so that no subject is mistakenly identified as not having at least one genetic variation when, in fact, they do have at least one genetic variation. On the other hand, an assessment is often made of the ability of a prediction algorithm to classify negatives correctly, a complementary measurement to sensitivity. Ideally, the number of false positives equals zero or close to zero, so that no subject is mistakenly identified as having at least one genetic variation when they do not have the genetic variation to be evaluated.

[424]In some embodiments, the presence or absence of a genetic variation (eg, chromosome aneuploidy) is determined for a fetus. In such modalities, the presence or absence of a fetal genetic variation (eg, fetal chromosome aneuploidy) is determined.

[425]In certain embodiments, the presence or absence of a genetic variation (eg, chromosome aneuploidy) is determined for a sample. In such embodiments, the presence or absence of a genetic variation in the sample's nucleic acid (eg, chromosome aneuploidy) is determined. In some embodiments, a detected variation or undetected residues in the nucleic acid sample from one source, but not in the nucleic acid sample from another source. Non-limiting examples of sources include placental nucleic acid, fetal nucleic acid, maternal nucleic acid, cancer cell nucleic acid, non-cancerous cell nucleic acid, the like, and combinations thereof. In non-limiting examples, a particular detected or undetected genetic variation (i) resides in placental nucleic acid, but not in fetal nucleic acid and not in maternal nucleic acid; (ii) resides in fetal nucleic acid but not maternal nucleic acid; or (iii) resides in maternal nucleic acid but not fetal nucleic acid.

[426]Once one or more results have been generated, a result is often used to provide a determination of the presence or absence of a genetic variation and/or associated medical condition. A result is typically provided to a healthcare professional (eg, laboratory technician or manager; physician or assistant). Often, a result is provided by a result module. In certain embodiments a result is provided by a graph module. In certain embodiments a result is provided on a peripheral or component of a machine or machine. For example, sometimes a result is provided by a printer or monitor. In some embodiments, a result determining the presence or absence of a genetic variation is provided to a healthcare professional in the form of a report, and in certain embodiments a report comprises a presentation of a result value and an associated confidence parameter. Generally, a result can be displayed in a suitable format that facilitates the determination of the presence or absence of a genetic variation and/or medical condition. Non-limiting examples of formats suitable for use for communicating and/or displaying sets of data or communicating a result include digital data, a graph, a 2D graph, a 3D graph, and 4D graph, an image, a graphic image , a graph, a bar graph, a pie graph, a diagram, a flowchart, a scatter plot, a map, a histogram, a density graph, a function graph, a circuit diagram, a block diagram , a bubble map, a constellation diagram, a contour diagram, a cartogram, spider chart, Venn diagram, nomogram, and the like, and a combination of the foregoing. Various examples of result representations are shown in the drawings and are described in the Examples.

[427]Generation of a result can be viewed as transforming fragments (reads) of nucleic acid sequences into a cellular nucleic acid representation of a subject, in certain embodiments. A cellular nucleic acid representation of a subject often reflects a dosage or copy number of a particular chromosome or portion thereof, and thus often the representation is a property of the subject's nucleic acid. The conversion of a multitude of relatively small sequence reads to a representation of a relatively large chromosome, for example, can be seen as a transformation. As an illustration, in a process for generating a representation of chromosome 21 that is about 47 million bases long, using fragments (reads) of approximately 36 base pairs in length, many thousands of fragments (reads) that are at least 100,000 times smaller than the chromosome are transformed into a significantly larger chromosome representation. Generating such a representation of a chromosome typically involves several manipulations of fragments (reads) (eg, mapping, filtering and/or normalization) to arrive at a relatively large representation of the chromosome, as described herein. Several manipulations are often used, which may require the use of one or more computers, often several computers coordinated in parallel.

[428] When providing a chromosome representation for a fetal chromosome using a sample from a pregnant woman, such a transformation is even more apparent, since the majority of reads are often maternal nucleic acid and a minority of fragments (reads) are often of fetal nucleic acid. Maternal nucleic acid reads often dominate fetal nucleic acid reads, and most maternal nucleic acid reads often mask a representation of a fetal chromosome. A typically large base of maternal reads can obscure differences between maternal and fetal chromosome nucleic acid, and obtaining a representation of a fetal chromosome against a base involves a process that deconvolves the contribution of maternal reads, as described here.

[429] In some embodiments, a result results from a transformation of fragments (reads) of sequences from a subject (for example, a pregnant woman), into a representation of an existing structure (for example, a genome, a chromosome or its segment) present in a subject (e.g., a mother and/or fetus). In some embodiments, an outcome comprises a transformation of sequence reads from a first subject (e.g., a pregnant woman), into a composite representation of structures (e.g., a genome, a chromosome, or a segment thereof), and a second composite representation transformation that produces a representation of a structure present in a first subject (eg, a pregnant woman) and/or a second subject (eg, a fetus). In some embodiments, a result comprises a transformation of fragments (reads) of sequences from a first subject (for example, a female subject, a pregnant woman), into a representation of structures (for example, a genome, a chromosome or its segment ) present in a second subject (e.g., a fetus).

[430] A transformation method here sometimes comprises determining the presence or absence of a trisomic chromosome (i.e., chromosome trisomy) in a fetus (e.g., T21, T18, and/or T13) from fragments (reads ) of nucleic acid in a sample obtained from a pregnant female subject carrying the fetus. In some embodiments, a transformation method here may comprise preparing (e.g., determining, visualizing, displaying, providing) a representation of a chromosome (e.g., chromosome copy number, chromosome dosage) for a fetus of fragments (reads) of nucleic acid in a sample obtained from a subject pregnant woman carrying the fetus. In the latter embodiments, a representation of a chromosome from a fetus is often for chromosome 13, chromosome 18 and/or chromosome 21.

Use of results

[431]A healthcare professional, or other qualified subject, who receives a report comprising one or more results determining the presence or absence of a genetic variation may use the data displayed in the report to make a call about the test subject's status or patient. The healthcare professional can make a recommendation based on the result provided, in some modalities. A healthcare professional or qualified subject can provide a test subject or patient with a link or score of the presence or absence of genetic variation based on the result value or values and associated confidence parameters provided in a report, in some embodiments. In certain embodiments, a score or link is made manually by a healthcare professional or qualified subject through visual observation of the report provided. In certain embodiments, a score or link is made by an automated routine, sometimes built into software, and reviewed by a healthcare professional or qualified subject for accuracy before providing information to a test subject or patient. The term "receiving a report" as used herein refers to obtaining, by means of communication, a written and/or graphical representation comprising a result, which upon review allows a healthcare professional or qualified subject to make a determination as to the presence or absence of a genetic variation in a test subject or patient. The report may be generated by a computer or human input, and may be communicated using electronic means (e.g., via the internet, a computer, fax, from one network location to another location at the same physical sites or different), or by another method of sending or receiving data (e.g. electronic mail services, courier service and the like). In some embodiments, the result is transmitted to a healthcare professional in an appropriate medium, including, without limitation, verbal, document, or archival form. The file may be, for example, but not limited to an audio file, a computer readable file, a paper file, a laboratory file or a medical record file.

[432]The term "providing a result" and its grammatical equivalents thereof, as used herein may also refer to a method for obtaining such information, including, without limitation, obtaining information from a laboratory (e.g., a laboratory file). A laboratory file may be generated by a laboratory that has performed one or more tests or one or more data processing steps to determine the presence or absence of the medical condition. The laboratory may be in the same location or a different location (for example, in another country) as employees identifying the presence or absence of the medical condition from the laboratory file. For example, the laboratory file can be generated in one location and transmitted to another location where the information in it will be transmitted to the pregnant woman subject. The laboratory file may be in tangible form or in electronic form (eg, computer-readable form), in certain embodiments.

[433]In some embodiments, a result may be provided to a healthcare professional, physician, or qualified subject of a laboratory, and the healthcare professional, physician, or qualified subject may make a diagnosis based on the result. In some embodiments, a result may be provided to a healthcare professional, physician or qualified subject of a laboratory and the healthcare professional, physician or qualified subject may make a diagnosis based, in part, on the results together with data and/or additional information and other results.

[434]The healthcare professional or qualified subject can provide an appropriate recommendation based on the result or results provided in the report. Non-limiting examples of recommendations that may be provided based on the report result provided include surgery, radiotherapy, chemotherapy, genetic counseling, post-birth treatment solutions (e.g., life planning, long-term assisted care, medications, symptomatic treatments ), termination of pregnancy, organ transplantation, blood transfusion, the like or combinations of the above. In some modalities, the recommendation is dependent on the result based on the classification provided (eg, Down syndrome, Turner syndrome, medical conditions associated with genetic variations at T13, medical conditions associated with genetic variations at T18).

[435]Laboratory staff (e.g., a laboratory manager) may analyze values (e.g., test counts, reference counts, level of deviation) underlying a determination of the presence or absence of a genetic variation (or determination euploid or non-euploid for a test region). For links regarding the presence or absence of a genetic variation that are close or questionable, laboratory personnel may reorder the same test, and/or request a different test (e.g., karyotype and/or amniocentesis in the case of fetal aneuploidy), which makes use of the same or different nucleic acid as a test subject's sample.

Genetic variations and medical conditions

[436]The presence or absence of a genetic variation can be determined using a method, apparatus, or machine described herein. In certain embodiments, the presence or absence of one or more genetic variations is determined according to a result provided by methods, machines and apparatus described herein. A genetic variation is usually a specific genetic phenotype present in certain individuals, and often a genetic variation is present in a statistically significant subpopulation of individuals. In some embodiments, a genetic variation is a chromosome abnormality (eg, aneuploidy), partial chromosome abnormality, or mosaicism, each of which is described in greater detail here. Non-limiting examples of genetic variations include one or more deletions (e.g., microdeletions), duplications (e.g., micro-duplications), insertions, mutations, polymorphisms (e.g., one nucleotide polymorphism), fusions, repeats ( eg, short tandem repeats), distinct methylation sites, distinct methylation patterns, similar, and combinations thereof. An insertion, repeat, deletion, duplication, mutation or polymorphism can be of any length, and in some embodiments, is from about 1 base or base pair (bp) to about 250 megabases (Mb) in length. In some embodiments, an insertion, repeat, deletion, duplication, mutation, or polymorphism is from about 1 base or base pair (bp) to about 1000 kilobases (kb) in length (e.g., about 10 bp, 50 bp, 100bp, 500bp, 1kb, 5kb, 1Okb, 50kb, 100kb, 500kb, or 1000kb in length).

[437]Genetic variation is sometimes a deletion. In certain embodiments a deletion is a mutation (eg, a genetic aberration) in which a part of a chromosome or DNA sequence is missing. A deletion is often the loss of genetic material. Any number of nucleotides can be deleted. A deletion may comprise the deletion of one or more entire chromosomes, a segment of a chromosome, an allele, a gene, an intron, an exon, any non-coding region, any coding region, its segment or a combination thereof. The deletion may comprise a microdeletion. A deletion may comprise the deletion of a single base.

[438]Genetic variation is sometimes genetic duplication. In certain embodiments a duplication is a mutation (eg, a genetic aberration) in which a part of a chromosome or DNA sequence is copied and inserted back into the genome. In certain embodiments a genetic duplication (i.e., duplication) is the duplication of a region of DNA. In some embodiments a duplication is a nucleic acid sequence that is repeated, often together, within a genome or chromosome. In some embodiments a duplication may comprise a copy of one or more entire chromosomes, a segment of a chromosome, an allele, a gene, an intron, an exon, any non-coding region, any coding region, its segment or a combination thereof. . A duplication may comprise a micro-duplication. A duplication sometimes comprises one or more copies of a duplicated nucleic acid. A duplication is sometimes characterized as a genetic region repeated one or more times (for example, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 times). Duplications can range from small regions (thousands of base pairs) to entire chromosomes in some cases. Duplications often occur as a result of an error in the homologous recombination process, or due to a retrotransposon event. Duplications have been associated with certain types of proliferative diseases. Duplications can be characterized using genomic DNA microarrays or comparative genetic hybridization (CGH).

[439]Genetic variation is sometimes an insertion. An insertion is sometimes the addition of one or more base pairs of nucleotides into a nucleic acid sequence. An insert is sometimes a micro-insert. In certain embodiments an insertion comprises adding a segment of a chromosome into a genome, the chromosome, or its segment. In certain embodiments an insertion comprises the addition of an allele, a gene, an intron, an exon, any non-coding region, any coding region, segment thereof or a combination thereof into a genome or segment thereof. In certain embodiments an insertion comprises adding (i.e., inserting) nucleic acid of unknown origin into a genome, chromosome, or segment thereof. In certain embodiments an insertion comprises the addition (i.e., insertion) of a single base.

[440]As used herein a "copy number variation" is generally a class or type of genetic variation or chromosomal aberration. A copy number variation can be a deletion (eg, micro-deletion), duplication (eg, a micro-duplication), or insertion (eg, a micro-insert). Often, the prefix "micro" as used herein is sometimes a nucleic acid segment less than 5 Mb in length. A copy number variation can include one or more deletions (eg, micro-deletion), duplications, and/or insertions (eg, a micro-duplication, micro-insertion) of a segment of a chromosome. In certain embodiments the duplication comprises an insertion. In certain embodiments an insertion is a duplication. In certain embodiments an insertion is not a duplication. For example, a duplication of a sequence in a chunk often increases the counts for a chunk in which the duplication is found. Often a doubling of a sequence in a portion raises the level. In certain embodiments, a duplication present in portions constituting a first tier increases the tier relative to a second tier where a duplication is absent. In certain embodiments an insert increases the scores of a slice and a sequence representing the insert is present (i.e., duplicated) elsewhere within the same slice. In certain embodiments an insertion does not significantly increase a chunk's scores or level and the sequence that is inserted is not a duplication of a sequence within the same chunk. In certain embodiments an insertion is not detected or represented as a duplication and a duplicated sequence representing the insertion is not present in the same portion.

[441]In some embodiments a copy number variation is a fetal copy number variation. Often, a fetal copy number variation is a copy number variation in the genome of a fetus. In some embodiments a copy number variation is a maternal and/or fetal copy number variation. In certain embodiments a maternal and/or fetal copy number variation is a copy number variation within the genome of a pregnant woman (e.g., a female subject bearing a fetus), a female subject who has given birth, or a woman able to have a fetus. A copy number variation can be a heterozygous copy number variation in which the variation (e.g., a duplication or deletion) is present in an allele of a genome. A copy number variation may be a homozygous copy number variation where the variation is present in both alleles of a genome. In some embodiments a copy number variation is a heterozygous or homozygous fetal copy number variation. In some embodiments the copy number variation is a heterozygous or homozygous maternal and/or fetal copy number variation. A copy number variation is sometimes present in a maternal genome and a fetal genome, a maternal genome and not a fetal genome, or a fetal genome and not a maternal genome.

[442] "Ploidy" is a reference to the number of chromosomes present in a fetus or mother. In certain embodiments "ploidy" is the same as "chromosome ploidy". In humans, for example, autosomal chromosomes are often present in pairs. For example, in the absence of genetic variation, most humans have two of each autosomal chromosome (eg, chromosomes 1-22). The presence of the normal complement of 2 autosomal chromosomes in a human is often referred to as euploid. "Microploidy" is similar in meaning to ploidy. "Microploidy" often refers to the ploidy of a segment of a chromosome. The term "microploidy" is sometimes a reference to the presence or absence of a copy number variation (e.g., a deletion, duplication, and/or an insertion) within a chromosome (e.g., a deletion, a duplication or a homozygous or heterozygous insertion, the like or absence thereof). "Ploidy" and "microploidy" are sometimes determined after normalizing one-level counts in a profile. Thus, a level that represents an autosomal pair of chromosomes (eg, an euploid) is often normalized to a ploidy of 1. Similarly, a level within a segment of a chromosome that represents the absence of a duplication, deletion, or insertion, is often normalized to a microploidy of 1. Ploidy and microploidy are often portion-specific (eg, portion-specific) and sample-specific. Ploidy is often defined as integer multiples of 1/2, with the values 1, 1/2, 0, 3/2, and 2 representing euploid (e.g., 2 chromosomes), 1 chromosome present (e.g., a deletion of chromosome), no chromosome present, 3 chromosomes (e.g. a trisomy) and 4 chromosomes respectively. Likewise, microploidy is generally defined as integer multiples of 1/2, with the values 1, ^, 0, 3/2, and 2 representing euploid (i.e., no copy number variation), a heterozygous deletion, homozygous deletion, heterozygous duplication and homozygous duplication, respectively. Some examples of ploidy values for a fetus are provided in Table 2.

[443]In certain embodiments, the microploidy of a fetus corresponds to the microploidy of the fetus's mother (ie, pregnant female subject). In certain embodiments the microploidy of a fetus matches the microploidy of the mother of the fetus and both mother and fetus carry the same heterozygous copy number variation, homozygous copy number variation, or both are euploid. In certain embodiments the microploidy of a fetus is different from the microploidy of the mother of the fetus. For example, sometimes the microploidy of a fetus is heterozygous for a copy number variation, the mother is homozygous for the copy number variation, and the microploidy of the fetus does not match (ie, is not equal to) the microploidy of the mother. for the specified copy number range.

[444]Microploidy is often associated with an expected level. For example, sometimes a level (for example, a level of a profile, sometimes a level that includes substantially any copy number variation) is normalized to a value of 1 (for example, a ploidy of 1, a microploidy of 1) and the microploidy of a homozygous duplication is 2, a heterozygous duplication is 1.5, a heterozygous deletion is 0.5, and a homozygous deletion is zero.

[445]A genetic variation for which the presence or absence is identified for a subject is associated with a medical condition in certain embodiments. Thus, the technology described herein can be used to identify the presence or absence of one or more genetic variations that are associated with a medical condition or medical condition. Non-limiting examples of medical conditions include those associated with mental retardation (e.g., Down Syndrome), aberrant cell proliferation (e.g., cancer), the presence of a nucleic acid in the microorganism (e.g., virus, bacteria, fungus, yeast), and pre-eclampsia.

[446]Non-limitative examples of genetic variations, conditions, and medical conditions are described below.

fetal gender

[447]In some embodiments, the prediction of a fetal gender or gender-related disorder (eg, sex chromosome aneuploidy) can be determined by a method, machine, or apparatus described herein. Gender determination is usually based on a sex chromosome. In humans, there are two sex chromosomes, the X and Y chromosomes. The Y chromosome contains a gene, SRY, which activates embryonic development as a male. The Y chromosomes of humans and other mammals also contain other genes necessary for normal sperm production. Individuals with XX are female and XY are male, and non-limiting variants, often referred to as sex chromosome aneuploidies, include XO, XYY, XXX, and XXY. In certain embodiments, males have two X chromosomes and one Y chromosome (XXY; Klinefelter syndrome), or one X chromosome and two Y chromosomes (XYY syndrome; Jacobs syndrome), and some females have three X chromosomes (XXX; Down syndrome). of Triple X), or a single X chromosome instead of two (XO; Turner Syndrome). In certain embodiments, only a portion of cells in an individual are affected by a sex chromosome aneuploidy that may be referred to as a mosaicism (eg, Turner mosaicism). Other cases include those where SRY is damaged (leading to an XY female), or copied to the X (leading to an XX male).

[448]In some embodiments, a method in which fetal gender is determined may also comprise determining fetal fraction and/or the presence or absence of a fetal genetic variation (eg, fetal chromosome aneuploidy). Determining the presence or absence of a fetal genetic variation can be performed in a suitable manner, non-limiting examples of which include karyotype analysis, amniocentesis, analysis of circulating cell-free nucleic acid, analysis of fetal cell-free DNA, analysis nucleotide sequence quantification, sequence read quantification, targeting approaches, amplification-based approaches, mass spectrometry-based approaches, differential methylation-based approaches, differential digestion-based approaches, polymorphism-based approaches, hybridization-based approaches (eg using probes), and the like.

[449]In certain cases, it may be beneficial to determine the gender of a fetus in utero. For example, a patient (eg, pregnant woman) with a family history of one or more sex-linked disorders may wish to determine the gender of the fetus she is carrying to help assess the risk of the fetus inheriting such a disorder. Sex-linked disorders include, without limitation, X-linked and Y-linked disorders. X-linked disorders include X-linked dominant and X-recessive disorders. Examples of X-linked recessive disorders include, without limitation, immunological disorders (e.g. , chronic granulomatous disease (CYBB), Wiskott-Aidrich syndrome, X-linked severe combined immunodeficiency, X-linked gammaglobulinemia, type 1 hyper-IgM syndrome, IPEX, X-linked lymphoproliferative disease, properdin deficiency), haematological disorders (e.g., hemophilia A, hemophilia B, X-linked sideroblastic anemia), endocrine disorders (e.g., androgen insensitivity syndrome/Kennedy's disease, Kallmann KAL1 syndrome, X-linked congenital adrenal hypoplasia), metabolic disorders (e.g., , ornithine transcarbamylase deficiency, oculocerebrorenal syndrome, adrenoleukodystrophy, glucose-6-phosphate dehydrogenase deficiency, pyruvate dehydrogenase deficiency, Danon disease/glycogen storage disease type IIb, Fabry disease, Hunter syndrome, Lesch-Nyhan syndrome , Menkes disease/occipital horn syndrome), nervous system disorders (eg, Coffin-Lowry syndrome, MASA syndrome, X-linked alpha thalassemia mental retardation syndrome, Siderium X-linked mental retardation syndrome , color blindness, ocular albinism, Norrie disease, choroideremia, Charcot-Marie-Tooth disease (CMTX2-3), Pelizaeus-Merzbacher disease, SMAX2), skin and tissue related disorders (e.g. dyskeratosis congenita, dysplasia hypohidrotic ectodermal dystrophy (EDA), X-linked ichthyosis, X-linked endothelial corneal dystrophy), neuromuscular disorders (e.g. Becker/Duchenne muscular dystrophy, centronuclear myopathy (MTM1), Conradi-Hunermann syndrome, Emery-Dreifuss muscular dystrophy 1), urologic disorders (eg, Alport's syndrome, Dent's disease, X-linked nephrogenic diabetes insipidus), bone/tooth disorders (eg, amelogenesis imperfecta AMELX), and other disorders (eg, Barth's syndrome, Mcleod syndrome, Smith-Fineman-Myers syndrome, Simpson-Golabi-Behmel syndrome, Mohr-Tranebjrerg syndrome, Naso-digito-acoustic syndrome). Examples of X-linked dominant disorders include, without limitation, X-linked hypophosphatemia, focal dermal hypoplasia, fragile X syndrome, Aicardi syndrome, Incontinence Pigmentosa, Rett syndrome, CHILD syndrome, Lujan-Fryns syndrome, and Orofaciodigital syndrome 1 Examples of Y-linked disorders include, without limitation, male infertility, retinitis pigmentosa, and azoospermia.

chromosomal abnormalities

[450]In some embodiments, the presence or absence of a fetal chromosome abnormality can be determined using a method, machine, or apparatus described herein. Chromosome abnormalities include, without limitation, a gain or loss of an entire chromosome or a region of a chromosome comprising one or more genes. Chromosome abnormalities include monosomies, trisomies, polysomies, loss of heterozygosity, translocations, deletions and/or duplications of one or more nucleotide sequences (eg, one or more genes), including duplications and deletions caused by unbalanced translocations. The term "chromosome abnormality", "aneuploidy" and/or "aneuploid", as used herein, refers to a deviation between the structure of the subject's chromosome and a normal homologous chromosome. The term "normal" refers to the predominant karyotype or banding pattern found in healthy individuals of a particular species, for example, an euploid genome (in humans, 46,XX or 46,XY). As different organisms have widely varying chromosome complements, the term "aneuploidy" and "aneuploidy" does not refer to a particular number of chromosomes, but rather to the situation where the chromosome content within a given cell or cells of an organism is not normal. In some embodiments, the term "aneuploidy" and "aneuploidy" herein refers to an imbalance of genetic material caused by a loss or gain of an entire chromosome, or part of a chromosome. An "aneuploidy" can refer to one or more deletions and/or insertions of a segment of a chromosome. The term "euploid", in some embodiments, refers to a normal complement of chromosomes.

[451]The term "monosomy" as used herein refers to the lack of a normal complement chromosome. Partial monosomy can occur in unbalanced translocations or deletions, in which only one segment of the chromosome is present in a single copy. Monosomy of sex chromosomes (45, X) causes Turner syndrome, for example. The term "disomy" refers to the presence of two copies of a chromosome. For organisms such as humans who have two copies of each chromosome (those who are diploid or "euploid"), disomy is the normal condition. For organisms that normally have three or more copies of each chromosome (those that are triploid or above), a disomy is an aneuploid state of the chromosome. In uniparental disomy, both copies of the chromosome come from the same parent (without any input from the other parent).

[452]The term "trisomy", as used herein, refers to the presence of three copies, rather than two copies, of a particular chromosome. The presence of an extra chromosome 21, which is found in human Down syndrome, is referred to as "Trisomy 21." Trisomy 18 and Trisomy 13 are two other human autosomal trisomies. Sex chromosome trisomy can be seen in females (eg, 47, XXX in Triple X syndrome) or males (eg, 47, XXY in Klinefelter syndrome, or 47, XYY in Jacobs syndrome). In some embodiments, a trisomy is a duplication of most or all of an autosome. In certain embodiments a trisomy is an aneuploidy of the entire chromosome resulting in three instances (e.g., three copies) of a particular type of chromosome (e.g., instead of two cases (e.g., one pair) of a particular type of chromosome). chromosome for an euploid).

[453]The terms "tetrasomy" and "pentasomy" as used herein refer to the presence of four or five copies of a chromosome, respectively. Although rarely seen with autosomes, sex chromosome tetrasomy and pentasomy have been reported in humans, including XXXX, XXXY, XXYY, XYYY, XXXXX, XXXXY, XXXYY, XXYYY, and XYYYY.

[454]Chromosome abnormalities can be caused by a variety of mechanisms. Mechanisms include, but are not limited to (i) nondisjunction occurring as the result of weakened mitotic checkpoints, (ii) inactive mitotic checkpoints causing nondisjunction in multiple chromosomes, (iii) merotelic linkage occurring when a kinetochore is turned on to both poles of the mitotic spindle, (iv) a multipolar spindle forming when more than two spindle poles form, (v) a monopolar spindle forming when only a single spindle pole forms, and (vi) a tetraploid intermediate occurring as an end result of the monopolar spindle mechanism.

[455]The terms "partial monosomy" and "partial trisomy" as used herein refer to an imbalance of genetic material caused by the loss or gain of part of a chromosome. A partial monosomy or partial trisomy can result from an unbalanced translocation, in which an individual carries a derived chromosome formed through the breakup and fusion of two different chromosomes. In this situation, the individual would have three copies of part of a chromosome (two normal copies and the segment that exists on the derived chromosome) and only one copy of part of the other chromosome involved in the derived chromosome.

[456] The term "mosaicism" as used herein refers to aneuploidy in some cells, but not all cells, of an organism. Certain chromosomal abnormalities can exist as mosaic and non-mosaic chromosome abnormalities. For example, certain trisomy 21 individuals have mosaic Down syndrome and some have non-mosaic Down syndrome. Different mechanisms can lead to mosaicism. For example, (i) an early zygote may have three chromosomes 21, which would normally result in simple trisomy 21, but during the course of cell division one or more cell lineages has lost one of its chromosomes 21; and (ii) an early zygote may have two chromosomes 21, but during the course of cell division one of the chromosomes 21 has been duplicated. Somatic mosaicism likely occurs through mechanisms other than those typically associated with genetic syndromes involving complete aneuploidy or mosaic. Somatic mosaicism has been identified in certain types of cancers and in neurons, for example. In certain cases, trisomy 12 has been identified in chronic lymphocytic leukemia (CLL) and trisomy 8 has been identified in acute myelogenous leukemia (AML). Furthermore, genetic syndromes in which an individual is predisposed to chromosome breakage (chromosome instability syndromes) are often associated with increased risk for various types of cancer, thereby highlighting the role of somatic aneuploidy in carcinogenesis. Methods and protocols described here can identify the presence or absence of mosaic and non-mosaic chromosome abnormalities.

[457]Tables 1A and 1B present a non-limiting list of conditions, syndromes, and/or chromosome abnormalities that can potentially be identified by the methods, machines, and apparatus described herein. Table 1B is from the DECIPHER database as of 6 October 2011 (eg version 5.1, based on positions mapped to GRCh37; available at uniform resource locator (URL) dechipher.sanger.ac.uk). Table 1A Table 1B

[458]Grade 1 conditions often have one or more of the following features: pathogenic abnormality; strong agreement among geneticists; highly penetrating; may still have variable phenotype but some common features; all cases in the literature have a clinical phenotype; no cases of healthy individuals with the anomaly; failed to report on DVG databases or found in healthy population; functional data confirming dose effect on one gene or multi-genes; strong or confirmed candidate genes; implications for defined clinical management; the known cancer risk with implication for surveillance; multiple information sources (OMIM, GeneReviews, Orphanet, Unique, Wikipedia); and/or available for diagnostic use (reproductive counseling).

[459]Grade 2 conditions often have one or more of the following characteristics; probably pathogenic anomaly; highly penetrating; variable phenotype with no consistent features other than DD; small number of cases/reports in the literature; all reported cases have a clinical phenotype; no functional data or confirmed pathogenic genes; various sources of information (OMIM, GeneReviews, Orphanet, Unique, Wikipedia); and/or may be used for diagnostic purposes and reproductive counseling.

[460]Grade 3 conditions often have one or more of the following characteristics; local susceptibility; healthy subjects or unaffected parents of a described proband; present in control populations; non-penetrating; mild and non-specific phenotype; are less consistent; there are no confirmed functional data or pathogenic genes; more limited data sources; the possibility of a second diagnosis remains a possibility for cases that deviate from the majority or if a new clinical finding is present; and/or caution when using for diagnostic purposes and advice guarded for reproductive counseling.

Pre eclampsia

[461]In some embodiments, the presence or absence of preeclampsia is determined using a method, machine, or apparatus described herein. Pre-eclampsia is a condition in which hypertension arises in pregnancy (ie, pregnancy-induced hypertension) and is associated with significant amounts of protein in the urine. In certain embodiments, pre-eclampsia is also associated with elevated levels of extracellular nucleic acid and/or changes in methylation patterns. For example, a positive correlation between extracellular levels of fetal-derived hypermethylated RASSF1A and the severity of preeclampsia has been observed. In certain examples, increased DNA methylation is observed for the H19 gene in preeclamptic placentas compared to normal controls.

[462] Pre-eclampsia is a leading cause of neonatal/maternal and fetal mortality and morbidity worldwide. Circulating cell-free nucleic acids in plasma and serum are new biomarkers with promising clinical applications in different medical areas, including prenatal diagnosis. Quantitative changes of cell-free (cff)DNA in maternal plasma as an indicator for impending pre-eclampsia have been reported in different studies, eg using quantitative real-time PCR for male-specific SRY or local DYS 14 . In cases of early onset preeclampsia, elevated levels may be seen in the first trimester. Increased levels of cffDNA before the onset of symptoms may be due to hypoxia/reoxygenation within the intervillous space leading to tissue oxidative stress and increased placental apoptosis and necrosis. In addition to evidence of increased excretion of cffDNA in the maternal circulation, there is also evidence of reduced renal clearance of cffDNA in pre-eclampsia. As the amount of fetal DNA is currently determined by quantification of specific Y-chromosome sequences, alternative approaches, such as measurement of whole cell-free DNA or the use of gender-independent fetal epigenetic markers, such as DNA methylation, offer an alternative. Cell-free RNA of placental origin is another alternative biomarker that can be used for screening and diagnosing pre-eclampsia in clinical practice. Fetal RNA is associated with subcellular placental particles that protect it from degradation. Fetal RNA levels are sometimes ten times higher in pregnant women with preeclampsia compared to controls, and therefore it is an alternative biomarker that can be used for screening and diagnosing preeclampsia in clinical practice. .

pathogens

[463]In some embodiments, the presence or absence of a pathogenic condition is determined using a method or apparatus described herein. A pathogenic condition can be caused by an infection of a host by a pathogenic agent, including, but not limited to, a bacterium, virus, or fungus. Since pathogens typically have nucleic acid (eg, genomic DNA, genomic RNA, mRNA) that can be distinguished from host nucleic acid, methods, apparatus, and machines provided herein can be used to determine the presence or absence of a pathogen. Often, pathogens possess nucleic acid with characteristics unique to a particular pathogen, such as, for example, epigenetic status and/or one or more sequence variations, duplications and/or deletions. As such, the methods provided here can be used to identify a particular pathogen or pathogen variant (eg, strain).

Cancers

[464]In some embodiments, the presence or absence of a cell proliferation disorder (eg, a cancer) is determined using a method, machine, or apparatus described herein. For example, serum cell-free nucleic acid levels may be elevated in patients with various types of cancer compared to healthy patients. Patients with metastatic disease, for example, can sometimes have serum DNA levels approximately twice as high as non-metastatic patients. Patients with metastatic disease can also be identified using cancer-specific markers and/or certain single nucleotide polymorphisms or short tandem repeats, for example. Non-limiting examples of cancer types that may be positively correlated with elevated levels of circulating DNA include breast cancer, colorectal cancer, gastrointestinal cancer, hepatocellular cancer, lung cancer, melanoma, non-Hodgkin's lymphoma, leukemia, multiple myeloma, bladder cancer, hepatoma, cervical cancer, esophageal cancer, pancreatic cancer and prostate cancer. Various cancers may possess, and sometimes may release into the bloodstream, nucleic acids with characteristics that distinguish them from the nucleic acids of healthy, non-cancerous cells, such as, for example, epigenetic status and/or variations , sequence duplications and/or deletions. Such features may, for example, be specific to a particular type of cancer. Thus, it is further contemplated that the method provided herein can be used to identify a particular type of cancer.

[465]The software may be used to perform one or more steps of the processes described herein, including but not limited to; counting, processing data, generating a result, and/or providing one or more recommendations based on generated results, as described in more detail below.

Machines, Software and Interfaces

[466]Certain processes and methods described herein (e.g., quantification, mapping, normalization, sizing, tuning, categorization, counting and/or determination of sequence reads, counts, levels (e.g., levels) and/or profiles) often cannot be performed without a computer, microprocessor, software, module, or other machine. Methods described herein are typically computer-implemented methods, and one or more portions of a method are sometimes performed by one or more processors (e.g., microprocessors), computers, or microprocessor controlled machines. Embodiments relating to methods described herein are generally applicable to the same or related processes implemented by instructions in systems, machines, and computer program products described herein. Embodiments related to the methods described in this document may be generally applicable to the same or related processes implemented by a non-transient computer-readable storage medium, with an executable program stored thereon, where the program instructs a microprocessor to execute the method, or a part of it. In some embodiments, processes and methods described herein (e.g., quantitation, counting, and/or sequence reads, counts, levels, and/or profiling) are performed by automated methods. In some embodiments one or more steps and a method described herein are performed by a microprocessor and/or computer, and/or performed in conjunction with memory. In some embodiments, an automated method is incorporated into software, modules, microprocessors, peripherals, and/or a similar understanding machine that determines sequence reads, counts, mapping, mapped sequence markers, levels, profiles, normalizations, comparisons, range definition, categorization, adjustments, plotting, results, transformations and identifications. As used herein, software refers to computer-readable program instructions, which, when executed by a microprocessor, perform computer operations as described herein.

[467]Sequence reads, counts, levels, and profiles obtained from a test subject (e.g., a patient, a pregnant woman) and/or a reference subject can be further analyzed and processed to determine the presence or absence of a genetic variation. Fragments (reads) of sequences, counts, levels and profiles are sometimes referred to as "data" or "datasets". In some embodiments, the data or datasets may be characterized by one or more characteristics or variables (e.g., sequence-based [e.g., GC content, specific nucleotide sequence, the like], specific function [e.g., expressed genes, cancer genes, the like], based on location [specific genome, specific chromosome, or portion or portion-specific], the like, and combinations thereof). In certain embodiments, data or datasets can be organized into a matrix that has two or more dimensions based on one or more characteristics or variables. Data organized into matrices can be organized using any suitable resources or variables. A non-limiting example of data in a matrix that includes data is organized by maternal age, maternal ploidy, and fetal contribution. In certain embodiments, data sets characterized by one or more characteristics or variables are sometimes processed after counting.

[468]Machines, software, and interfaces can be used to perform the methods described here. Using machines, software, and interfaces, a user can enter, request, query, or determine options for using particular information, programs, or processes (e.g., mapping sequence reads, processing mapped data, and/or providing of a result), which may involve the execution of statistical analysis algorithms, statistical significance algorithms, statistical algorithms, interactive steps, validation algorithms and graphical representations, for example. In some embodiments, a set of data may be entered by a user as input information, a user may download one or more sets of data via suitable hardware means (e.g., flash drive), and/or a user may submit a set of data from one system to another for subsequent processing and/or providing a result (e.g., sending readable sequence data from a sequencer to a computer system for sequence fragment mapping, sending mapped sequence data to a computer system for processing and providing a result and/or report).

[469]A system typically comprises one or more machines. Each machine comprises one or more memory, one or more microprocessors, and instructions. In case a system includes two or more machines, some or all the machines can be located in the same place, some or all the machines can be located in different places, all the machines can be located in one place and/or all the machines may be located in different places. Where a system includes two or more machines, some or all of the machines may be located at the same location as a user, some or all of the machines may be located at a different location than a user, all of the machines may be located at the same location that the user, and/or the entire machine may be located in one or more different locations than the user.

[470] A system sometimes comprises a computing machine and a sequencing apparatus or machine, wherein the sequencing apparatus or machine is configured to receive physical nucleic acid and generate sequence reads, and the apparatus Compute is configured to process the fragments (reads) from the sequencing apparatus or machine. The computing machine is sometimes configured to determine the presence or absence of genetic variation (eg, copy number variation; fetal chromosome aneuploidy) from sequence reads.

[471]A user may, for example, consult software which then may acquire a dataset via internet access, and in certain embodiments, a programmable microprocessor may be required to acquire a suitable dataset based on given parameters. A programmable microprocessor may also prompt a user to select one or more data options selected by the microprocessor based on given parameters. A programmable microprocessor may prompt a user to select one or more data options selected by the microprocessor based on information found over the internet, other internal or external information, or the like. Options can be chosen by selecting one or more data characteristic selections, one or more statistical algorithms, one or more statistical analysis algorithms, one or more statistical significance algorithms, interactive steps, one or more validation algorithms, and one or more plus graphical representations of methods, machines, apparatus, computer programs, or a non-transient computer-readable storage medium with an executable program stored thereon.

[472]Systems discussed here may include general components of computer systems, such as, for example, network servers, laptop systems, desktop systems, handheld systems, personal digital assistants, computing kiosks, and the like. A computer system may comprise one or more input means, such as a keyboard, touch screen, mouse, speech recognition, or other means to allow the user to enter data into the system. A system may further comprise one or more outputs including, but not limited to, a display screen (e.g. CRT or LCD), speaker, fax machine, printer (e.g. laser printer, inkjet , impact, black and white or color), or other useful output to provide visual, auditory and/or printed information output (e.g. result and/or report).

[473]In a system, input and output means may be connected to a central processing unit, which may comprise, among other components, a microprocessor for executing program instructions and memory for storing program code and data. In some embodiments, the processes may be implemented as a one-user system located in a geographic location. In certain embodiments, the processes may be implemented as a multi-user system. In the case of a multiuser implementation, several central processing units can be connected through a network. The network can be local, spanning a single department in a portion of a building, an entire building, spanning multiple buildings, spanning a region, spanning an entire country, or worldwide. The network can be private, being owned and controlled by a vendor, or it can be implemented as an internet-based service where the user accesses a web page to log in and retrieve information. Thus, in certain embodiments, a system includes one or more machines, which may be local or remote from a user. More than one machine at a site or multiple sites can be accessed by a user, and data can be mapped and/or processed serially and/or in parallel. In this way, proper setup and control can be used for data mapping and/or processing using multiple machines, such as on local area network, wide area network and/or "cloud" computing platforms.

[474]A system may include a communications interface in some embodiments. A communication interface allows the transfer of software and data between a computer system and one or more external devices. Non-limiting examples of communications interfaces include a modem, a network interface (such as an Ethernet card), a communications port, a PCMCIA card, and the like. Software and data transferred over a communications interface are generally in the form of signals, which may be electronic, electromagnetic, optical and/or other signals that can be received by a communications interface. Signals are often provided to a communications interface over a channel. A frequency channel carries signals and can be implemented using wire or cable, fiber optics, a telephone line, a cell phone link, an RF link, and/or other communication channels. Thus, in one example, a communications interface can be used to receive signal information that can be detected by a signal detection module.

[475]Data may be input by a suitable device and/or method, including, but not limited to, manual input devices or direct data input devices (DDEs). Non-limiting examples of handheld devices include keyboards, concept keyboards, touchscreens, light pens, mice, tracker balls, joysticks, graphics tablets, scanners, digital cameras, video digitizers and speech recognition devices. Non-limiting examples of DDes include bar code readers, magnetic stripe codes, smart cards, magnetic ink character recognition, optical character recognition, optical mark recognition, and overturn documents.

[476]In some embodiments, the output of a sequencing apparatus or machine can serve as data that can be input through an input device. In certain embodiments, the mapped sequence reads may serve as inputable data through an input device. In certain embodiments, the size of the nucleic acid fragment (e.g., length) can serve as data that can be input through an input device. In certain embodiments, the output of a nucleic acid capture process (e.g., genomic region source data) can serve as data that can be input through an input device. In certain embodiments, a combination of nucleic acid fragment size (e.g., length) and output from a nucleic acid capture process (e.g., genomic region origin data) can serve as data that can be introduced via an input device. In certain embodiments, the simulated data is generated by a silicon process and the simulated data serves as inputtable data through an input device. The term "in silico" refers to research and experiments performed using a computer. Silicon processes include, but are not limited to, mapping sequence reads and processing mapped sequence reads according to the processes described herein.

[477] A system may include software useful for performing a process described herein, and software may include one or more modules for performing such processes (e.g., sequencing module, logic processing module, display organization module of data). The term "software" refers to computer-readable program instructions that, when executed by a computer, perform computer operations. Instructions executable by one or more microprocessors are sometimes provided as executable code that, when executed, can cause one or more microprocessors to implement a method described herein. A module described herein may exist as software, and instructions (eg, processes, routines, subroutines) embodied in the software may be implemented or executed by a microprocessor. For example, a module (for example, a software module) can be a part of a program that performs a particular process or task. The term "module" refers to a self-contained functional unit that can be used in a larger machine or software system. A module may comprise a set of instructions for performing a function of the module. A module can transform data and/or information. The data and/or information may be in a suitable form. For example, data and/or information can be digital or analogue. In certain embodiments, data and/or information can sometimes be packets, bytes, characters or bits. In some embodiments, data and/or information may be any data or information that is gathered, joined or usable. Non-limiting examples of data and/or information include a suitable medium, images, video, sound (e.g. frequencies, audible or non-audible), numbers, constants, a value, objects, time, functions, instructions, maps, references , sequences, fragments (reads), mapped fragments (reads), levels, tracks, boundaries, signs, exposures, representations or transformations thereof. A module can accept or receive data and/or information, transform the data and/or information into a second form, and provide or transfer the second form to a machine, peripheral, component, or other module. A module may perform one or more of the following non-limiting functions: mapping fragments (reads) of mapping sequences, providing counts, assembling portions, providing or determining a level, providing a count profile, normalizing (for example, normalizing reads, normalizing counts, and the like), providing a profile of counts or levels of normalized counts, comparing two or more levels, providing uncertainty values, providing or determining expected levels and ranges expected ranges (e.g., expected level ranges, threshold ranges, and threshold levels), providing level adjustments (e.g., first level adjustment, second level adjustment, adjustment of a profile of a chromosome or its segment, and/or padding), providing identification (eg, identifying a copy number variation, genetic variation, or aneuploidy), categorization, plotting, and/or determining a result, for example. A microprocessor may, in certain embodiments, carry instructions in a module. In some embodiments, one or more microprocessors are required to carry out instructions in a module or group of modules. A module can provide data and/or information to another module, machine or source and can receive data and/or information from another module, machine or source.

[478] A computer program product is sometimes embodied in a tangible computer-readable medium, and sometimes it is tangible embodied in a non-transient computer-readable medium. A module is sometimes stored on a computer-readable medium (eg, disk, hard) or in memory (eg, random-access memory). A module and microprocessor capable of executing a module's instructions may be located in one machine or another machine. A module and/or the microprocessor that is capable of implementing an instruction for a module may be located in the same location as a user (e.g. local area network) or in a different location than a user (e.g. remote network, computer system). cloud). In embodiments where a method is performed in conjunction with two or more modules, the modules may be located on the same machine, one or more modules may be located on a different machine at the same physical location, and one or more modules may be located on different machines in different physical locations.

[479]A machine, in some embodiments, comprises at least one microprocessor for carrying out the instructions of a module. Fragment counts (reads) of sequences mapped to portions of a reference genome are sometimes accessed by a microprocessor that executes configured instructions to perform a method described here. The counts that are accessed by a microprocessor can be within system memory, and the counts can be accessed and placed in system memory after they have been obtained. In some embodiments, a machine includes a microprocessor (e.g., one or more microprocessors) wherein the microprocessor can execute and/or implement one or more instructions (e.g., processes, routines and/or subroutines) of a module. In some embodiments, a machine includes multiple microprocessors, such as microprocessors coordinated and working in parallel. In some embodiments, a machine operates with one or more external microprocessors (e.g., an internal or external network, server, storage system, and/or storage network (e.g., a cloud)). In some embodiments, a machine comprises a module. In certain embodiments a machine comprises one or more modules. A machine comprising a frequency module can receive and transfer one or more of data and/or information to and from other modules. In certain embodiments, a machine comprises peripherals and/or components. In certain embodiments a machine may comprise one or more peripherals or components that can transfer data and/or information to and from other modules, peripherals and/or components. In certain embodiments a machine interacts with a peripheral and/or component that provides the data and/or information. In certain embodiments peripherals and components assist a machine in performing a function or interact directly with a module. Non-limiting examples of peripherals and/or components include a suitable computer peripheral, I/O or storage method or device, including but not limited to scanners, printers, monitors (e.g. monitors, LED, LCT or CRTs) , cameras, microphones, pads (e.g. ipads, tablets), touch screens, smart phones, cell phones, USB I/O devices, USB mass storage devices, keyboards, a computer mouse, digital pens, modems , hard disks, jump drives, flash drives, a microprocessor, a server, COS, DVDs, graphics cards, specialized I/O devices (e.g. sequencers, photocells, photo multiplier tubes, optical readers, sensors, etc.), one or more flow cells, fluid handling components, network interface controllers, ROM, RAM, wireless transfer methods and devices (Bluetooth, Wi-Fi, and the like,), the world wide web (www), the internet, a computer and/or another module.

[480]Software is often provided in a program product containing program instructions recorded on a computer-readable medium, including, but not limited to, magnetic media, including floppy disks, hard disks, and magnetic tape; and optical media including CD-ROM disks, DVD disks, magneto-optical disks, flash drives, RAM, floppy disks, the like, and other such media on which program instructions may be recorded. In an online application, a server and web site maintained by an organization can be configured to provide software downloads to remote users, or remote users can access a remote system maintained by an organization to remotely access software. Software can get or receive input information. Software may include a module that specifically obtains or receives data (e.g., a data-receiving module that receives readable sequence data and/or readable mapped data) and may include a module that specifically processes the data (e.g., a module process that processes incoming data (e.g., filters, normalizes, provides a result, and/or report). The terms "getting" and "receiving" input information refers to receiving data (e.g., fragments ( reads) of sequences, mapped fragments (reads) by computer communication means from a local or remote site, human input, or any other method of receiving data. Input information can be generated at the same location as is received, or it may be generated at a different location and transmitted to the receiving location. In some embodiments, the input information is modified before it is processed (e.g., put into a processable format (e.g., tabulated) ). In some embodiments, computer program products are provided, such as, for example, a computer program product comprising a computer usable medium having computer readable program code incorporated therein, the computer readable program code adapted to be performed to implement a method comprising: (a) obtaining fragments (reads) of sample nucleic acid sequences from a test subject sample; (b) mapping the fragments (reads) of sequences obtained in (a) to a known genome, whose known genome was divided into portions; (c) count of fragments (reads) of sequences mapped within the portions; (d) generating a normalized sample count profile by normalizing the counts to the portions obtained in (c); and (e) determining the presence or absence of a genetic variation of the normalized count profile in (d).

[481]The software may include one or more algorithms in certain embodiments. An algorithm can be used for processing data and/or providing a result or report according to a finite sequence of instructions. An algorithm is often a list of defined instructions for completing a task. Starting from an initial state, the instructions may describe a computation that proceeds through a defined series of successive states, eventually ending in a definite final state. The transition from one state to another is not necessarily deterministic (for example, some algorithms incorporate randomness). By way of example, and without limitation, an algorithm may be a search algorithm, sorting algorithm, sorting algorithm, numerical algorithm, graphing algorithm, string algorithm, modeling algorithm, computational geometric algorithm, combinatorial algorithm, learning algorithm machine code, encryption algorithm, data compression algorithm, parser algorithm, and the like. An algorithm can include one algorithm or two or more algorithms working in combination. An algorithm can be of any suitable complexity and/or parameterized complexity. An algorithm may be used for computation and/or data processing, and in some embodiments, may be used in a deterministic or probabilistic/predictive approach. An algorithm may be implemented in a computing environment by use of a suitable programming language, non-limiting examples of which are C, C++, Java, Peri, Python, Fortran and the like. In some embodiments, an algorithm can be configured or modified to include margin of error, statistical analysis, statistical significance, and/or comparison to other information or datasets (e.g., applicable when using a neural network algorithm or of clustering).

[482]In certain embodiments, various algorithms may be implemented for use in software. These algorithms can be formed with the raw data in some embodiments. For each new sample of raw data, the trained algorithms can produce a representative set of processed data or results. A processed dataset is sometimes of reduced complexity compared to the parent dataset that was processed. Based on a processed set, the performance of a trained algorithm can be evaluated based on sensitivity and specificity, in some modalities. An algorithm with the greatest sensitivity and/or specificity can be identified and used, in certain embodiments.

[483]In certain embodiments, simulated data (simulation) can aid data processing, for example, training an algorithm or testing an algorithm. In some embodiments, the simulated data includes several hypothetical samplings of different groups of fragments (reads) of sequences. Simulated data may be based on what would be expected from a real population or may be biased to test an algorithm and/or to assign a correct classification. Simulated data is also referred to here as "virtual" data. Simulations can be performed by a computer program, in certain ways. One possible step for using a simulated dataset is to assess the confidence of identified results, for example, how well a random sampling matches or better represents the original data. One approach is to calculate a probability value (p value), which estimates the probability of a random sample scoring better than selected samples. In some embodiments, an empirical model may be evaluated, where at least one sample is assumed to correspond to a reference sample (with or without resolved variances). In some embodiments, another distribution, such as a Poisson distribution, for example, can be used to define the probability distribution.

[484]A system may include one or more microprocessors in certain embodiments. A microprocessor can be connected to a communication bus. A computer system can include main memory, often random access memory (RAM), and it can also include secondary memory. Memory in some embodiments comprises a non-transient computer readable storage medium. Secondary memory may include, for example, a hard disk drive and/or a removable storage unit, representation of a floppy disk drive, magnetic tape drive, optical disk drive, memory card, and the like. A removable storage drive often reads and/or writes to a removable storage drive. Non-limiting examples of removable storage units include a floppy disk, magnetic tape, optical disk, and the like, which can be read by and written to, for example, a removable storage unit. A removable storage unit may include a computer usable storage medium having computer software and/or data stored thereon.

[485]A microprocessor can run software on a system. In some embodiments, a microprocessor can be programmed to automatically perform a task described herein that a user can perform. Accordingly, a microprocessor, or algorithm driven by such a microprocessor, may require little or no supervision or input from a user (for example, software may be programmed to perform a function automatically). In some embodiments, the complexity of a process is so great that a single person or group of people cannot perform the process in a short enough period of time for the determination of the presence or absence of a genetic variation.

[486]In some embodiments, secondary memory may include other similar means for allowing computer programs or other instructions to be loaded into a computer system. For example, a system might include a removable storage unit and an interface device. Non-limiting examples of such systems include a program cartridge and cartridge interface (such as those found in video game devices), a removable memory chip (such as EPROM or PROM) and associated socket, and other storage units. removable storage devices and interfaces that allow software and data to be transferred from the removable storage unit to a computer system.

[487]An entity may generate counts of reads of sequences, map the reads of sequences into chunks, count the mapped fragments (reads), and use the counted mapped fragments (reads) in a method, system, machine, apparatus, or computer program product described herein, in some embodiments. Fragment counts (reads) of sequences mapped to portions are sometimes transferred by one entity to a second entity for use by the second entity of a method, system, apparatus, or a computer program product described herein, in certain embodiments .

[488]In some embodiments, one entity generates sequence reads and a second entity maps those sequence reads into chunks into a reference genome in some embodiments. The second entity sometimes counts mapped reads and uses the counted mapped reads in a method, system, machine, or computer program product described herein. In certain embodiments the second entity transfers the mapped fragments (reads) in a third entity, and the third entity counts the mapped fragments (reads) and uses the mapped fragments (reads) in a method, system, equipment or computer program product described here. In certain embodiments, the second entity counts the mapped fragments (reads) and transfers the counted mapped fragments (reads) to a third entity, and the third entity uses the counted mapped fragments (reads) in a method, system, equipment or program product computer described here. In modalities involving a third entity, the third entity is sometimes the same as the first entity. That is, the first entity sometimes transfers fragments (reads) of sequences to a second entity, where the second entity can map the fragments (reads) of sequences into portions in a reference genome and/or count fragments (reads) ) mapped, and the second entity can transfer the mapped and/or counted fragments (reads) to a third entity. A third entity, sometimes, can use the fragments (reads) mapped and/or counted in a method, system, equipment or computer program product described herein, in which the third entity, sometimes, is the same as the first entity, and sometimes the third entity is different from the first or second entity.

[489]In some embodiments, one entity obtains blood from a pregnant woman, optionally isolates nucleic acid from the blood (e.g., from plasma or serum), and transfers the blood or nucleic acid to a second entity that generates fragments ( reads) of nucleic acid sequences.

[490] Figure 24 illustrates a non-limiting example of a computing environment 510 in which various systems, methods, algorithms, and data structures described herein may be implemented. Computing environment 510 is only one example of a suitable computing environment and is not intended to suggest any limitation on the scope of use or functionality of the systems, methods, and data structures described herein. Neither computing environment 510 should be construed as having any dependency or requirement on any one or combination of components illustrated in computing environment 510. A subset of systems, methods, and data structures shown in FIG. 24 may be utilized in certain embodiments. . Systems, methods, and data structures described herein are operative with numerous other special-purpose or general-purpose computing system environments or configurations. Examples of known computing system environments or configurations that may be suitable include, but are not limited to, personal computers, server computers, thin clients, thick clients, laptop or handheld devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable electronics, networked PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

[491] The operating environment 510 of Figure 24 includes a general purpose computing device in the form of a computer 520, which includes a processing unit 521, a system memory 522, and a system bus 523 operationally couple various components from system memory 522 to processing unit 521. There may be only one or there may be more than one processing unit 521, such that computer microprocessor 520 includes a central processing unit (CPU), or a plurality of processing units, commonly referred to as a parallel processing environment. Computer 520 can be a conventional computer, a distributed computer, or any other type of computer.

[492]The system bus 523 can be any of several types of bus structures including a memory bus or a memory bus, a peripheral bus, and a local bus using any of a variety of bus architectures. System memory may also be referred to simply as memory, and includes read only memory (ROM) 524 and random access memory (RAM). A basic input/output system (BIOS) 526, containing the basic routines that help transfer information between elements within computer 520, such as during boot, is stored in ROM 524. Computer 520 may further include an interface for hard disk drive 527 for reading from and writing to a hard disk, not shown, a magnetic disk drive 528 for reading from or writing to a removable magnetic disk 529, and an optical disk drive 530 for reading from or writing to a removable optical disk 531, such as a CD-ROM or other optical medium.

[493] The hard disk drive 527, magnetic disk drive 528, and optical disk drive 530 are connected to the system bus 523 by a hard disk drive interface 532, a magnetic disk drive interface 533, and a optical disk drive interface 534, respectively. The drives and their associated computer-readable media provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for computer 520. Any type of computer-readable media that can store accessable data by a computer, such as magnetic cassettes, flash memory cards, digital video discs, Bernoulli cartridges, random access memories (RAM), read-only memories (ROMs), and the like, can be used in the operating environment.

[494] A number of program modules may be stored on the hard disk, magnetic disk 529, optical disk 531, ROM 524, or RAM, including an operating system 535, one or more application programs 536, other program modules 537, and program data 538. A user may input commands and information to personal computer 520 through input devices, such as a keyboard 540 and pointing device 542. Other input devices (not shown) may include a microphone, joystick, game pad, satellite dish, scanner or the like. These and other input devices are often connected to processing unit 521 through a serial port interface 546 which is coupled to the system bus, but may be connected through other interfaces such as a parallel port, game port, or a universal serial bus (USB). A monitor 547 or other type of display device is also connected to the system bus of 523 via an interface, such as a video adapter 548. In addition to the monitor, computers typically include other peripheral output devices (not shown), such as speakers and printers.

[495] Computer 520 may operate in a network environment using logical connections to one or more remote computers, such as remote computer 549. These logical connections may be achieved by a communication device coupled to or a part of computer 520, or in other ways. Remote computer 549 may be another computer, server, router, network PC, client, peer device, or other common network node, and typically includes many or all of the elements described above in relation to computer 520, although only one memory storage device 550 has been illustrated in Figure 24. The logical connections shown in Figure 24 include a local area network (LAN) 551 and a wide area network (WAN) 552. Such network environments are common on office networks, enterprise-wide computer networks, intranets, and the Internet, which are all types of networks.

[496] When used in a LAN environment, the computer 520 is connected to the local network 551 through a network interface or adapter 553, which is a type of communications device. When used in a WAN network environment, computer 520 often includes a modem 554, a type of communications device, or any other type of communication device for establishing communication over wide area network 552. Modem 554, which may be internal or external, is connected to system bus 523 through serial port interface 546. In a network environment, program modules presented in connection with personal computer 520, or portions thereof, may be stored on the storage device remote memory. It is noted that the network connections shown are non-limiting examples and other communications devices to establish a communications link between computers may be used.

Modules

[497]One or more modules may be used in a method described herein, non-limiting examples of which include a logic processing module, data display organization module, sequencing module, mapping module, counting module, module filtering module, weighting module, normalization module, GC trending module, level module, comparison module, range adjustment module, categorization module, adjustment module, plotting module, representation module, relationship module , result module and/or data display organization module, the like or a combination thereof. Modules are sometimes controlled by a microprocessor. In certain embodiments a module or a machine comprising one or more modules, collect, assemble, receive, obtain, access, retrieve, provide and/or transfer data and/or information to or from another module, machine, component, peripheral or operator of a machine. In some embodiments, data and/or information (e.g., sequencing reads) is provided to a module of a machine comprising one or more of the following: one or more flow cells, a chamber, a detector (e.g., e.g. a photodetector, a photocell, an electrical detector (e.g. an amplitude modulation detector, a phase and frequency modulation detector, a phase locked loop detector), a counter, a sensor (e.g. , a pressure, temperature, volume, flow, weight sensor), a fluid handling device, a printer, a monitor (eg, an LED, LCT, or CRT), the like, or combinations thereof. For example, by Sometimes a machine operator supplies a constant, limit value, formula, or predetermined value to a module. A module is often configured to transfer data and/or information to or from another module or machine. A module can receiving data and/or information from another module, non-limiting examples of which include a logic processing module, sequencing module, mapping module, counting module, filtering module, weighting module, normalization module, trending module CG module, level module, comparison module, range adjustment module, categorization module, plotting module, representation module, relationship module, result module and/or data display organization module, the like or combination of these. A module can manipulate and/or transform data and/or information. Data and/or information derived from or transformed by a module can be transferred to another machine and/or suitable module, non-limiting examples of which include a logic processing module, sequencing module, mapping module, counting module , filtering module, weighting module, normalization module, CG trending module, level module, comparison module, range adjustment module, categorization module, plotting module, representation module, relationship module, module result and/or data display organization module, the similar or a combination thereof. A machine comprising a module may comprise at least one microprocessor. In some embodiments, data and/or information is received by and/or provided by a machine comprising a module. A machine comprising a module may include a microprocessor (e.g. one or more microprocessors) which microprocessor may execute and/or implement one or more instructions (e.g. processes, routines and/or subroutines) of a module. In some embodiments, a module operates with one or more external microprocessors (e.g., an internal or external network, server, storage system, and/or storage network (e.g., a cloud)).

logic processing module

[498] In certain embodiments a logic processing module orchestrates, controls, limits, organizes, orders, distributes, divides, transforms and/or regulates data and/or information or the transfer of data and/or information to and from one or more other modules, peripherals or devices.

Data display organization module

[499] In certain embodiments a data display organization module processes and/or transforms data and/or information into a suitable visual medium and non-limiting examples of which include images, video and/or text (e.g. numbers, letters and symbols). In some embodiments, a data display organization module processes, transforms and/or transfers data and/or information for presentation on a suitable display (e.g., a monitor, LED, LCD, CRT, the like, or combinations thereof), a printer, a peripheral, or a suitable device. In some embodiments a data display organization module processes, transforms and/or transfers data and/or information into a visual representation of a fetal or maternal genome, chromosome or part thereof.

sequencing module

[500]In some embodiments, a sequence module obtains, generates, joins, assembles, manipulates, transforms, processes, transforms, and/or transfers fragments (reads) of sequences. A "sequence receiving module", as used herein, is the same as a "sequencing module". A machine comprising a sequencing module can be any machine that determines the sequence of a nucleic acid using sequencing technology known in the art. In some embodiments a sequencing module may align, assemble, fragment, complement, reverse complement, error check, or correct sequence reads.

mapping module

[501]Fragments (reads) of sequences may be mapped by a mapping module or by a machine comprising a mapping module, which mapping module generally maps fragments (reads) to a reference genome or segment thereof. A mapping module may map sequencing reads by a suitable method known in the art. In some embodiments, a mapping module or a machine comprising a mapping module is required to provide reads of mapped sequences.

counting module

[502] Counts can be provided by a counting module or by a machine comprising a counting module. In some embodiments a counting module counts fragments (reads) of sequences mapped to a reference genome. In some embodiments a counting module generates, assembles, and/or provides counts according to a counting method known in the art. In some embodiments, a counting module or a machine comprising a counting module is required to provide counts.

filtering module

[503]Filtering portions (eg, portions of a reference genome) may be provided by a filtering module (eg, by a machine comprising a filtering module). In some embodiments, a filtering module is required to provide filtered portion data (e.g., filtered portions) and/or to remove portions from consideration. In certain embodiments a filtering module removes counts mapped to a consideration portion. In certain embodiments a filtering module removes counts mapped to a portion of a determination of a level or a profile. A filtering module can filter the data (e.g., counts, counts mapped to portions, portions, portion levels, normalized counts, raw counts, and the like) by one or more filtering methods known in the art or described herein.

weighting module

[504]Weighting portions (eg, portions of a reference genome) may be provided by a weighting module (eg, by a machine comprising a weighting module). In some embodiments, a weighting module is required to weight genomic sections and/or provide portion weighted values. A weighting module can weight portions by one or more weighting methods known in the art or described herein.

normalization module

[505]Normalized data (eg, normalized counts) may be provided by a normalization module (eg, by a machine comprising a normalization module). In some embodiments, a normalization module is required to provide normalized data (eg, normalized counts) obtained from sequencing reads. A normalization module can normalize data (e.g., counts, filtered counts, raw counts) with one or more normalization methods described herein (e.g., PERUN, hybrid normalization, the like, or combinations thereof) or known in the art.

GC Trend Module

[506]Determining the GC bias (eg, determining the GC bias for each of the portions of a reference genome (eg, portions, portions of a reference genome)) can be provided by a GC bias module (e.g. by a machine comprising a CG trending module). In some embodiments, a GC trend module is required to provide a GC trend determination. In some embodiments, a GC trend module provides a determination of the GC trend of a fitted relationship (e.g., a fitted linear relationship) between sequence read counts mapped to each of the portions of a reference genome and GC content of each serving. A CG trend module is sometimes part of a normalization module (eg PERUN normalization module).

level module

[507]Determining levels (e.g., levels) and/or calculating genomic section levels for portions of a reference genome may be provided by a level module (e.g., by a machine comprising a level module ). In some embodiments, a level module is required to provide a calculated level or genomic section level (eg, according to Equation A, B, L, M, N, O and/or Q). In some embodiments a level module provides a level of a fitted relationship (eg, a fitted linear relationship) between a CG trend and sequence read counts mapped to each of the portions of a reference genome. In some embodiments a level module computes a genomic section level as part of PERUN. In some embodiments, a level module provides a genomic level section (i.e., Li) according to the equation Li = (mi - GiS)I-1 where Gi is the GC trend, mi is the measured count mapped for each portion of a reference genome, i is a sample, and I is the intercept, and S is the slope of the fitted relationship (for example, a fitted linear relationship) between a GC trend and sequence read counts mapped to each of the portions of a reference genome.

Comparison module

[508]A first tier may be identified as significantly different from a second tier by a comparison module or a machine comprising a comparison module. In some embodiments, a comparison module or a machine comprising a comparison module is required to provide a comparison between two levels.

Range adjustment module

[509]Expected ranges (e.g., expected level ranges) for various copy number variations (e.g., duplications, insertions, and/or deletions) or ranges for the absence of a copy number variation may be provided by a swathing module or a machine comprising a swathing module. In certain embodiments, expected levels are provided by a ranging module or by a machine comprising a ranging module. In some embodiments, a range adjustment module or by a machine comprising a range adjustment module is required to provide the expected levels and/or ranges.

categorization module

[510]A copy variation number (e.g., a maternal and/or fetal copy number variation, a copy number variation, a fetal duplication, insertion, deletion) may be categorized by a categorization module or by a machine comprising a categorization module. In certain embodiments a copy number variation (eg, a maternal and/or fetal copy number variation) is categorized by a categorization module. In certain embodiments a tier (eg, a first tier) determined to be significantly different from another tier (eg, a second tier) is identified as representative of copy number variation by a categorization module. In certain embodiments, the absence of a copy number variation is determined by a categorization module. In some embodiments, a determination of a copy number variation may be determined by a machine comprising a categorization module. A categorization module can be specialized to categorize a maternal and/or fetal copy number variation, a fetal copy number variation, duplication, deletion or insertion or lack thereof, or combination of the foregoing. For example, a categorization module that identifies a maternal deletion may be different than and/or distinct from a categorization module that identifies a fetal duplication. In some embodiments, a categorization module or a machine comprising a categorization module is required to identify a copy number variation or a determinant result of a copy number variation.

adjustment module

[511]In some embodiments, one-level adaptations (e.g., adjustments to genomic section levels, one level to a profile, one level to a copy number variation, one level to one or more moieties, the like, or combinations thereof) are made by an adjustment module or by a machine comprising an adjustment module. In some embodiments, an adjustment module or a machine comprising an adjustment module is required to adjust a level. A level adjusted by methods described herein can be independently verified and/or adjusted by other tests (eg, by target sequencing of maternal or fetal nucleic acid).

plotting module

[512]In some embodiments a plotting module processes and/or transforms data and/or information into a suitable visual medium, non-limiting examples of which include a graph, table, chart, the like or combinations thereof. In some embodiments a plotting module processes, transforms and/or transfers data and/or information for presentation on a suitable monitor (e.g., a monitor, LED, LCD, CRT, the like or combinations thereof), a printer, a peripheral or suitable device. In certain embodiments a plot module provides a visual display of a count, a level, and/or a profile. In some embodiments, a data display organization module transforms data and/or information into a visual representation of a fetal or maternal genome, chromosome, or part thereof. In some embodiments, a plotting module or a machine comprising a plotting module is required to plot a count, level or profile.

Relationship Module

[513]In certain embodiments, a relationship module processes and/or transforms data and/or information in a relationship. In certain embodiments, a relationship is generated by and/or transferred from a relationship module.

result module

[514]The presence or absence of a genetic variation (an aneuploidy, a fetal aneuploidy, a copy number variation) is, in some embodiments, identified by a result module or by a machine comprising a result module. In certain embodiments a genetic variation is identified by an outcome module. Often, the determination of the presence or absence of an aneuploidy is identified by a result module. In some embodiments, a determinant result of a genetic variation (an aneuploidy, a copy number variation) can be identified by a result module or by a machine comprising a result module. An outcome module can be specialized to determine a specific genetic variation (eg a trisomy, trisomy 21, trisomy 18). For example, an output module that identifies a trisomy 21 may be different than and/or distinct from an output module that identifies a trisomy 18. In some embodiments, an output module or a machine comprising an output module is necessary to identify a genetic variation or a determinant result of a genetic variation (eg, an aneuploidy, a copy number variation). A genetic variation or a determinative result of a genetic variation identified by methods described herein can be verified by other tests (eg, by target sequencing of maternal and/or fetal nucleic acid).

Transformations

[515]As mentioned above, data is sometimes transformed from one form to another form. The terms "transformed", "transformation", and grammatical derivations or their equivalents, as used herein, refer to a data alteration of a physical starting material (e.g. test subject and/or sample nucleic acid of the reference subject) into a digital representation of the physical starting material (e.g., readable sequence data), and in some embodiments further includes a transformation into one or more numeric values or graphical representations of the digital representation that can be used to provide a result (for example, determination of the fetal fraction or estimation of a test sample). In certain embodiments, one or more numeric values and/or graphical representations or digitally represented data may be used to represent the appearance of a test subject's physical genome (e.g., virtually represent or visually represent the presence or absence of an insert , genomic duplication or deletion; representing the presence or absence of a variation in the physical quantity of a sequence associated with medical conditions). A virtual representation is sometimes further transformed into one or more numerical values or graphical representations of the digital representation of the starting material. These methods can transform the physical starting material into a numerical value or graphical representation, or a representation of the physical appearance of a test subject's genome.

[516]In some embodiments, transforming a dataset facilitates providing a result by reducing data complexity and/or data dimensionality. Dataset complexity is sometimes reduced during the process of transforming a physical starting material into a virtual representation of the starting material (eg, fragments (reads) of sequences representative of physical starting material). A suitable characteristic or variable can be used to reduce the complexity and/or dimensionality of the data set. Non-limiting examples of traits that may be chosen for use as a target trait for data processing include GC content, fetal gender prediction, fragment size (e.g. length of CCF fragments, reads) or a suitable representation (e.g. FRS)), fragment sequence, identification of chromosomal aneuploidy, identification of particular genes or proteins, identification of cancer, diseases, inherited genes/traits, chromosomal abnormalities, a biological category, a chemical category, a biochemical category , a category of genes or proteins, a gene ontology, a protein ontology, co-regulated genes, cell signaling genes, cell cycle genes, proteins related to previous genes, genetic variants, protein variants, co-genes regulated, co-regulated proteins, amino acid sequence, nucleotide sequence, protein structure data and the like, and combinations of the foregoing. Non-limiting examples of dataset complexity and/or dimensionality reduction include; reducing a plurality of sequence reads to profile graphs, reducing a plurality of sequence reads to numerical values (eg, normalized values, Z scores, p values); reducing multiple analysis methods to probability plots or single points; principal component analysis of derived quantities; and the like or combinations thereof.

Certain system, machine and computer program product modes

[517] There is in certain aspects provided a computer-implemented method for determining the presence or absence of a genetic variation, comprising (a) obtaining fragment counts (reads) of nucleotide sequences mapped to genomic sections of a reference genome , the sequence reads of which are: (i) circulating cell-free nucleic acid reads from a test sample from a pregnant woman, and (ii) nucleic acid fragment reads having lengths that are less than a length of the selected fragment; (b) normalizing the counts, thereby generating normalized counts of sequence reads mapped to the genomic sections; and (c) determining the presence or absence of a genetic variation according to the normalized counts.

[518] Also provided in certain aspects is a system comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors, and which memory comprises counts of fragments (reads) of nucleotide sequences mapped to genomic sections of a reference genome, the sequence reads of which are (i) circulating cell-free nucleic acid reads from a test sample of a pregnant woman, and (ii) nucleic acid fragment reads having lengths that are less than a selected fragment length; and that instructions executable by one or more microprocessors are configured to (a) normalize the counts, thereby generating normalized counts of fragments (reads) of sequences mapped to genomic sections; and (b) determining the presence or absence of a genetic variation according to the normalized counts.

[519] There is also provided in certain aspects a machine comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors and which memory comprises counts of fragments (reads) of nucleotide sequences mapped to genomic sections of a reference genome, the sequence reads of which are (i) circulating cell-free nucleic acid reads from a test sample of a pregnant woman, and (ii) nucleic acid fragment reads having lengths that are less than a selected fragment length; and that instructions executable by one or more microprocessors are configured to (a) normalize the counts, thereby generating normalized counts of fragments (reads) of sequences mapped to genomic sections; and (b) determining the presence or absence of a genetic variation according to the normalized counts.

[520] Also provided in certain embodiments is a computer program product tangibly embodied in a computer-readable medium, comprising instructions that, when executed by one or more microprocessors, are configured to (a) access fragment counts (reads ) of nucleotide sequences mapped to genomic sections of a reference genome, the sequence reads of which are: (i) circulating cell-free nucleic acid reads from a test sample from a pregnant woman, and ( ii) fragments (reads) of nucleic acid fragments having lengths that are smaller than a selected fragment length; and that instructions executable by one or more microprocessors are configured to (a) normalize the counts, thereby generating normalized counts of fragments (reads) of sequences mapped to genomic sections; and (b) determining the presence or absence of a genetic variation according to the normalized counts.

[521] Also provided herein is a system comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors, and which memory comprises fragments (reads) of nucleotide sequences mapped to portions of a reference genome, whose Sequence reads are circulating cell-free nucleic acid reads from a test specimen from a pregnant woman, and which instructions executable by one or more microprocessors are configured to (a) weight, using a microprocessor, ( i) the sequence reads counts mapped to each chunk, or (ii) another chunk-specific parameter, to a specific fraction of the chunk of fetal nucleic acid according to a weighting factor independently associated with each chunk providing, thus, fetal portion-specific fraction estimates according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of several samples , and (ii) counts of sequence reads mapped to each chunk, or other chunk-specific parameter, for multiple samples, and (b) estimating a fraction of fetal nucleic acid for the test sample based on the fraction estimates specific to the fetal portion.

[522] Also provided herein is a machine comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors, and which memory comprises fragments (reads) of nucleotide sequences mapped to portions of a reference genome, which sequence reads are circulating cell-free nucleic acid reads of a test sample from a pregnant woman, and which instructions executable by one or more microprocessors are configured to (a) weight, using a microprocessor, (i) counts of sequence reads mapped to each chunk, or (ii) another chunk-specific parameter, to a specific fraction of the chunk of fetal nucleic acid according to a weighting factor independently associated with each chunk providing , therefore, fetal portion-specific fraction estimates according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of several samples, and (ii) counts of sequence reads mapped to each chunk, or other chunk-specific parameter, for multiple samples, and (b) estimating a fraction of fetal nucleic acid for the test sample based on the estimates of specific fraction of the fetal portion.

[523] Also provided herein is a non-transient computer-readable storage medium with an executable program stored thereon, where the program instructs a microprocessor to perform the following: (a) access fragments (reads) of nucleotide sequences mapped to portions of a reference genome, the sequence reads of which are circulating cell-free nucleic acid reads from a test sample from a pregnant woman, (b) weight, using a microprocessor, (i) the counts of sequence reads mapped to each chunk, or (ii) another chunk-specific parameter, to a specific fraction of the chunk of fetal nucleic acid according to a weighting factor independently associated with each chunk, thereby providing estimates specific fraction of the fetal portion according to the weighting factors, wherein each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of several samples, and (ii) counts of sequence reads mapped to each portion, or other portion-specific parameter, for multiple samples, and (c) estimating a fetal nucleic acid fraction for the test sample based on the fetal portion-specific fraction estimates

[524] In certain embodiments, a system, machine, and/or computer program product comprises a counting module configured to count fragments (reads) mapped to genomic sections of a reference genome or portion thereof (e.g., subset of genomic sections, selected set of genomic sections). A counting module is often configured to count fragment reads of nucleic acid having lengths that are less than a selected fragment length. Counts are sometimes filtered counts, normalized counts, or a combination of the above. In some embodiments, a counting module may normalize the counts, for example, using any suitable normalization process described herein or known in the art.

[525] In some embodiments, a system, machine, and/or computer program product comprises a count comparison module. A count comparison module is often configured to compare the number of fragment counts (reads) counted by a count module, thereby making a count comparison. A count comparison module is often configured to access, receive, use, store, search, and/or align counts of fragments (reads) (for example, from a count module or normalization module). A count comparison module is often configured to provide a suitable comparison between counts, non-limiting examples of which comparison include a simple comparison (e.g. match or mismatch between counts of fragments (reads) mapped to a first set of genomic sections compared to a second set of genomic sections), mathematical comparison (eg, ratio, percentage), statistical comparison (eg, multiple comparisons, multiple tests, standardization (eg, Z-score analyses)), the similar and combinations thereof. A suitable count comparison value can be provided by a count comparison module, non-limiting examples of which include the presence or absence of a match between the counts, a ratio, percentage, z-score, a value coupled with a measure of variance or uncertainty (eg, standard deviation, median absolute deviation, confidence interval), like, and combinations thereof. A count comparison module is sometimes configured to transmit a comparison value to another module or machine, such as a genetic variation module, display machine or printing machine, for example.

[526] In certain embodiments, a system, machine, and/or computer program product comprises a module of genetic variation. A genetic variation module is sometimes configured to provide a determination of the presence or absence of a genetic variation according to fragment counts (reads) mapped to genomic sections of a reference genome. A genetic variation module is sometimes configured to provide a determination of the presence or absence of a genetic variation based on a comparison of counts. A genetic variation module is often configured to access, receive, use, store, search and/or align one or more comparisons of a counts comparison module and/or counts of a count module. A genetic variation module can determine the presence or absence of a genetic variation from one or more comparisons or counts in an appropriate manner. A module of genetic variation sometimes determines whether there is a significant difference between counts for different sets of genomic sections in a reference genome. Difference significance can be determined through a genetic variation module in an appropriate manner (eg percent difference, Z score analysis). A genetic variation module sometimes determines whether a count determination or a count comparison is in a particular category. For example, a module of genetic variation can categorize a particular comparison to a threshold of a particular proportion or range of ratios associated with an euploid determination, or a threshold of a particular proportion or range of ratios associated with an aneuploid determination. In another non-limiting example, a genetic variation module can categorize a particular count determination to a particular ratio threshold or range of ratios associated with an euploid determination, or a threshold of a particular ratio or range of ratios. proportions associated with an aneuploid determination. A genetic variation module can provide a result in a suitable format, which is sometimes a binding pertaining to a genetic variation optionally associated with a measure of variance or uncertainty (e.g. standard deviation, median absolute deviation, precision (e.g. , within a particular confidence interval.) A genetic variation module is sometimes configured to transmit a determination of the presence or absence of a genetic variation to another module or machine, such as a display machine or printer, for example .

[527] A machine or system comprising a module described herein (e.g., a reference comparison module) may comprise one or more microprocessors. In some embodiments, a machine or system may include multiple microprocessors, such as microprocessors coordinated and working in parallel. A microprocessor (eg, one or more microprocessors) in a system or machine can execute and/or implement one or more instructions (eg, processes, routines and/or subroutines) in a module described herein. A module described here is sometimes located in memory, or associated with a machine or system. In some embodiments, a module described herein operates with one or more external microprocessors (e.g., an internal or external network, server, storage system, and/or storage network (e.g., a cloud)). In some embodiments, a module described herein is configured to access, join, gather and/or receive data and/or information from another module, machine or system (eg, component, peripheral). In some embodiments, a module described herein is configured to provide and/or transfer data and/or information to another module, machine, or system (e.g., component, peripheral). In some embodiments, a module described herein is configured to access, accept, receive and/or merge input data and/or information from an operator of a machine or system (i.e., user). For example, sometimes a user provides a constant, threshold value, formula, and/or predetermined value for a module. A module described here is sometimes configured to transform data and/or information that it accesses, receives, unites and/or gathers.

[528] In certain embodiments, a system, machine and/or computer program product comprises (i) a sequencing module configured to obtain and/or access fragments (reads) of nucleic acid sequences and/or fragments (reads) of partial nucleotide sequences; (ii) a mapping module configured to map fragments (reads) of nucleic acid sequences to portions of a reference genome; (iii) a counting module configured to provide counts of fragments (reads) of nucleic acid sequences mapped to portions of a reference genome; (iv) a normalization module configured to provide normalized counts; (v) a comparison module configured to provide an identification of a first elevation that is significantly different from a second elevation; (vi) a range adjustment module configured to provide one or more expected level ranges; (vii) a categorization module configured to identify a representative elevation of a copy number variation; (viii) an adjustment module configured to adjust a level identified as a copy number variation; (ix) a plotting module configured to plot and display a level and/or a profile; (x) a result module configured to determine the presence or absence of a genetic variation, or determine an outcome (eg, result determining the presence or absence of a fetal aneuploidy); (xi) a data display organization module configured to display a determination of genetic variation; (xii) a logic processing module configured to execute one or more mapped sequence reads, count mapped sequence reads, normalize counts and generate a result; (xiii) a count comparison module, (xiv) a fetal fraction module configured to provide a fetal fraction determination; (xv) a genetic variation module configured to provide a determination of the presence or absence of a genetic variation; or (xvi) combination of two or more of the foregoing.

[529] In some embodiments a sequencing module and mapping module are configured to transfer fragments (reads) of sequences from the sequencing module to the mapping module. The counting module and mapping module are sometimes configured to transfer reads of mapped sequences from the mapping module to the counting module. In some embodiments, the normalization module and/or the comparison module are configured to transfer normalized counts to the comparison module and/or the sizing module. The comparison module, the range adjustment module and/or the categorization module are independently configured to transfer (i) an identification of a first elevation that is significantly different from a second elevation and/or (ii) a range of the level expected from the comparison module and/or range adjustment module to the categorization module, in some embodiments. In certain embodiments, the categorization module and the adjustment module are configured to transfer a categorized elevation as a copy number variation from the categorization module to the adjustment module. In some embodiments, the fit module, plot module, and result module are configured to transfer one or more adjusted levels from the fit module to the plot module or result module. The normalization module is sometimes configured to transfer readable counts of mapped normalized sequences to one or more of the compare module, range fit module, categorization module, fit module, result module, or plot module.

Examples

[530]The following examples are provided by way of illustration only and not by way of limitation. Thus, the examples presented below illustrate certain modalities and do not limit the technology. Those skilled in the art will readily recognize a variety of non-critical parameters that could be altered or modified to obtain essentially the same or similar results.

Example 1: PERUN and general methods for detecting conditions associated with genetic variations

[531]The methods and underlying theory described herein can be used to detect various conditions associated with genetic variation and provide a determinant result, or to determine the presence or absence of a genetic variation.

Removal of uninformative portions of a reference genome

[532]Several attempts to remove uninformative portions of a reference genome have indicated that portion selection has the potential to improve classification. Equation A: M = LI + GS

[533]The various terms in Equation A have the following meanings: • M: measured counts, representing primary information polluted by unwanted variation. • L: chromosome level - this is the desired output of the data processing process. L indicates fetal and/or maternal euploid aberrations. This is the amount that is masked both by stochastic errors and by systematic biases. The L chromosome level is both sample-specific and portion-specific. • G: GC trend coefficient measured using linear model, LOESS, or any equivalent approach. G represents secondary information, extracted from M and a set of portion-specific GC content values, usually derived from the reference genome (but can be derived from actually observed GC contents as well). G is sample-specific and does not vary across genomic position. It encapsulates a portion of the unwanted variation. • I: Linear model intercept. This model parameter is fixed for a given experimental setup, sample-independent, portion-specific. • S: Slope of the linear model. This model parameter is fixed for a given experimental setup, sample-independent, and portion-specific.

[534]Quantities M and G are measured. Initially, the specific values of the I and S portion are unknown. To evaluate unknown I and S, we must assume that L = 1 for all portions of a reference genome in euploid samples. The assumption is not always true, but it can reasonably be expected that any samples with deletions/duplications will be overwhelmed by samples with normal chromosome levels. A linear model applied to the euploid samples extracts the specific I and S parameter values for the selected portion (assuming L = 1). The same procedure is applied to all portions of a reference genome in the human genome, obtaining a set of I intercepts and S slopes for each genomic locus. Cross-validation randomly selects a working set that contains 90% of all LDTv2CE euploids and uses this subset to train the model. The random selection is repeated 100 times, obtaining a set of 100 slopes and 100 intercepts for each portion.

Chromosomal level extraction of measured counts

[535]Assuming that model parameter values I and S are available for each portion, M measurements collected on a new test sample are used to assess the chromosome level according to the following Equation B: L = (M-GS )/I (B) As in Equation A, coefficient of GC bias is evaluated as the linear regression coefficient between raw counts measured in M portions and the GC content of the reference genome. The L chromosome level is then used for further analysis (Z values, maternal deletions/duplications, maternal microdeletions/microduplications, fetal gender, sex aneuploidies, and so on). The procedure encapsulated by Eq. B is called Parameterized Error Removal and Unbiased Normalization (PERUN).

Example 2: Examples of formulas

[536]Non-limitative examples of mathematical and/or statistical formulas that can be used in methods described herein are provided below.

[537]Z scores and p-values calculated from Z scores associated with deviations from the expected level of 1 can then be evaluated in light of the estimate for the uncertainty at the mean level. The p values are based on a t distribution whose order is determined by the number of portions of a reference genome in a peak. Depending on the desired confidence level, a cut can suppress noise and allow unambiguous detection of the real signal. Equation 1: Equation 1 can be used to directly compare peak levels from two different samples, where N and n refer to the numbers of portions of a reference genome across the chromosome and within the aberration, respectively. The order of the t-test that will produce a p-value measuring the similarity between two samples is determined by the number of portions of a reference genome in the shorter of the two deviant stretches.

[538]Equation 8 can be used to incorporate fetal fraction, maternal ploidy, and mean reference counts into a scoring scheme for determining the presence or absence of a genetic variation with respect to fetal aneuploidy. Equation 8: where Yi represents the measured counts for a portion in the test sample corresponding to the portion in the median count profile, F represents the fetal fraction, X represents fetal ploidy, and Mi represents maternal ploidy assigned to each portion. Possible values for X used in equation (8) are: 1 if the fetus is euploid; 3/2, if the fetus is triploid; and, 5/4, if there are twin fetuses and one is affected and one is not. 5/4 is used in the case of twins where one fetus is affected and the other is not, because the F term in equation (8) represents the total DNA of the fetus, therefore, all fetal DNA must be taken into account. In some embodiments, large deletions and/or duplications in the maternal genome can be accounted for by assigning maternal ploidy, Mi to each portion or portion. Maternal ploidy is often designated as a multiple of 1/2, and can be estimated using normalization and portions, in some modalities. Because maternal ploidy is often a multiple of 1/2, maternal ploidy can be easily explained, and therefore will not be included in other equations to simplify derivations.

[539]When evaluating equation (8) with X = 1, (eg euploid assumption), the fetal fraction is canceled and the following equation results from the sum of squared residuals. Equation 9: To simplify equation (9) and subsequent calculations, the following equations are used. Equation 10: Equation 11: Equation 12: When evaluating equation (8) with X = 3/2 (eg, triploid assumption), the following equation results from the sum of squared residuals. Equation 13: The difference between equations (9) and (13) forms the functional result (e.g. phi) which can be used to test the null hypothesis (e.g. euploid, X = 1) against the alternative hypothesis (trisomy singlet, X = 3/2): Equation 14: Equation 18: Ideal ploidy value is sometimes given by Equation 20: The term for maternal ploidy, Mi, may be omitted for some mathematical derivations. The resulting expression for X corresponds to the relatively simple and often-frequently occurring special case of when the mother has no deletions or duplications in the chromosome or chromosomes being evaluated. Equation 21: Xiff and Xify are provided by equations (11) and (12), respectively. In embodiments where all experimental errors are negligible, solution equation (21) results in a value of 1 for euploids where Xiff = Xify. In certain embodiments where all experimental errors are negligible, solution equation (21) results in a paratriploid value of 3/2 (see equation (15) for triploid relationship between Xiff and Xify. Table 2

Example 3: Portion selection using FRS.

[541] Portions of the human reference genome designated HG19 were first prefiltered using a PERUN-based method that removes portions with high variability, low mappability, and binds with a high percentage of repetitive elements. Portions (as selected for LDTv2) with high variability, low mappability, and a large fraction of repeated sequences were excluded. For each 50 kb chunk (eg, chunk), a fetal statistical ratio was calculated for fragments (reads) of final sequence paired CCF fragments less than 150 bases and CCF fragments less than 600 bases. The FRS was then averaged across 264 unpooled samples processed using TruSeq biochemistry library preparation with the automated bead cleanup. Portions with FRS > median (FRS) were selected and are presented in Table 4 with reference to specific chromosome start and end positions. Chromosome-specific start and end positions in Table 4 reference nucleotide base positions in human reference genome HG19.

[542]All portions with FRS > median (FRS) were plotted concurrently with the number of single exon start positions in each respective portion. A significant correlation was shown for gene regions that contain an overrepresentation of small fragments (Figures 1 - 9). A significantly stronger correlation was shown with GC content (percentage of GC bases in a 50kb portion) and FRS (Table 3).

[543] Portion selection was further restricted to portions (ie, portions) of the genome where FRS > median (FRS) for detection of chromosomal trisomy. Applying this approach to a preliminary dataset of 264 samples provided consistent ranking margins for weighting to discard 50% of the data. On the other hand, by restricting portions where FRS < median (FRS), the classification margin was drastically reduced, which suggests a dilution of fetal DNA for analysis (Figures 10 - 11).

[544]In Figure 10 and Figure 11 there are two regression lines, one for non-T21 samples only (dash-dot line) and another for T21 samples (dotted line). The regression line for T21 samples based on high FRS portions was above the regression line for non-T21 samples based on high FRS (Figure 10). On the other hand, this similar regression was lower than the non-T21 samples when comparing Z scores calculated on low portions of FRS (Figure 11). This suggests that using high FRS portions can improve the accuracy of outcome determinations, as Z scores tend to be higher for T21 samples. Table 3

Example 4 Detection of trisomy 21 using a combination of sequence-based sorting and length-based analyzes

[545]Plasma samples containing the circulating cell-free DNA obtained from pregnant women are tested for trisomy 21 using the following method.

Sequence based sorting

[546] A SURESELECT custom capture library is obtained from Agilent that includes a set of custom designed biotinylated capture RNAs. Capture RNAs are designed according to specific nucleotide sequences on chromosome 21 (test chromosome) and specific on chromosome 14 (reference chromosome) and are identified by Agilent's EARRAY web-based design tool. 100 independent capture RNAs are designed for each of chromosomes 14 and 21. Single copy nucleotide sequences in the range of 40 to 60 base pairs that are unique to chromosome 14 or 21 and are rich in AT are selected for the design of Custom RNA.

[547]Nucleic acid sample, which is nucleic acid from circulating cell-free plasma of a pregnant woman in the first trimester of pregnancy, is divided into two tubes and incubated with either chromosomal capture RNA 21 or capture RNA 14 for 24 hours at 65°C, according to the manufacturer's instructions. After hybridization, the captured target fragments and captured reference fragments (collectively referred to as captured fragments) are selected by pulling down the RNA/biotinylated fragment hybrids using streptavidin-coated magnetic beads (DYNAL DYNAMAG-2, Invitrogen, Carlsbad, CA ), and purified with the MinElute PCR purification kit (Qiagen, Germantown, MD). Capture RNA is digested and the remaining DNA fragments are amplified according to the manufacturer's instructions.

Length based analysis

[548]Samples containing the above separated nucleic acid fragments are hybridized under stringent hybridization conditions to poly-inosine probes comprising biotinylated inosine, which probes are longer than the DNA fragments to which they hybridize and 500 base pairs long. In some embodiments, hybridization is performed overnight at 65°C in 6 x SSC and 1% SDS In some embodiments, hybridization is performed overnight at 43°C in 1.0 M NaCl, sodium phosphate buffer 50 mM sodium (pH 7.4), 1.0 mM EDTA, 2% (w/v) sodium dodecyl sulfate, 0.1% (w/v) gelatin, 50 μg/mL tRNA and 30% (v/v) formamide. Four 30-minute washes are performed at 55°C in 1.2X SSC (1X SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 10 mM sodium phosphate (pH 7.4), 100 ml EDTA 1.0 mM and 0.5% (w/v) sodium dodecyl sulfate. After hybridization, portions of the unhybridized probe are digested using exonuclease I (New England Biolabs, Ipswich, MA) and phosphodiesterase II (Worthington Biochemical Corp., Lakewood, NJ). The probe-fragment duplexes are denatured at 95°C for two minutes, and the probes are separated away from the fragments (i.e., pulled down) using streptavidin-coated magnetic beads (Dynal DYNAMAG-2, Invitrogen, Carlsbad, CA) , and purified with the MinElute PRC Purification Kit (Qiagen, Germantown, MD). Trimmed, isolated and purified polyinosine probes are measured for mass using MALDI mass spectrometry. Probe length, and thus the corresponding fragment length, is extrapolated from the mass peaks for each probe length species by comparison with mass peaks for biotinylated polyinosine standards of known length.

Determination of trisomy 21

[549]The relative amount of each fragment length species is determined based on the amplitude of the mass peaks for each probe length species. Fragments of 150 base pairs or less are quantified for chromosome 14 and chromosome 21. Samples with substantially equal amounts of chromosome 14 and chromosome 21 fragments are determined to be euploid for chromosome 21. chromosome 21 versus chromosome 14 (for example, the 2% elevation in fragments of chromosome 21 versus chromosome 14) are determined as triploid to chromosome 21.

Example 5: Trisomy detection using fragment length filtering and chromosome representation

[550]In this example, maternal samples containing cell-free nucleic acid were classified as carrying a euploid fetus or a fetus having an aneuploidy (i.e., trisomy 13, trisomy 18, trisomy 21) based on readable sequence counts of nucleotides from a subset of fragments having certain length parameters. Samples were obtained from the Women's and Children's Hospital (WI study; Palomaki et al. (2011) Genet Med 13(11):913-20). Nucleotide sequence reads (base 36 reads) for each sample were obtained using an Illumina paired-end sequencing platform (Illumina, Inc., San Diego, CA). Fragments (reads) of paired-end nucleotide sequences were aligned to a reference genome (build 37 (hg19)) using the BOWTIE 2 beta 3 alignment program and fragment length was determined based on the alignments of the fragments (reads) paired end.

[551]Certain fragments (reads) of nucleotide sequences were filtered according to the following nucleic acid fragment length parameters and: 1) fragments having lengths greater than or equal to 120 bases; 2) fragments having lengths greater than or equal to 130 bases; 3) fragments having lengths greater than or equal to 140 bases; 4) fragments having lengths greater than or equal to 150 bases; 5) fragments having lengths greater than or equal to 160 bases; or 6) fragments having lengths greater than or equal to 170 bases. Thus, end-paired reads corresponding to fragments equal to or greater than a given length threshold (e.g., 120 bases, 130 bases, 140 bases, 150 bases, 160 bases, 170 bases) were filtered and fragments ( end-paired reads) corresponding to fragments shorter than a given length threshold were retained for analysis.

[552]Chromosome representations for chromosome 13, chromosome 18 and chromosome 21 were calculated for datasets shown in Figure 23 using 1) unfiltered sequence reads and 2) length-filtered sequence reads on a threshold of 150 base fragments. Chromosome representations for each chromosome 13, 18, and 21 were calculated according to the following: Chromosome 13 (Chr 13) representation = ∑ Chr 13 (unfiltered) sequence readable counts/ ∑ all autosomal (unfiltered) sequence readable counts (filtered) Chromosome 13 (Chr 13) representation = ∑ Chr 13 sequence readable counts (filtered)/ ∑ all autosomal sequence readable counts (filtered) Chromosome 18 (Chr 18) representation = ∑ Chr 18 sequence readable counts ( unfiltered)/ ∑ all readable autosomal sequence counts (unfiltered) Chromosome 18 (Chr 18) representation = ∑ readable counts of Chr 18 sequence (filtered)/ ∑ all readable autosomal sequence counts (filtered) Chromosome representation 21 (Chr 21) = ∑ readable counts of Chr 21 sequence (unfiltered)/ ∑ all readable counts of autosomal sequence (unfiltered) Representation of chromosome 21 (Chr 21) = ∑ readable counts of Chr 21 sequence (filtered)/ ∑ all readable autosomal sequence counts (filtered) Figures 14, 16 and 18 show chromosome representations for chromosomes 13, 18 and 21, respectively, using unfiltered sequence reads. Figures 15, 17 and 19 show chromosome representations for chromosomes 13, 18 and 21, respectively, using length-filtered sequence reads. For filtered datasets, chromosome representation increased for the trisomy samples, due in part to an increase in contributed fetal sequence data. While this increase in chromosome representation may increase the power to detect chromosomal abnormalities, the chromosome representation variance for non-trisomy samples increased due to an approximate 63-82% reduction in fragment counts (read). Example distributions of readable counts at various threshold values of fragment length are illustrated in Figure 13 and presented in Table 5 below. Table 5

[553]The average area under the curve (AUC) for fragment reads of less than a certain length was determined to illustrate the total reduction of fragments (reads) (i.e. sequence coverage) seen on average . For a given assay that generates about 15 million sequence reads (or 0.2X coverage of the human genome), excluding reads longer than 150 bases, for example, is equivalent to about 0.035X coverage.

[554]To determine an optimal fragment size threshold for chromosome representation, fragment size threshold ranged from 120 to 170 bases, in 10 base increments. Chromosome representation (i.e., for chromosomes 13, 18, and 21) was calculated after normalization of readable sequence counts (i.e., PERUN filled with LOESS) for each length-filtered dataset (end-paired reads). ) and for an unfiltered dataset (single-ended reads; also referred to as “whole”). Representations of chromosomes 13, 18 and 21 are presented in Figures 20, 21 and 22, respectively. Chromosome representation for the filtered datasets at the limit of 150, 160, and 170 bases was quite consistent with the unfiltered dataset. The following tables present the observed specificity and sensitivity for detecting trisomy 13, 18, and 21 at the respective Z score cutoff values (i.e., 3.95 for chromosome 13, 3.95 for chromosome 18, and 3 for chromosome 21). Z score values were based on MAD values from the population and historical dataset-specific and flow cell-specific medians. Additionally, 10-fold cross-validation of receiver operating characteristic (ROC) analyzes were conducted (i.e., 10-fold stratified cross-validation, repeated 100 times) and the mean area under the curve (AUC; i.e., a measurement of precision) for each analysis (calculated by summing all sensitivity time values (1-specificity) and implemented using R ROCR package) is presented in Tables 6, 7 and 8 below.

[555]The data show that weighting a significant reduction in sequence coverage of filtered-length samples, trisomies can be identified using samples filtered at certain fragment length limits (e.g., 150 bases, 160 bases), with an accuracy , similar sensitivity and specificity compared to unfiltered samples.

Example 6:

[556]This example illustrates, in part, a relationship between fetal fraction and fetal statistical proportion (FRS). As shown in Figures 25A and 25B, a plot of Z v. Median FRS per sample showed remarkable age similarity to a plot of Z scores versus FQA-based estimates of the fetal fraction. In addition, median FRS for samples with trisomy 21 restricted to high FRS portions (Figure 25A, dashed band above) was 0.188 and median FRS for samples with trisomy 21 for all portions (Figure 25B, dotted line above) was 0.172 . For non-trisomy Chr21 samples, the median FRS of high FRS portions was 0.181 (Figure 25A, dashed line below) and the median FRS for all portions was 0.166 (Figure 25B, dashed line below). This suggested that the trisomy 21 samples, in fact, have a slightly higher portion representation than the non-trisomy 21 samples, in particular portions with a higher propensity for fetal contribution.

[557]As shown in Figure 26, it was determined that reads of different fragment lengths comprise different GC contents. Smaller fragments, which are known to be more fetal in origin, showed higher CC content compared to larger fragments. The difference in GC content was also related to how FRS correlated with GC content and gene density, as servings (bins) with higher FRS were positively correlated with GC content per serving (bin). These subtle GC differences in fragment length can be leveraged to provide fetal fraction information. For example, difference in GC, fragment length and/or dispersion of fragment length across the reference human genome can be used to predict maternal or fetal origin of fragments. These data demonstrate that GC content per reading can be used to estimate fetal contribution.

[558]PERUN is region-specific additive correction to remove GC bias in readable depth of coverage. This normalization procedure involved a trained estimation of two region-specific parameters, the slope, ie the impact of the GC trend, and the intercept, ie the coverage of the baseline level in the absence of the GC trend. The distribution of PERUN intercepts divided into FRS quantiles suggested that increasing FRS increases PERUN intercepts (Figure 27). Overall, genomic regions with the smallest FRS tend to have the lowest intercepts, possibly due to reduced fetal contribution relative to full coverage representation. In addition, initial region selection efforts incorporated maximum cross-validation errors, where higher values indicated an increase in coverage variability. Figure 28 shows a distribution of maximum cross-validation errors broken down into quantiles. The extreme quantiles (high and low) exhibited greater variability in the stability of the region. As extreme FRS genomic regions are potentially more sensitive to fetal input, the increased variability in maximum cross-validation errors may in fact be due to fetal signal variability.

Example 7: Fetal Fraction based on Portion(bin)

[559]This example demonstrates a method for quantifying the amount of circulating cell-free fetal DNA in a maternal blood sample using sequencing coverage data. The technology encompasses a method known here as Portion-Based Fetal Fraction (bin) (BFF) that uses sequencing coverage maps to quantify the fraction of fetal DNA in a sample of maternal blood. The method takes advantage of machine learning methods to build a model relating to sequencing coverage for fetal fraction.

[560]The first step of the BFF method was to obtain genomic coverage data. Genomic coverage data were obtained from a sequencing and alignment run. These coverage data then served as an indicator for the fetal fraction. Coverage predictor variables may be generated by any suitable method, including, but not limited to, discrete genomic (bin) portion, variable size (bin) portion, or point-based views of a leveled coverage map.

[561]The second step of the BFF method was to train a model to estimate the fetal fraction from coverage data predictors (eg parameters). In this example, a general multiple regression model was trained using simple least squares to estimate fetal fraction directly from the known proportional sequencing level of a particular portion (bin). This approach can be extended to a multivariate multiple regression model to predict bins that are known to be proportional to the fetal fraction (from which the fetal fraction, in turn, can be derived). Similarly, if bins are correlated, multivariate response models can be trained to explain correlated responses. The following is an example in its simplest form: The multiple regression model was chosen as equation 30 below; where Xbin is an mxp matrix of counts of chunks (bins), yff is a vector mx 1 of m number of training samples and p number of predictor chunks (bins), ε is a noise vector with the expectation of E(ε ) = 0, where the covariance Cov(ε) = u2I where I is the identity matrix (that is, the errors are homoscedastic), and rank(Xbin)< p. The yff vector corresponded to a portion (bin) with levels known to be proportional to the fetal fraction.

[562]Without loss of generality, we assume that Xbin was centered by its mean. Thus β the px 1 vector of regression coefficients, can be estimated from the solution of the normal equations for β as;

[563]The extension to the multivariate multiple response model simply extended the previous model to have multiple response variables, or as a yff matrix of size mxn, where n is a number of different bins that have levels proportional to the fraction fetal. The model is therefore; where E is a noise matrix with assumptions parallel to the multiple model. The matrix of coefficients B can be estimated by solving for B in; where B is a matrix of px n.

[564]If rank(Xbin)< p, then the problem can be decomposed into any number of regression models suitable to explain multicollinearity. In addition to this, reduced rank B estimators can also be found such that rank(B) <= min(n, p), which represents the potential correlation within the multivariate response. The resulting estimators can then be averaged or weighted together by a suitable method.

[565]The BFF approach is not limited to this regression method. Many machine learning methods can be used, including but not limited to other multiple regression methods, multivariate response regression, decision trees, support vector machines, and neural networks, to improve estimation. There are also methods that can relax the assumptions and provide the high dimensional estimate so that all relevant bins can be incorporated into the model. Non-limiting examples of such estimators are those based on constraints such as reduced rank, LASSO, weighted rank selection criteria (WRSC), rank selection criteria (RSC), and elastic network estimators that have been shown to improve predictive power.

[566]Fetal fraction predictions have also been improved by measuring and incorporating the genomic coverage trend into the plumber. These biases can come from a number of sources, including but not limited to GC content, DNase1-hypersensitivity, mappability, and chromatin structure. Such profiles can be quantified on a per-sample basis and used to fit genomic coverage data, or added as predictors or constraints to the fetal fraction model.

[567]For example, the multiple regression approach was trained on 6000 male euploid samples, using the relative level of Y chromosome coverage across all portions (bins) as the true fetal fraction value (ChrFF). To avoid circularity with the detection of common trisomies, the model was trained only on portion (bin) of autosomal coverage, and does not include chromosomes 13, 18, or 21. The model demonstrated strong performance on test data consisting of 19,312 samples independent (Figure 29).

[568]BFF's strong performance is driven by the portions (bins) and regions that tend to attract fetal DNA. These regions tend to have greater coverage variance, and the model makes use of this variation. A bootstrap approach was used to compare models formed exclusively in portion (bin) with representation of high or low fetal fraction (based on FRS). Portions (bins) with higher fetal content were shown to be better predictors of fetal fraction (Figure 30). This corresponded with the finding that models built in portion (bin) with greater fetal representation tend to have higher regression coefficients (Figure 31).

[569]While the training set example included only male samples, predictions were made on both female samples and male trisomy samples, for which the fetal fraction can be estimated independently using the trisomy chromosome representation. Fetal fraction estimation of male and female samples showed no difference in total distribution (Figure 32). This demonstrates that BFF is not systematically biased towards estimating fetal fraction over one gender compared to the other.

Example 8: Examples of modalities

[570]The following examples illustrate certain embodiments and do not limit the technology.

[571]A1. A method for estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising: (a) obtaining counts of sequence reads mapped to portions of a reference genome, the sequence reads of which are circulating cell-free nucleic acid reads from a test specimen from a pregnant woman; (b) weight, using a microprocessor, (i) counts of sequence reads mapped to each chunk, or (ii) other chunk-specific parameter, to a specific fraction of the fetal nucleic acid chunk according to a weighting factor independently associated with each portion, thereby providing estimates of portion-specific fetal fraction according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i ) a fraction of fetal nucleic acid for each of the various samples, and (ii) counts of fragments (reads) of sequences mapped to each portion, or other portion-specific parameter, for the various samples; and (c) estimating a fetal nucleic acid fraction for the test sample based on the portion-specific fetal fraction estimates.

[572]A2. The method according to claim A1, wherein weighting factors are associated with moieties in a plurality of moieties on all autosomes and X and Y chromosomes.

[573]A2.1. The method of claim A1, wherein the weighting factors are associated with portions in a plurality of portions that do not include portions on the Y chromosome.

[574]A3. The method of claim A2.1, wherein weighting factors are associated with portions in a plurality of portions that do not include portions on X and Y chromosomes.

[575]A4. The method of claim A2, wherein the weighting factors are associated with moieties in a plurality of moieties that include moieties on autosomes or a subset thereof.

[576]A5. The method of claim A3 or A4, wherein the weighting factors are associated with moieties in a plurality of moieties that do not include moieties on chromosomes 13, 18 and 21.

[577]A6. The method according to any one of claims A1 to A5, wherein the counts in (b)(i) or (b)(ii) are normalized counts.

[578]A7. The method of claim A6, wherein the normalized counts have reduced guanine-cytosine (GC) bias relative to the raw counts.

[579]A8. The method of claim A6 or A7, wherein the normalized counts are a product of normalization by portions (bins), normalization by GC content, linear least squares regression, nonlinear least squares regression, LOESS, GC LOESS, LOWESS, PERUN, GC-normalization and repeated masking (GCRM), conditional quantile normalization (cQn), or a combination thereof.

[580]A9. The method according to any one of claims A1 to A8, wherein the fetal nucleic acid fraction estimate for the test sample comprises the average or sum of the portion-specific fetal fraction estimates.

[581]A10. The method according to any one of claims A1 to A9, wherein the portion-specific parameter is a portion-specific parameter or is one of two or more portion-specific parameters.

[582]A11. The method according to any one of claims A1 to A10, wherein the specific parameter of the portion is chosen from genomic coverage, an amount of fragments (reads) having a length smaller than a selected fragment length, mappability, sensitivity to DNasel, methylation status, acetylation, histone distribution and chromatin structure.

[583]A12. The method according to any one of claims A1 to A10, wherein the portion specific parameter is guanine cytosine (GC) content.

[584]A13. The method according to any one of claims A1 to A10, wherein the portion specific parameter is not guanine cytosine (GC) content.

[585]A14. The method according to claim A11, wherein the amount of fragments (reads) having a length smaller than a selected fragment length is determined according to an X to Y ratio, where X is the amount derived from circulating cell-free (CCF) fragments having a length less than a first fragment length selected, and Y is the amount of fragments (reads) derived from CCF fragments having a length less than a second fragment length selected.

[586]A15. The method of claim A14, wherein the first selected fragment length is about 140 to about 160 bases and the second selected fragment length is about 500 to about 700 bases.

[587]A16. The method of claim A15, wherein the length of the first selected fragment is about 150 bases in length and the second selected fragment is about 600 bases.

[588]A17. The method according to any one of claims A14 to A16, wherein the weighting factor for each portion is related to the average proportion for the portion for the various samples.

[589]A18. The method according to any one of claims A1 to A16, wherein the weighting factor for each portion is proportional to the average amount of fragments (reads) from fetal CCF nucleic acid fragments mapped to the portion for the various samples .

[590]A19. The method according to any one of claims A1 to A18, wherein the portions are chosen from discrete genomic portions (bins), genomic portions (bins) having continuous sequences of predetermined length, portion (bin) of variable length, point-based views of a graded coverage map, and a combination thereof.

[591]A20. The method according to any one of claims A1 to A19, wherein the various samples are from subjects having an euploid foetus.

[592]A21. The method according to any one of claims A1 to A19, wherein the various samples are from subjects having a fetus with trisomy.

[593]A22. The method according to any one of claims A1 to A19, wherein the various samples are from subjects having a euploid fetus and from subjects having a fetus with trisomy.

[594]A23. The method according to any one of claims A1 to A22, wherein the various samples are from subjects having a male fetus.

[595]A24. The method of claim A23, wherein the fetal nucleic acid fraction is determined according to a Y chromosome assay.

[596]A25. The method of any one of claims A1 to A24, wherein the counts in about 1,500 servings to about 200,000 servings are adjusted.

[597]A25.1. The method of claim A25, wherein each portion is about 10 contiguous kilobases to about 75 contiguous kilobases of the reference genome.

[598]A26. The method according to any one of claims A1 to A25.1, wherein about 75% or more of the weighting factors are greater than zero.

[599]A26.1. The method of claim A26, wherein about 85% or more of the weighting factors are greater than zero.

[600]A26.2. The method of claim A26.1, wherein about 95% or more of the weighting factors are greater than zero.

[601]A27. The method according to any one of claims A1 to A26.2, wherein the width of a distribution of weighting factors is dependent on the amount of fragments (reads) of fetal CCF nucleic acid fragments.

[602]A28. The method according to any one of claims A1 to A27, wherein the distribution of the weighting factors is substantially symmetrical.

[603]A28.1. The method according to any one of claims A1 to A27, wherein the distribution of the weighting factors is substantially normal.

[604]A29. The method according to any one of claims A1 to A28.1, the weighting factors are estimated coefficients from the fitted ratios.

[605]A30. The method according to any one of claims A1 to A29 comprising estimating coefficients from the relationship for each portion between (i) the fetal nucleic acid fraction for each of several samples, and (ii) fragment counts (reads) of sequences mapped to each chunk, or other chunk-specific parameter, for the various samples.

[606]A31. The method according to claim A29 or A30, wherein each of the fitted relationships is a regression model and the weighting factors are, or are based on, regression coefficients from the fitted relationships.

[607]A32. The method according to claim A31, wherein the regression model is chosen from a linear regression model, simple regression model, ordinary least squares regression model, multiple regression model, general multiple regression model, polynomial regression model, general linear model, generalized linear model, discrete choice regression model, logistic regression model, the multinomial logit model, mixed logit model, probit model, multinomial probit model, ordered logit model, ordered probit model, model Poisson model, multivariate response regression model, multilevel model, fixed effects model, random effects model, mixed model, nonlinear regression model, nonparametric model, semiparametric model, robust model, quantile model, isotonic model , principal components model, minor angle model, local model, segmented model, and errors in variables model.

[608]A33. The method of claim A29 or A30, wherein each of the fitted relationships is not a regression model.

[609]A34. The method according to claim A33, wherein each of the fitted relations is chosen from a decision tree model, support vector machine model and neural network model.

[610]A35. The method according to any one of claims A1 to A34, wherein the fitted relationships are fitted by least squares estimation, common least squares, linear, partial, total, generalized, weighted, nonlinear, interactively reweighted regression, ridge regression, minus absolute deviations, Bayesian, multivariate Bayesian, reduced rank, LASSO, elastic network estimator and combination of these.

[611]A36. The method of any one of claims A1 to A35 wherein it comprises, prior to (a), determining sequence reads by sequencing circulating cell-free nucleic acid from a test subject.

[612]A37. The method according to claim A36, comprising, prior to (a), mapping sequence reads to portions of the reference genome.

[613]A38. The method of claim A36 or A37 wherein it comprises, prior to (a), isolating circulating cell-free nucleic acid from the test sample.

[614]A39. The method of claim A38 wherein it comprises, prior to (a), isolating the test sample from the test subject.

[615]A40. The method according to any one of claims A1 to 39 wherein it comprises determining the presence or absence of a fetal chromosome aneuploidy for the test sample based on the estimated fetal nucleic acid fraction.

[616]A41. The method of claim A40 wherein the fetal chromosome aneuploidy is a trisomy.

[617]A42. The method according to claim A41 wherein the trisomy is chosen from a trisomy of chromosome 21, chromosome 18, chromosome 13 or a combination thereof.

[618]A43. The method of claim A41 or A42 wherein the presence or absence of trisomy is determined with a sensitivity of 95% or greater, or a specificity of 95% or greater, or a sensitivity of 95% or greater and a specificity of 95% or greater.

[619]A44. A system comprising one or more microprocessors and memory, wherein the memory comprises instructions executable by one or more microprocessors, and the memory comprising nucleotide sequence reads mapped to portions of a reference genome, whose reads ) of sequences are fragments (reads) of circulating cell-free nucleic acid from a test sample of a pregnant woman, and which instructions executable by one or more microprocessors are configured to: (a) weight, using a microprocessor, ( i) the fragment counts (reads) of sequences mapped to each portion, or (ii) another portion-specific parameter, to a specific fraction of the portion of fetal nucleic acid according to a weighting factor independently associated with each portion providing, thus, portion-specific fetal fraction estimates according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of the various samples, and (ii) counts of fragments (reads) of sequences mapped to each portion, or other portion-specific parameter, for the various samples; and (b) estimating a fetal nucleic acid fraction for the test sample based on the portion-specific fetal fraction estimates.

[620]A45. Machine comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors, and which memory comprises nucleotide sequence reads mapped to portions of a reference genome, which sequence reads are circulating cell-free nucleic acid reads from a test sample of a pregnant woman, and which instructions executable by one or more microprocessors are configured to: (a) weigh, using a microprocessor, (i) the counts of sequence reads mapped to each chunk, or (ii) another chunk-specific parameter, to a specific fraction of the fetal nucleic acid chunk according to a weighting factor independently associated with each chunk, thereby providing portion-specific fetal fraction estimates according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i) a fraction of fetal nucleic acid for each of several samples , and (ii) counts of fragments (reads) of sequences mapped to each portion, or other portion-specific parameter, for the various samples; and (b) estimating a fetal nucleic acid fraction for the test sample based on the portion-specific fetal fraction estimates.

[621]A46. A non-transient computer-readable storage medium with an executable program stored thereon, where the program instructs a microprocessor to perform the following: (a) access fragments (reads) of nucleotide sequences mapped to portions of a reference genome, which sequence reads are circulating cell-free nucleic acid reads from a test sample from a pregnant woman; (b) weight, using a microprocessor, (i) the sequence reads counts mapped to each chunk, or (ii) other chunk-specific parameter, to a specific fraction of the fetal nucleic acid chunk according to a weighting factor independently associated with each portion, thereby providing estimates of portion-specific fetal fraction according to the weighting factors, where each of the weighting factors was determined from a ratio adjusted for each portion between (i ) a fraction of fetal nucleic acid for each of the various samples, and (ii) counts of fragments (reads) of sequences mapped to each portion, or other portion-specific parameter, for the various samples; and (c) estimating the fetal nucleic acid fraction for the test sample based on the portion-specific fetal fraction estimates.

[622]B1. A method for estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising: (a) obtaining counts of sequence reads mapped to portions of a reference genome whose sequence reads are circulating cell-free nucleic acid fragments (reads) from a test sample from a pregnant woman; (b)(i) adjust, using a microprocessor, the sequence reads counts mapped to each chunk according to an independently assigned weighting factor for each chunk, thereby providing the adjusted counts for the chunks, or (b)(ii) selecting, using a microprocessor, a subset of portions, thereby providing a subset of counts, where the adjustment in (b)(i) or the selection or in (b)(ii) is in accordance with according to the portions to which an increasing amount of fetal nucleic acid reads are mapped; and (c) estimating a fetal nucleic acid fraction for the test sample based on the adjusted counts or the subset of counts.

[623]B2. The method according to claim B1, wherein the portions to which an increased amount of fetal nucleic acid fragments (reads) are mapped are determined according to an X to Y ratio, where X is the amount of fragments (reads) derived from circulating cell-free (CCF) fragments having a length less than one length of the first selected fragment, and Y is the amount of fragments (reads) derived from CCF fragments having a length less than one second fragment of selected length.

[624]B3. The method according to claim B2, wherein the ratio is an average ratio of several samples.

[625]B4. Method according to claim B3, wherein the weighting factor is determined, or portions are selected, according to a portion having an average proportion greater than the average proportion calculated for the portions.

[626]B5. The method according to any one of claims B2 to B4 wherein the first selected fragment length is about 140 to about 160 bases and the second selected fragment length is about 500 to about 700 bases.

[627]B6. The method of claim B5, wherein the length of the first selected fragment is about 150 bases and the second length of the selected fragment is about 600 bases.

[628]B7. A system comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors, and which memory comprises fragments (reads) of nucleotide sequences mapped to portions of a reference genome, which fragments (reads) of sequences are circulating cell-free nucleic acid fragments (reads) of a test sample from a pregnant woman, and which instructions executable by one or more microprocessors are configured to: (a)(i) adjust, using a microprocessor, the counts of sequence reads mapped to each chunk according to an independently assigned weighting factor for each chunk, thereby providing adjusted counts for the chunks, or (a)(ii) selecting, using a microprocessor, a subset of portions thus providing a subset of counts, where the adjustment in (b)(i) or the selection in (b)(ii) is in accordance with the portions for which an increase in the amount of fragments (reads) of fetal nucleic acid are mapped; and (b) estimating a fetal nucleic acid fraction for the test sample based on the adjusted counts or the subset of counts.

[629]B8. A machine comprising one or more microprocessors and memory, which memory comprises instructions executable by one or more microprocessors, and which memory comprises fragments (reads) of nucleotide sequences mapped to portions of a reference genome, which fragments (reads) of sequences are circulating cell-free nucleic acid fragments (reads) of a test sample from a pregnant woman, and which instructions executable by one or more microprocessors are configured to: (a)(i) adjust, using a microprocessor, the counts of sequence reads mapped to each chunk according to an independently assigned weighting factor for each chunk, thereby providing adjusted counts for the chunks, or (a)(ii) selecting, using a microprocessor, a subset of portions thus providing a subset of counts, where the adjustment in (b)(i) or the selection in (b)(ii) is in accordance with the portions for which an increase in the amount of fragments (reads) of fetal nucleic acid are mapped; and (c) estimating a fetal nucleic acid fraction for the test sample based on the adjusted counts or the subset of counts.

[630]B9. A non-transient computer-readable storage medium with an executable program stored thereon, where the program instructs a microprocessor to perform the following: (a) access fragments (reads) of nucleotide sequences mapped to portions of a reference genome, which sequence reads are circulating cell-free nucleic acid reads from a test sample from a pregnant woman; (b)(i) adjust, using a microprocessor, the sequence reads counts mapped to each chunk according to an independently assigned weighting factor for each chunk, thereby providing adjusted counts for the chunks, or ( b)(ii) selecting, using a microprocessor, a subset of portions, thereby providing a subset of counts, where the adjustment in (b)(i) or the selection in (b)(ii) conforms to the portions to which an increased amount of fetal nucleic acid reads are mapped; and (c) estimating a fetal nucleic acid fraction for the test sample based on the adjusted counts or the subset of counts.

[631]C1. A method for increasing the accuracy of estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising: obtaining counts of fragments (reads) of sequences mapped to portions of a reference genome, which fragments (reads) of sequences are circulating cell-free nucleic acid reads from a test sample from a pregnant woman; wherein at least a subset of the counts obtained are derived from a region of the genome that contributes a greater number of counts derived from fetal nucleic acid relative to total counts of the region than fetal nucleic acid counts relative to total counts of another region of the genome.

[632]C2. The method of claim C1, comprising: adjusting, using a microprocessor, the sequence reads counts mapped to each portion according to an independently assigned weighting factor for each portion, thereby providing adjusted counts for the portions, or selecting, using a microprocessor, a subset of portions, thereby providing a subset of counts; and estimating a fetal nucleic acid fraction for the test sample based on the adjusted counts or the subset of counts.

[633]C3. Method according to claim C1 or C2, in which the region of the genome that contributes a greater number of counts derived from fetal nucleic acid is determined according to a ratio of X to Y, where X is the amount of fragments (reads) derived from circulating cell free (CCF) fragments having a length less than a length of the first selected fragment, and Y is the amount of fragments (reads) derived from CCF fragments having a length less than a second fragment of selected length.

[634]C4. Method according to claim C3, wherein the ratio is an average ratio for several samples.

[635]C5. Method according to claim C4, wherein the weighting factor is determined, or portions are selected, according to a portion having an average proportion greater than the average proportion calculated for the portions.

[636]C6. The method of any one of claims C3 to C5, wherein the first selected fragment length is about 140 to about 160 bases and the second selected fragment length is about 500 to about 700 bases.

[637]C7. The method of claim C6, wherein the length of the first selected fragment is about 150 bases and the second length of the selected fragment is about 600 bases.

[638]The entirety of each patent, patent application, publication and document referenced herein is hereby incorporated by reference. Citation of patents, patent applications, publications and documents above is not an admission that any of the foregoing is pertinent prior art, nor does it constitute any admission to the contents or date of these publications or documents.

[639]Modifications can be made to the foregoing without departing from the basic aspects of the technology. Although the technology has been described in detail with substantial reference to one or more specific embodiments, those skilled in the art will recognize that changes may be made to the embodiments specifically described in this application, still such modifications and improvements are within the scope and spirit of the technology.

[640] The technology illustratively described herein may be suitably practiced in the absence of any element(s) not specifically disclosed herein. Thus, for example, in each case here any of the terms "comprising", "consisting essentially of" and "consisting of" may be replaced by any of the other two terms. The terms and expressions that have been employed are used as terms of description and not of limitation, and the use of such terms and expressions does not exclude any equivalents of the shown and described features or segments thereof, and various modifications are possible within the scope of the claimed technology . The term "a" or "an" may refer to one or a plurality of the elements they modify (e.g. "a reactant" may mean one or more reactants) unless it is contextually clear any one of the elements or more than one of the elements is described. The term "about" as used herein refers to a value within 10% of the underlying parameter (i.e., plus or minus 10%), and the use of the term "about" at the beginning of a series of values modifies each of the values (that is, "about 1, 2, and 3" refers to about 1, about 2, and about 3). For example, a "weight of about 100 grams" can include weights between 90 grams and 110 grams. Furthermore, when a list of values is described here (e.g. about 50%, 60%, 70%, 80%, 85% or 86%) the list includes all intermediate and fractional values thereof (e.g. 54%, 85.4%). Accordingly, it is to be understood that, although the present technology has been specifically disclosed by the representative embodiments and optional features, modification and variation of the concepts disclosed herein may be invoked by those skilled in the art, and such modifications and variations are considered within the scope of this technology.

[641]Certain embodiments of the technology are set out in the claim(s) that follow(s).

Claims (9)

1. Method for estimating a fraction of fetal nucleic acid in a test sample from a pregnant woman, comprising: (a) obtaining sequence fragment counts mapped to portions of a reference genome, whose sequence fragments (reads) are circulating cell-free nucleic acid fragments of a test sample from a pregnant woman, and wherein such portions are: (i) defined on the basis of a reference genome partition, wherein the reference genome is partitioned accordingly with one or more factors selected from size, guanine-cytosine (GC) content, contiguous regions, contiguous regions of arbitrarily defined size, and a combination thereof; and (ii) selected from discrete genomic bins, genomic bins having sequential sequences of predetermined length, varying size bins, point-based views of a flattened coverage map, and a combination thereof; (b) selecting a subset of portions, thus providing a subset of counts, where: (i) the selection is according to the portions to which a greater number of fragments (reads) of fetal nucleic acid are mapped; and (ii) the portions to which a greater number of fragments (reads) of fetal nucleic acid are mapped are determined according to a ratio of X to Y, where X is the amount of fragments (reads) derived from exempt fragments (CCF) having a length less than a length of the first selected fragment, where the length of the first selected fragment is 140 to 160 bases, and Y is the number of fragments (reads) derived from CCF fragments having a length less than a second selected fragment length, wherein the length of the second selected fragment is 500 to 700 bases; and (c) estimating a fetal nucleic acid fraction for the test sample based on the subset of counts. 2. Method according to claim 1, characterized in that the proportion is an average proportion for several samples. 3. Method according to claim 2, characterized in that the portions are selected according to a portion that has an average proportion greater than the average proportion calculated for the portions. 4. Method according to any one of claims 1 to 3, characterized in that the length of the first selected fragment is 145 to 155 bases and the length of the second selected fragment is 550 to 650 bases 5. Method according to any one of claims 1 to 4, characterized in that the length of the first selected fragment is 145 bases, 150 bases or 155 bases, and the length of the second selected fragment is 550 bases, 600 bases or 650 bases. 6. Method according to any one of claims 1 to 5, characterized in that the counts are normalized counts. 7. Method according to claim 6, characterized in that the counts are normalized according to the guanine-cytosine (GC) content. 8. Method according to any one of claims 1 to 7, characterized in that the subset of portions are portions of one or more autosomes. 9. Method according to any one of claims 1 to 8, characterized in that the subset of the portions are portions of one or more euploid autosomes.