WO2016112539A1 - Method and device for determining fetal nucleic acid content - Google Patents

Abstract

Disclosed in the present invention is a device for constructing distribution area of different combined genotypes of SNP site, comprising: construction unit of first area and second area for constructing the first area and the second area, wherein the first area is the distribution area of the first combined genotype of SNP site, and the second area is the distribution area of the second combined genotype of SNP site; construction unit of third area and fourth area for constructing the third area and the fourth area from the second area, wherein the third area is the distribution area of the first predetermined ratio AAaa SNP site in the second combined genotype of SNP site, and the fourth area is the distribution area of AAab SNP site in the second combined genotype of SNP site. Also disclosed in the present invention are a method for constructing distribution area of different combined genotypes of SNP site, a method and device for determining the fetal nucleic acid content from a pregnant woman sample.

Description

Method and apparatus for determining fetal nucleic acid content Technical field

The present invention relates to the field of biological information. Specifically, the present invention relates to a device for constructing a distribution region of different combined genotype SNP sites, a method for constructing a distribution region of different combined genotype SNP sites, and a method for distinguishing different combinations. A method of genotypic SNP locus, a method of determining fetal nucleic acid content in a pregnant sample, a device for determining fetal nucleic acid content in a pregnant sample, and a computer readable medium.

Background technique

Since the discovery of fetal free DNA in maternal plasma, prenatal testing techniques have undergone major innovations. Nowadays, with the continuous reduction of the price of second-generation sequencing technology and technological innovation, the development of non-invasive prenatal testing is rapidly applied. For example, the diagnosis of genetic diseases such as prenatal hemophilia, gender confusion and monogenic diseases. In these diagnoses, fetal concentration is an important parameter. In addition, abnormal fetal concentrations can also be used to aid in predicting some disease risks, such as higher levels of fetal concentration may be associated with preterm birth, and lower levels of fetal concentration may be used to aid in the identification of moderate or severe eclampsia.

There are several methods for detecting fetal DNA concentration in pregnant women's plasma, such as directly counting the proportion of Y chromosome and autosome in pregnant women's plasma in free DNA. However, when pregnant women are pregnant, this method is not available. Row. Studies have calculated fetal concentrations by differences in epigenetic markers such as methylation and unmethylation in the maternal and fetal genomes, but this method is affected by bisulfite conversion or methylation restriction enzyme digestion, often with precision. not tall. Other studies are based on next-generation sequencing technology to analyze differences in fetal and maternal genomes. However, in some non-invasive prenatal tests, fetal genomic information cannot be obtained in advance. Further research is based on the genomic information of fathers and mothers to identify fetal-specific genetic loci, but in many cases it is impossible to obtain the fetal father's genomic information, and additional analysis of the parent's genomic information requires additional costs.

Summary of the invention

According to an aspect of the present invention, the present invention provides a device for constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a first source nucleic acid and a gene in a second source nucleic acid a combination of types, the apparatus comprising: a first region-second region building unit for constructing a first region and a second region, the first region being a distribution region of the first combined genotype SNP site, and the second region is a The distribution region of the two combined genotype SNP sites, the first region and the second region are divided based on the difference between the first relationship and the second relationship, and the first relationship is the second highest base of the first combined genotype SNP site The relationship between the depth and the depth of the highest base, the second relationship is the relationship between the depth of the second highest base of the second combined genotype SNP and the depth of the highest base, the first combination The genotypes are ABaa and ABab, the second combined genotypes are AAaa and AAab; the third region-fourth region building unit is configured to construct a third region and a fourth region from the second region, and the third region is the second combination Genotype SNP locus Distribution area AAaa SNP site of a first predetermined ratio, the second region is a combination of the fourth group The distribution area of the AAab SNP locus in the type-dependent SNP locus, the third region and the fourth region are divided based on the difference between the third relationship and the fourth relationship, and the third relationship is in the second combined genotype SNP locus. The relationship between the depth of the next highest base of the first predetermined ratio of AAaa SNP sites and the depth of the highest base, and the fourth relationship is the number of AAab SNP sites in the second combined genotype SNP locus The relationship between the depth of the high base and the depth of the highest base; wherein AA and AB respectively represent homozygous and heterozygous SNP sites from the first source nucleic acid, and aa and ab respectively represent homozygous and heterozygous The same SNP site from the second source nucleic acid, defining A or a represents the highest base of the SNP site, and B or b represents the next highest base of the same SNP site. Since the SNP is generally considered to be dimorphic, it is a di-allelic, consisting of two bases, the highest base and the second highest base, respectively. Here, A and B respectively represent a SNP locus in a source nucleic acid. The highest base and the second highest base in the corresponding one, and a and b respectively represent the highest base and the next highest base of the same SNP site in the nucleic acid of another source. In this aspect of the invention, the device is based on four possible combinations of genotype formation in the two source nucleic acids by the same SNP site, assuming that the first source nucleic acid is greater than the second source nucleic acid content, and if in the same sequence In the measurement results, the distribution regions of the four different combined genotype SNP sites were constructed by the difference in the relationship between the sub-high base depth and the highest base depth of the four combined SNP sites. The distribution region constructed using the device can be used to determine the combined genotype of SNP sites in the data to be detected, and/or to distinguish different combined genotype SNP sites in the data to be detected and to obtain a certain combination genotype thereof. Data information of SNP sites.

According to an aspect of the present invention, the present invention provides a device for constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a first source nucleic acid and a gene in a second source nucleic acid a combination of types, the apparatus comprising: a first region-second region building unit for constructing a first region and a second region, the first region being a distribution region of the first combined genotype SNP site, and the second region is a The distribution region of the two combined genotype SNP sites, the first region and the second region are divided based on the difference between the first relationship and the second relationship, and the first relationship is the second highest base of the first combined genotype SNP site The relationship between the depth and the depth of the highest base, the second relationship is the relationship between the depth of the second highest base of the second combined genotype SNP and the depth of the highest base, the first combination The genotypes are ABaa and ABab, the second combined genotypes are AAaa and AAab; the third region-fourth region building unit is configured to construct a third region and a fourth region from the second region, and the third region is the second combination Genotype SNP locus a distribution area of the first predetermined ratio of AAaa SNP sites, the fourth region is a distribution region of AAab SNP sites in the second combined genotype SNP site, and the third region and the fourth region are based on the third relationship and the third region The relationship between the four relationships is divided, and the third relationship is the relationship between the depth of the second highest base of the AAaa SNP site in the first predetermined proportion of the second combined genotype SNP site and the depth of the highest base. a fourth relationship is a relationship between a depth of a sub-high base of the AAab SNP site in the second combined genotype SNP site and a depth of the highest base thereof; and a closed fourth region building unit for A closed fourth region is constructed in the fourth region, and the closed fourth region is a distribution region of a second predetermined ratio of AAab SNP sites in the second combined genotype SNP site, and the closed fourth region is based on the AAab SNP site. Both the high base and the highest base depth follow a normal distribution, and a second predetermined ratio is set, constructed from the fourth region; wherein AA and AB respectively represent homozygous and heterozygous nucleic acids from the first source SNP locus, aa and ab Representing homozygous and heterozygous respectively from the second The same SNP site of the source nucleic acid, defining A or a represents the highest base of the SNP site, and B or b represents the next highest base of the same SNP site.

According to an aspect of the present invention, the present invention provides a method for constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a first source nucleic acid and a gene in a second source nucleic acid a combination of types, the method comprising: constructing a first region and a second region based on a difference between the first relationship and the second relationship, the first region is a distribution region of the first combined genotype SNP site, and the second region is a second region The distribution region of the combined genotype SNP locus, the first relationship is the relationship between the depth of the second highest base of the first combined genotype SNP locus and the depth of the highest base, and the second relationship is the second combined gene The relationship between the depth of the next highest base of the SNP site and the depth of the highest base, the first combined genotypes are ABaa and ABab, and the second combined genotype is AAaa and AAab; based on the third relationship and the third The difference between the four relationships is to construct a third region and a fourth region from the second region, and the third region is a distribution region of the first predetermined ratio of AAaa SNP sites in the second combined genotype SNP site, and the fourth region is Second combination base The distribution of the AAab SNP locus in the type-dependent SNP locus, the third relationship being the depth of the sub-high base of the first predetermined proportion of the AAaa SNP locus in the second combined genotype SNP locus and its highest base The relationship between the depths of the fourth relationship is the relationship between the depth of the sub-high base of the AAab SNP site in the second combined genotype SNP site and the depth of the highest base; wherein AA and AB denotes homozygous and heterozygous SNP sites from the first source nucleic acid, respectively, and aa and ab represent homozygous and heterozygous identical SNP sites from the second source nucleic acid, respectively, and A or a represents the SNP site. The highest base, B or b, represents the next highest base of the same SNP site.

According to another aspect of the present invention, the present invention provides a method for distinguishing different combinations of genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a first source nucleic acid and in a second source nucleic acid. a combination of genotypes, the method comprising: sequence determining at least a portion of the nucleic acid in the mixed nucleic acid sample, obtaining sequencing data, the sequencing data consisting of a plurality of reads, the mixed nucleic acid sample comprising the first source nucleic acid and the second source nucleic acid; The sequencing data is aligned with the reference sequence to obtain a comparison result; based on the comparison result, the SNP site is identified and the distribution region where the SNP site is located is determined, and the distribution region is a different combination according to one aspect of the present invention. A method for constructing a distribution region of a genotype SNP locus; determining a combined genotype of the SNP locus based on a distribution region in which the SNP locus is located.

Further, according to an aspect of the present invention, the proportion of the second source nucleic acid in the mixed nucleic acid is estimated based on the information of the SNP site falling into the fourth region. Since the combined genotype of the SNP locus falling into the fourth region or falling into the closed fourth region is AAab, the sub-high base is only from the second source nucleic acid, and the number of reads supported by the second highest base is overwriting the position. The proportion of the total number of reads in the point, the concentration of the second source nucleic acid can be estimated, which can be expressed as the second source nucleic acid concentration = 2 * second highest base depth / (highest base depth + second highest base depth), The next highest base and the highest base in the formula are from the same AAab SNP site. In a specific embodiment of the present invention, the second source nucleic acid concentration is estimated by using each SNP site falling into the closed fourth region to obtain a set of second source nucleic acid concentration values, and the median value is the second. Source nucleic acid concentration. For mixed nucleic acid samples containing only two sources of nucleic acid, the nucleic acid content of one of the sources is determined and the other is determined.

According to still another aspect of the present invention, the present invention provides a method for determining a fetal nucleic acid content in a pregnant woman sample, the method comprising: obtaining a sequencing result, the obtaining of the sequencing result comprising sequencing a nucleic acid of at least a portion of the sample of the pregnant woman, The sequencing result consists of multiple reads. The maternal sample contains maternal nucleic acid and fetal nucleic acid; the sequencing result is compared with the reference sequence to obtain the alignment result; based on the comparison result, the SNP locus is identified; based on the comparison result, the SNP is determined. a distribution region in which the site is located, the distribution region is constructed according to an aspect of the present invention for constructing a distribution region of different combined genotype SNP sites; based on the fourth region in the distribution region or the SNP site in the fourth region; The fetal nucleic acid content in the pregnant woman sample is determined. In the method of this aspect of the invention, the sample of the pregnant woman to be tested is a sample of the pregnant woman's body fluid, for example, from the peripheral blood of the pregnant woman, the urine of the pregnant woman, and the like.

According to an aspect of the present invention, the present invention provides a method for determining a fetal nucleic acid content in a pregnant woman sample, the method comprising: obtaining a sequencing result, the obtaining of the sequencing result comprising sequencing, sequencing, and sequencing at least a portion of the nucleic acid in the pregnant woman sample The result consists of multiple readings. The maternal sample contains maternal nucleic acid and fetal nucleic acid; the sequencing result is compared with the reference sequence to obtain the alignment result; based on the comparison result, the SNP locus is identified; based on the comparison result, the SNP position is determined. a distribution region in which the point is located, the distribution region is constructed according to an aspect of the present invention for constructing a distribution region of different combined genotype SNP sites; based on the fourth region in the distribution region or the SNP site in the closed fourth region, The fetal nucleic acid content of the pregnant woman sample; and, when the sequencing result contains less than 65X of data and/or the determined fetal nucleic acid content is less than 10%, the bias correction model is used to correct the fetal nucleic acid content, and the corrected fetal nucleic acid content is obtained. .

According to still another aspect of the present invention, the present invention provides an apparatus for determining a fetal nucleic acid content in a pregnant woman sample, comprising: a data input unit for inputting data; a data output unit for outputting data; and a storage unit for storing Data, including an executable program; a processor coupled to the data input unit, the data output unit, and the storage unit for executing an executable program, the method of performing the method comprising determining the fetal nucleic acid content in the pregnant woman sample .

According to still another aspect of the present invention, a computer readable medium for storing a program for execution by a computer, the method comprising performing any of the above methods for determining a fetal nucleic acid content in a sample of a pregnant woman is provided. One of ordinary skill in the art will appreciate that all or part of the above-described methods of determining fetal nucleic acid content in a sample of a pregnant woman can be accomplished by instructing the associated hardware while executing the procedure. The storage medium may include: a read only memory, a random access memory, a magnetic disk or an optical disk, and the like.

According to the apparatus and/or method of the present invention, the distribution regions of different combined genotype SNP loci can be obtained, so that the SNP loci of different combined genotypes can be distinguished, the combined genotype of the SNP can be determined, and the SNP position of the specific combined genotype can be obtained. The point and the SNP site information of the specific combination genotype in the mixed nucleic acid sample are used to determine the content of the nucleic acid of different sources in the mixed nucleic acid sample, including the fetal nucleic acid content in the pregnant sample, and the content of the nucleic acid from the tumor cell in the tumor circulating blood sample. Further, according to the deviation correction model of the present invention, it is possible to correct the calculation due to the fact that the amount of data acquired is small or the concentration of the target-derived nucleic acid in the mixed nucleic acid sample is relatively low, and the range of the delineated distribution region is relatively strict. The deviation of the concentration of the target source nucleic acid is obtained such that the nucleic acid content of the second source is determined.

DRAWINGS

The above and/or additional aspects and advantages of the present invention will become apparent and readily understood from

Figure 1 shows the ratio of the standard deviation to the probability of a standard normal distribution in one embodiment of the invention.

Figure 2 shows a schematic representation of a distribution region constructed in a specific embodiment of the invention.

Figure 3 shows the relationship between predicted fetal nucleic acid concentration and real fetal nucleic acid concentration in one embodiment of the invention, wherein the predicted fetal nucleic acid concentration in Figure 3a is not corrected using the bias correction model, the predicted in Figure 3b The fetal nucleic acid concentration is the predicted fetal nucleic acid concentration corrected by the bias model.

Figure 4 shows the effect of different sequencing depths on determining fetal nucleic acid content in one embodiment of the invention.

Figure 5 shows the effect of the number of sites in the AAab SNP distribution region on determining fetal nucleic acid content in one embodiment of the invention.

Figure 6 shows the difference in male fetus fetal nucleic acid concentration predicted using Y chromosome depth and the fetal fetal nucleic acid concentration predicted by the method of the present invention in one embodiment of the present invention.

detailed description

According to an embodiment of the present invention, there is provided a device for constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a nucleic acid of a first source and a nucleic acid at a second source a combination of genotypes, the apparatus comprising: a first region-second region building unit for constructing a first region and a second region, the first region being a distribution region of the first combined genotype SNP site, and second The region is a distribution region of the second combined genotype SNP site, and the first region and the second region are divided based on the difference between the first relationship and the second relationship, and the first relationship is the time of the first combined genotype SNP site The relationship between the depth of the high base and the depth of the highest base, and the second relationship is the relationship between the depth of the second highest base of the second combined genotype SNP and the depth of the highest base. The first combined genotype is ABaa and ABab, the second combined genotype is AAaa and AAab; and the third region-fourth region building unit is configured to construct a third region and a fourth region from the second region, the third region is Second combined genotype S a distribution area of a first predetermined ratio of AAaa SNP sites in the NP site, the fourth region is a distribution region of AAab SNP sites in the second combined genotype SNP site, and the third region and the fourth region are based on The difference between the third relationship and the fourth relationship is the third relationship being the depth of the second highest base of the first predetermined proportion of the AAaa SNP sites in the second combined genotype SNP site and the depth of the highest base The relationship between the fourth relationship is the relationship between the depth of the sub-high base of the AAab SNP site in the second combined genotype SNP site and the depth of the highest base; wherein AA and AB respectively represent Homozygous and heterozygous SNP sites from nucleic acids of the first source, aa and ab represent homozygous and heterozygous identical SNP sites from the second source of nucleic acid, respectively, and A or a represents the highest base of the SNP site. , B or b represents the next highest base of the same SNP site. It is generally considered that the SNP is dimorphic, that is, the second allele is composed of two bases, which are the highest base and the second highest. Base, in the same sequence measurement data, the base that obtains the most data support in a SNP site is called the highest base, and the data that supports the second most is called the second highest base, for example, covers a SNP. In the read segment obtained by sequencing of the site, the corresponding position is the same as the one base of the site, and the number of reads supported by the base is the read segment supporting the base. The most bases. Here, A and B respectively represent the highest base and the second highest base of a SNP site in one source nucleic acid, and correspondingly, a and b respectively represent the highest of the same SNP site in another source nucleic acid. Base and second highest base. The depth of the highest base and the depth of the next highest base refer to the amount of data supported by the highest base and the second highest base in the same measurement data, for example, the reads obtained by sequencing are compared. On the reference sequence, the number of reads of the SNP site on the alignment (cover) reference sequence is the depth of the site, also known as the sequencing depth, and the corresponding position in the read of the site is aligned with the bit The number of reads of the highest base of the point is the depth of the highest base, and correspondingly, the number of reads of the corresponding position in the read of the site is the same as the next highest base of the site. The depth of this high base. A read that matches the corresponding position in the read on the site to the same position as the base of the site is referred to as a read that supports the base. The device is used to construct the distribution region, and it is not necessary to obtain specific SNP site data in advance, that is, the construction of the distribution region does not depend on specific SNP site data information, including the highest base and the second highest base depth independent of the SNP site. The device is based on four possible combinations of genotype formation in the two source nucleic acids by the same SNP site, assuming that the content of the first source nucleic acid is greater than the nucleic acid content of the second source, and if in the same sequence determination results, four combinations The distribution of the four different combined genotype SNP sites was constructed by the difference in the relationship between the sub-high base depth and the highest base depth of the SNP site. The four combined genotypes are shown in Table 1. It should be noted that, in theory, when a site is homozygous in both the first source nucleic acid and the second source nucleic acid, the site is not a SNP site, but There are random errors in the process of actually obtaining SNP loci, resulting in a SNP locus with a combined genotype of AAaa. Random errors include sequencing errors. According to previous studies, SNP loci due to sequencing errors obey the binomial distribution, and according to the center Limit Theorem: When the number is large enough, the binomial distribution can be considered as a normal distribution. This embodiment considers these sites to be normally distributed.

Table 1

In a specific embodiment of the present invention, a method is provided for visualizing the distribution region, establishing a two-dimensional coordinate system, the y-axis represents the depth of the next-high base of the SNP site, and the x-axis represents the highest of the SNP site. The depth of the base, the first relationship can be expressed as x/2≥y≥x/3, the second relationship can be expressed as 0<y<x/3, and the third relationship can be expressed as 0<y<(x+y) *e+m*δ, the fourth relationship can be expressed as (x+y)*e+m*δ<y<3/x, where e is the sequencing error rate, generally e≤1%, and δ is the standard deviation. δ=((x+y)*e)^0.5,m depends on the first predetermined ratio, m is a non-negative number, and the relationship between m*δ and the first predetermined ratio is the standard deviation in the standard normal distribution The ratio of the probability to the probability. Correspondingly, the difference between the first relationship and the second relationship and the difference between the third relationship and the fourth relationship are also visualized, which are displayed as boundaries of different regions, and the difference between the first relationship and the second relationship may be expressed as y=x/ 3, The difference between the third relationship and the fourth relationship is y=(x+y)*e+m*δ. The probability density function curve of the standard normal distribution has a bell shape, and those skilled in the art can understand that the ratio of the standard deviation to the probability in the standard normal distribution is as shown in FIG. 1, such as: plus or minus one standard deviation. Between the total area of 68.26%; between plus or minus 1.96 standard deviations, including 95% of the total area; between plus and minus 2.58 standard deviations, including 99% of the total area; m is corresponding to the standard deviation The number, the first predetermined ratio is the percentage corresponding to the total area. Preferably, the first predetermined ratio is not less than 95%. In a specific embodiment of the invention, the first predetermined ratio is 99.9%, corresponding to m=3.

According to a specific embodiment of the present invention, the apparatus further comprises: a closed fourth region building unit for constructing a closed fourth region from the fourth region, the closed fourth region being the second of the second combined genotype SNP sites a distribution area of a predetermined ratio of AAab SNP sites, wherein the closed fourth region is subjected to a normal distribution based on the sub-high base and the highest base depth of the AAab SNP site, and the second is set A predetermined ratio is obtained from the construction of the fourth region. Generally, the second predetermined ratio is set to be not less than 95%. In a y-axis representing the depth of the next highest base of the SNP site, the x-axis represents the depth of the highest base of the SNP site in a two-dimensional coordinate system, y=x/3, y=(x+y)*e+ m*δ, y=D ₀ -n*δ-x and y=D ₀ +n*δ-x constitute the fourth closed region, where e is the sequencing error rate, generally e≤1%, δ Is the standard deviation, δ = ((x + y) * e) ^ 0.5, D ₀ is the average depth of the SNP site, m, n is non-negative, depending on the first predetermined ratio, m * δ and the The relationship of the first predetermined ratio is a ratio relationship of the standard deviation and the probability in the standard normal distribution, n depends on the second predetermined ratio, and the relationship between n*δ and the second predetermined ratio is in the standard normal distribution The ratio of standard deviation to probability. Preferably, the second predetermined ratio is not less than 95%. In a specific embodiment of the invention, the first predetermined ratio and the second predetermined ratio are both 99.9%, corresponding to m=n=3. The average depth of the so-called SNP site refers to the average of the depth of the SNP site, and the depth of the SNP site is the amount of support data obtained by the SNP site, for example, the reads obtained by sequencing are compared to the reference. In sequence, the number of reads of the SNP site in the reference sequence is the depth of the site, also referred to as the sequencing depth, ie the number of times the site is covered, preferably, D ₀ ≥ 100X.

By using the device in any of the above embodiments, according to the difference in the functional relationship between the sub-high base depth and the highest base depth of different combined genotype SNP sites, a distribution region for obtaining different combined genotype SNP sites can be constructed. After the division determines the respective distribution regions, the converse genotype of the SNP locus can be determined according to the distribution region in which the SNP locus falls in the data to be detected, the SNP locus of the specific combined genotype is obtained, and the mixed nucleic acid sample is utilized. The SNP site information for a particular combination genotype determines the amount of nucleic acid from a different source in the mixed nucleic acid sample, including the fetal nucleic acid content in the pregnant sample.

According to an embodiment of the present invention, there is provided a method for constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a first source nucleic acid and in a second source a combination of genotypes in a nucleic acid, the method comprising: constructing a first region and a second region based on a difference between the first relationship and the second relationship, the first region being a distribution region of the first combined genotype SNP site, and second The region is a distribution region of the second combined genotype SNP locus, and the first relationship is a relationship between the depth of the second highest base of the first combined genotype SNP locus and the depth of the highest base, and the second relationship is The relationship between the depth of the second highest base of the second combined genotype SNP locus and the depth of the highest base, the first combined genotype is ABaa And ABab, the second combined genotype is AAaa and AAab; based on the difference between the third relationship and the fourth relationship, the third region and the fourth region are constructed from the second region, and the third region is the second combined genotype SNP site a distribution area of the first predetermined ratio of AAaa SNP sites, the fourth region is a distribution region of AAab SNP sites in the second combined genotype SNP site, and the third relationship is in the second combined genotype SNP site The relationship between the depth of the next highest base of the first predetermined ratio of AAaa SNP sites and the depth of the highest base, and the fourth relationship is the number of AAab SNP sites in the second combined genotype SNP locus The relationship between the depth of the high base and the depth of the highest base; wherein AA and AB respectively represent homozygous and heterozygous SNP sites from the first source nucleic acid, and aa and ab respectively represent homozygous and heterozygous The same SNP site from the second source nucleic acid, with A or a representing the highest base of the SNP site, and B or b representing the next highest base of the same SNP site. Using this method to construct a distribution region, it is not necessary to obtain specific SNP locus data in advance, that is, the construction of the distribution region does not depend on specific SNP locus data information, including the highest base and the second highest base depth independent of the SNP locus. The method is based on four possible combinations of genotype formation in the two source nucleic acids of the same SNP site, assuming that the content of the first source nucleic acid is greater than the second source nucleic acid content, and if in the same sequence determination result, four combinations The distribution of the four different combined genotype SNP sites was constructed by the difference in the relationship between the sub-high base depth and the highest base depth of the SNP site. The foregoing description of the technical features and advantages of the apparatus for constructing the distribution regions of different combinations of genotype SNP sites in any of the embodiments of the present invention is equally applicable to the method of this embodiment of the present invention.

For example, in one embodiment of the invention, a method is provided for visualizing said distribution region, establishing a two-dimensional coordinate system, the y-axis representing the depth of the next-high base of the SNP site, and the x-axis representing the SNP site The depth of the highest base, the first relationship can be expressed as x/2 ≥ y ≥ x / 3, the second relationship can be expressed as 0 < y < x / 3, and the third relationship can be expressed as 0 < y < (x + y) *e+m*δ, the fourth relationship can be expressed as (x+y)*e+m*δ<y<3/x, where e is the sequencing error rate, generally e≤1%, δ is the standard Poor, δ=((x+y)*e)^0.5, m depends on the first predetermined ratio, m is a non-negative number, and the relationship between m*δ and the first predetermined ratio is in a standard normal distribution The ratio of standard deviation to probability. Correspondingly, the difference between the first relationship and the second relationship and the difference between the third relationship and the fourth relationship are also visualized, which are displayed as boundaries of different regions, and the difference between the first relationship and the second relationship may be expressed as y=x/ 3. The difference between the third relationship and the fourth relationship is y=(x+y)*e+m*δ. The probability density function curve of the standard normal distribution has a bell shape, and those skilled in the art can understand that the ratio of the standard deviation to the probability in the standard normal distribution is as shown in FIG. 1 , where σ refers to the standard deviation. Between positive and negative one standard deviation, including 68.26% of the total area enclosed by the curve, between plus or minus 1.96 standard deviations, including 95% of the total area, between plus and minus 2.58 standard deviations, including 99% of the total area ;m corresponds to the number of standard deviations therein, and the first predetermined ratio corresponds to the percentage of the total area therein. Preferably, the first predetermined ratio is not less than 95%. In a specific embodiment of the invention, the first predetermined ratio is 99.9%, corresponding to m=3.

According to a particular embodiment of the invention, the method further comprises: constructing a closed fourth region from the fourth region, the closed fourth region being a second predetermined ratio of AAab SNP sites in the second combined genotype SNP site The distribution region, the construction of the closed fourth region includes obeying a normal distribution based on the sub-high base and the highest base depth of the AAab SNP site, and setting the second predetermined ratio. Generally, the second predetermined ratio is set to be not less than 95%. In a y-axis representing the depth of the next highest base of the SNP site, the x-axis represents the depth of the highest base of the SNP site in a two-dimensional coordinate system, y=x/3, y=(x+y)*e+ m*δ, y=D ₀ -n*δ-x and y=D ₀ +n*δ-x constitute the fourth closed region, where e is the sequencing error rate, generally e≤1%, δ Is the standard deviation, δ = ((x + y) * e) ^ 0.5, D ₀ is the average depth of the SNP site, m, n is non-negative, depending on the first predetermined ratio, m * δ and the The relationship of the first predetermined ratio is a ratio relationship of the standard deviation and the probability in the standard normal distribution, n depends on the second predetermined ratio, and the relationship between n*δ and the second predetermined ratio is in the standard normal distribution The ratio of standard deviation to probability. Preferably, the second predetermined ratio is not less than 95%. In a specific embodiment of the invention, the first predetermined ratio and the second predetermined ratio are both 99.9%, corresponding to m=n=3. The average depth of the so-called SNP site refers to the average of the depth of the SNP site, and the depth of the SNP site is the amount of support data obtained by the SNP site, for example, the reads obtained by sequencing are compared to the reference. In sequence, the number of reads of the SNP site in the reference sequence is the depth of the site, also referred to as the sequencing depth, ie the number of times the site is covered, preferably, D ₀ ≥ 100X.

According to another embodiment of the present invention, there is provided a method of distinguishing different combinations of genotype SNP sites, wherein the combined genotype is a genotype of a SNP site in a first source nucleic acid and a genotype in a second source nucleic acid In combination, the method comprises: sequence determining at least a portion of the nucleic acid in the mixed nucleic acid sample, the sequencing data comprising a plurality of reads, the mixed nucleic acid sample comprising the first source nucleic acid and the first a nucleic acid of two origins; aligning the sequencing data with a reference sequence to obtain a comparison result; identifying a SNP site based on the comparison result; determining a distribution region where the SNP site is located, the distribution region is based on The method for constructing a distribution region of different combined genotype SNP sites in any of the above embodiments is constructed; and based on the distribution region in which the SNP site is located, the combined genotype of the SNP site is determined. The first source nucleic acid and the second source nucleic acid may be nucleic acids derived from different individuals, or may be nucleic acids of different tissues or parts of the same individual, such as nucleic acids from tumor cells and from non-tumor cells. Obtaining sequencing data includes sequencing library preparation of mixed nucleic acid samples, and sequencing of the libraries. Sequencing can be performed using existing sequencing platforms, and library preparation can be performed according to the selected sequencing platform. The optional sequencing platforms include, but are not limited to, CG (Complete Genomics) CGA, Illumina/Solexa, Life Technologies/Ion Torrent, and Roche 454. Preparation of single-ended or double-end sequencing libraries according to the selected sequencing platform. The comparison can be performed by using software such as SOAP (Short Oligonucleotide Analysis Package), BWA, etc., and the embodiment does not limit this. In the comparison process, according to the setting of the comparison parameter, for example, each reading in the sequencing data is allowed to be allowed at most. h base mismatch, h is preferably 1 or 2. If more than h bases in a reads are mismatched, it is considered that the reads cannot be compared to the reference sequence. The identification of SNP sites can be performed according to software default parameter settings using software such as SOAPsnp and GATK. The reference sequence is a known sequence, and may be any reference template in the biological category to which the target individual belongs, such as a published genome assembly sequence of the same biological category, if the mixed nucleic acid sample is from a human, its genome The reference sequence (also referred to as the reference genome) can be selected from the HG19 provided by the NCBI database. The comparison result includes the comparison of each read segment with the reference sequence, including whether the read segment can compare the reference sequence, the position of the reference sequence on the read alignment, how many reads at a certain point are aligned, and the comparison The base type of the corresponding position of the read of a certain site, and the like. Identifying SNPs based on alignment results The locus, based on the relationship between the highest base of the SNP locus and the depth of the sub-high base, determines which distribution region the SNP locus falls into, because each distribution region corresponds to the four combined genotypes, such as the distribution region. The first region is the distribution region of the combined genotype ABaa and ABab SNP sites, the third region is the distribution region of the first predetermined ratio of AAaa SNP sites, and the fourth region or the closed fourth region is the AAab SNP site. In the distribution region, by determining the distribution region in which the SNP locus falls, the combined genotype of the SNP can be determined, that is, the SNP is typed. Any description of the advantages and technical features of the distribution area in the specific embodiments of the present invention may also bring the same advantages and technical features to the method of the embodiment, and details are not described herein again.

Further, according to a specific embodiment of the present invention, the proportion of the second source nucleic acid in the mixed nucleic acid is estimated based on the information of the SNP site falling into the fourth region. Since the combined genotype of the SNP locus falling into the fourth region or falling into the closed fourth region is AAab, the sub-high base is only from the second source nucleic acid, and the number of reads supported by the second highest base is overwriting the position. The proportion of the total number of reads in the point, the concentration of the second source nucleic acid can be estimated, which can be expressed as the second source nucleic acid concentration = 2 * second highest base depth / (highest base depth + second highest base depth), The next highest base and the highest base in the formula are from the same AAab SNP site. In a specific embodiment of the present invention, the second source nucleic acid concentration is estimated by using each SNP site falling into the closed fourth region to obtain a set of second source nucleic acid concentration values, and the median value is the second. Source nucleic acid concentration.

Fetal concentrations are an important parameter for prenatal genetic testing, such as copy number variation, eclampsia, preterm birth, and genetic disease research. According to still another embodiment of the present invention, a method for determining fetal nucleic acid content in a pregnant woman sample is provided, the method comprising: obtaining a sequencing result, the obtaining of the sequencing result comprising sequencing, sequencing, and sequencing at least a portion of the nucleic acid in the pregnant woman sample The result consists of multiple readings. The maternal sample contains maternal nucleic acid and fetal nucleic acid; the sequencing result is compared with the reference sequence to obtain the alignment result; based on the comparison result, the SNP locus is identified; based on the comparison result, the SNP position is determined. The distribution area in which the point is located, the distribution area is constructed according to the method of constructing the distribution area of different combined genotype SNP sites in any of the foregoing embodiments; the SNP based on the fourth region in the distribution region or the closed fourth region A site that determines the fetal nucleic acid content in the sample of the pregnant woman. In this embodiment, the sample of the pregnant woman to be tested is a sample of a pregnant woman's body fluid, for example, from pregnant women's peripheral blood, pregnant women's urine, and the like. The free DNA of pregnant women's body fluid contains the genomic information of mother and fetus, which can be divided into four categories: mother homozygous fetus is also homozygous (AAaa), mother homozygous fetal heterozygous (AAab), mother heterozygous but fetus Homozygous (ABaa), the mother's heterozygous fetus is also heterozygous (ABab). This embodiment uses the constructed distribution region to distinguish these four categories, and then selects the SNP site of the mother homozygous fetal heterozygous (AAab) as an effective site for calculating the fetal concentration. The description of the advantages and technical features of the constructed distribution area in any of the foregoing embodiments is equally applicable to this embodiment, and details are not described herein again.

In a specific embodiment of the invention, the fetal nucleic acid content is a median of 2*y ₄ /(x ₄ +y ₄ ), and y ₄ is in the fourth region or in each of the SNP sites in the closed fourth region. The depth of the next highest base, x ₄ is the depth of the highest base in the fourth region or the corresponding respective SNP site of the closed fourth region, wherein the depth of the base is the number of supported reads for which it is obtained.

When determining the fetal nucleic acid concentration by the method of this embodiment, if the amount of data acquired is small or the concentration of fetal nucleic acid in the pregnant woman sample is too low, the delineated distribution area may appear relatively strict, and the calculated fetal is easily caused. The concentration of nucleic acid is biased. Preferably, the sequencing result contains a data volume of not less than 65X, that is, the sequencing depth is not less than 65X. In a specific embodiment of the present invention, when the sequencing result contains less than 65X of data and/or the determined fetal nucleic acid content is less than 10%, in order to more accurately determine the fetal nucleic acid concentration, the deviation correction model is used to correct and calculate The fetal nucleic acid content is obtained to obtain a corrected fetal nucleic acid content. The so-called deviation correction model is capable of correcting the calculated deviation of the fetal nucleic acid concentration due to the small amount of data or the concentration of fetal nucleic acid in the pregnant woman sample being too low and the defined distribution area being relatively strict. The deviation correction model may be established when the data amount of the pregnant woman sample to be tested is not less than 65X or the estimated fetal nucleic acid concentration is less than 10%, or may be established in advance and saved for use.

In a specific embodiment of the present invention, when the sequencing result contains less than 65X of data and/or the determined fetal nucleic acid content is less than 10%, in order to more accurately determine the fetal nucleic acid concentration, before using the deviation correction model for correction Adjusting the first predetermined ratio, increasing the fourth region or closing the fourth region, so that the SNP sites falling into the fourth region or closing the fourth region are more and theoretically closer, which is beneficial to improve the calculated fetal nucleic acid. The accuracy of the concentration. The first predetermined ratio is lowered, and the third region is narrowed to increase the range of the fourth region or the fourth region.

In a specific embodiment of the present invention, the establishment of the deviation correction model comprises: acquiring K simulation sites, K=K ₁ +K ₂ +K ₃ +K ₄ , and K ₁ is the number of combined genotype AAaa simulation sites. K ₂ is the number of mimic sites of the combined genotype AAab, K ₃ is the number of combinatorial genotype ABaa mimic sites, K ₄ is the number of combinatorial genotype ABab mimetic sites, K ₂ /K ≥ 0.5%, K ₂ ≥ 35; set different standard fetal nucleic acid content f, based on the hypothesis, using the simulated locus of the combined genotype AAab, that is, in the fourth region or the closed region of the fourth region, the corresponding fetal nucleic acid content f _{0 is} calculated; Polynomial regression is performed on the obtained plurality of sets (f, f ₀ ) to establish the deviation correction model; the hypothesis includes that the depths of the highest base and the second highest base of the combined genotype AAaa analog sites are respectively obeyed by N ( De*D, (De*D)) and N(e*D, e*D), the depth of the highest base and the next highest base of the combined genotype AAab mimic sites are respectively obeyed by N(D*(1-f) /2), D*(1-f/2)) and N(D*f/2, (D*f/2)), the depth of the highest base and the next highest base of the combined genotype ABaa analog site Obey N separately (D*(1/2-f/2), D*(1/2-f/2)) and N(D*(1/2+f/2), D*(1/2+f/2 )), the depth of the highest base and the second highest base of the combined genotype ABab mimetic site are respectively obeyed by N(D/2, D/2) and N

Where e is the sequencing error rate, e = K ₁ / K ≤ ₁ %, D is the average sequencing depth of the simulated site, f is the standard fetal nucleic acid content, 0.5% ≤ f ≤ 25%. N(μ, σ^2) represents a normal distribution with a mean (expected) of μ and a variance of σ^2, and N(μ, σ^2) is also often expressed as N(μ, σ). The deviation correction model is Normal mixed model. Based on the above assumptions, under a fixed average sequencing depth, taking a series of f, can obtain a series of f ₀ , f ₀ can be calculated by the above method, and the equations are fitted to multiple groups (f, f ₀ ). A formula suitable for correcting f ₀ at this average sequencing depth is obtained.

In a specific embodiment of the invention, an equation suitable for correcting the calculated fetal nucleic acid concentration at different sequencing depths is provided, the deviation correction model comprising: D=50X, f=10.981f ₀ ³ -8.401f ₀ ² +3.1292f ₀ -0.1883; when D=60X, f=14.449f ₀ ³ -9.757f ₀ ² +3.2212f ₀ -0.1759; when D=70X, f=18.57f ₀ ³ -11.595f ₀ ² +3.4261f ₀ -0.1745; when D=80X, f=18.693f ₀ ³ -11.293 f ₀ ² +3.279f ₀ -0.1566; when D=90X, f=20.076f ₀ ³ -11.749f ₀ ² +3.2816f ₀ -0.1494;D When =100X, f=19.126f ₀ ³ -11.025f ₀ ² +3.098f ₀ -0.1337; when D=110X, f=19.81f ₀ ³ -11.159f ₀ ² +3.0725f ₀ -0.1279; when D=120X, f=20.61f ₀ ³ -11.38f ₀ ² +3.0554f ₀ -0.1226; when D=130X, f=19.808f ₀ ³ -10.82f ₀ ² +2.9285f ₀ -0.1128; when D=140X, f=20.752f ₀ ³ -10.892f ₀ ² +2.8731f ₀ -0.1061; when D=150X, f=16.71f ₀ ³ -9.1447f ₀ ² +2.623f ₀ -0.0937; when D=160X, f=16.878f ₀ ³ -9.1543 f ₀ ² +2.6011f ₀ -0.0904; when D=170X, f=15.433f ₀ ³ -8.3874f ₀ ² +2.4715f ₀ -0.0831; when D=180X, f=17f ₀ ³ -8.9749f ₀ ² +2.5224f ₀ -0.0828; when D=190X, f=14.627f ₀ ³ -7.8187f ₀ ² +2.3464f ₀ -0.0743; when D=200X, f=13.3f ₀ ³ -7.2048f ₀ ² +2.2491f ₀ -0.0688. Equation regression can utilize existing methods. In this embodiment, since each group (f, f ₀ ) exhibits a relatively fixed difference, the fitted curve corrects the calculated value f ₀ to the corresponding theoretical value. Like f, in order to derive the above equations, these fitted polynomial equations are significant. Since the sequencing depth (average sequencing depth of the simulated sites) assumed by either equation is greater than or equal to 50X, in practice, the sequencing depth can be corrected by the same equation in the range of ±5X of the assumed sequencing depth, for example, The amount of sequencing data for the sample is 55X. When used, f=10.981f ₀ ³ -8.401f ₀ ² +3.1292f ₀ -0.1883 (D=50X); or f=14.449f ₀ ³ -9.757f ₀ ² +3.2212f ₀ -0.1759 (D=60X) can achieve offset correction for f ₀ . In another embodiment, the amount of sequencing data for the same sample to be inspected is 55X. After the foregoing explanation, those skilled in the art can understand that D=55X can also be obtained, and multiple groups (f, f ₀ ) can be obtained. Correct the equation to obtain the optimal deviation correction equation.

A person skilled in the art can understand that all or part of the steps of the various methods in the above embodiments may be completed by a program to instruct related hardware, and the program may be stored in a computer readable storage medium, and the storage medium may include: a read only memory. , random access memory, disk or CD, etc.

According to still another embodiment of the present invention, there is provided an apparatus for determining a fetal concentration in a pregnant woman sample, comprising: a data input unit for inputting data; a data output unit for outputting data; and a storage unit for storing data, The program includes an executable program; the processor is connected to the data input unit, the data output unit, and the storage unit, and is configured to execute an executable program stored in the storage unit, and the execution of the program includes completing various modes in the foregoing implementation manner. All or part of the steps of the method.

The method and device for detecting fetal nucleic acid concentration in a pregnant woman sample according to the present invention are based on the distribution regions of different combined genotype SNP sites and/or combined with a Gaussian distribution (normal distribution) mixed model to classify SNP sites. Direct detection of fetal DNA concentration based on large fragment parallel sequencing sequences of pregnant women samples eliminates the need for experimental analysis of existing methods to determine specific types of SNP sites, saving trial and analysis costs. It is assumed that the depth of each type of SNP obeys the binomial distribution, and the normal distribution is obeyed under certain conditions to construct the distribution region and/or the deviation correction model in the present invention. Using the method of the present invention for determining fetal nucleic acid content in a pregnant woman sample, it is possible to accurately detect fetal DNA concentrations as low as 2.5% with only 65X data.

The following is a detailed description of constructing different combined genotype SNP locus distribution regions according to the device/method of the present invention in combination with specific parameters, and combining the specific nucleic acid sample data to determine the combined genotype of SNP loci according to the method of the present invention, and distinguishing different combinatorial genes The type and determination of fetal nucleic acid content in pregnant women samples are described in detail. In the description of the present invention, "first", "second", "third", "Fourth" and the like are meant to be convenient or to describe, and are not to be understood as having a sequential relationship or a relative importance indication, unless otherwise stated, "multiple", "multiple groups" have the meaning of two (groups) or two. (group) or more.

Unless otherwise stated, the reagents, software, and instruments not specifically addressed in the following examples are conventionally commercially available or disclosed.

Embodiment 1

A mixed nucleic acid sample comprising two different source nucleic acids, such as a maternal plasma sample. The genomic information of the mother and the fetus is contained in the free DNA of the pregnant woman's plasma. The SNP sites can be divided into four categories: the mother homozygous fetus is also homozygous (AAaa); the mother is homozygous but the fetus is heterozygous (AAab); The mother is heterozygous but the fetus is homozygous (ABaa); the mother's heterozygous fetus is also heterozygous (ABab). The following four types are distinguished by the difference in the relationship between the highest base depth and the second highest base depth of various types of sites, and the distribution regions of various SNP sites are constructed.

The key to the division of the distribution area is to determine the difference between the various SNP loci. In the two-dimensional coordinate system in which y represents the next highest base depth and x represents the highest base depth, the difference can be expressed as a boundary line. It is necessary to determine at least 2 curves (demarcation lines) in order to delineate the distribution area belonging to only one type of SNP site. First, the distribution of the ABaa and ABab sites is divided. The fetal concentration has not been reported to exceed 50%, so the expected depth of the sub-base of ABaa is between x/2 and x/3. The expected value of the maximum depth of the second highest base is at x/2, so the distribution area of the SNP sites of ABaa and ABab can be divided by the fitting line of y=x/3. The next step is to separate the SNP sites of the AAaa and AAab types. In theory, when the mother and the fetus are homozygous, the site is not a SNP, but an AAaa-type SNP site is generated due to sequencing errors and assembly errors. According to previous studies, these SNP loci due to sequencing errors obey the binomial distribution. In addition, according to the central limit theorem: when the number is large enough, the binomial distribution can be regarded as a normal distribution, so these sites are regarded as It is a normal distribution, and according to the nature of the binomial distribution: when the number of sites is greater than 20 and the difference in depth between each is significant (p is less than 5%), the binomial distribution can be approximated as a Poisson distribution. We set the sequencing quality value to Q20 (sequencing error rate is less than 1%). Therefore, the P of these SNP sites is theoretically less than or equal to 1%, which is consistent with the requirement that the binomial distribution is approximated as a Poisson distribution. Therefore, we assume that The variance of the state distribution is equal to the expectation. The sequencing error is assumed to be 1%, and the expected (mean) is assumed to be the sequencing depth of each site multiplied by 1%, ie the expectation and the variance are both (x _i + y _i ) * 0.01, x _i and y _i represent the SNP bits, respectively The highest base depth and the second highest base depth of point i. In the normal distribution, 99.9% of the points will fall within three standard deviations from the mean, as shown in Figure 1. Therefore, it is possible to distinguish AAab from AAaa by y=(x _i +y _i )*e+3*δ, ie y=(x _i +y _i )*0.01+3*((x _i +y _i )*0.01)^0.5 Open. Finally, the left and right boundaries are delineated for the AAab type SNP site distribution area to eliminate the influence of other uncertain factors. According to the previous study, the depth distribution of the sequencing tends to Poisson distribution. According to the Poisson distribution principle, the Poisson distribution can be approximated as a normal distribution when λ>5, and the variance and mean of the Poisson distribution are λ. When the sequencing depth is 200X and the fetal DNA concentration is 5%, the sub-high base depth of AAab is 5, and the highest base depth is 195, so it can be assumed that the highest base and the second highest base of AAab are also obeyed. State distribution, and then delineate the left and right boundaries of AAab according to the principle of triple standard deviation mentioned above. The overall result of the divided distribution area is shown in Fig. 2. In the figure, uppercase letters (A/B) represent bases from the site of the mother's nucleic acid, lowercase letters (a/b) represent bases from the site of the fetal nucleic acid, and A/a represents the highest base, B/b The next highest base representing the same site, the x-axis represents the highest base depth, and the y-axis represents the next highest base depth.

Embodiment 2

To determine the fetal nucleic acid concentration in a pregnant woman's peripheral blood sample, the overall operational procedures include:

1) Extract DNA from maternal plasma,

2) Build a library for at least a part of the extracted DNA, for example, using a chip enrichment to capture a target region on the genome, and then construct a library for sequencing, the sequencing depth is 100X, and the sequencing data is obtained.

3) Compare the sequencing data to Hg19 with software SOAP2,

4) Identify SNP sites, such as using SAMTOOLS software to generate files containing SNP information,

5) According to the SNP locus in the distribution area of the AAab type SNP locus delineated in the first embodiment (hereinafter, the distribution area of the AAab type SNP locus is simply referred to as the AAab region), that is, the mother is homozygous but the fetus is selected. The SNP site of the AAab is used as an effective site to calculate the predicted fetal nucleic acid concentration. Specifically, each SNP falling within the distribution area of the AAab SNP site is expressed according to the formula

Calculate the fetal nucleic acid concentration corresponding to each locus, and finally take the median as the fetal nucleic acid concentration of the sample, and y ₄ and x ₄ are the times of a SNP locus in the distribution area of the AAab SNP locus, respectively. High base depth and highest base depth.

Embodiment 3

When the sequencing depth of the pregnant woman sample to be tested is relatively small (<65X) or the fetal concentration is relatively low (<10%), the boundary between the defined AAab and AAaa may appear too strict and may cause deviation. For example, if the number of sites falling into the AAab region is too small to estimate or because the median value of a set of fetal nucleic acid concentrations is taken as the fetal nucleic acid concentration of the sample, when the SNP site of the AAab near the x-axis is removed too much, more support will be provided. The removal of the SNP site with low fetal concentration results in a large result, as shown in Figure 2, so when the sequencing depth is lower than 65X or the fetal concentration is less than 10%, we adjust the boundary between AAab and AAaa and establish a normal model. Correct the fetal nucleic acid concentration value. In the first step, the boundary between AAab and AAaa is adjusted from the original standard deviation of three times to two standard deviations. From a statistical point of view, 97.7% of the points in the normal distribution fall within two standard deviations from the mean.

According to the description of the construction process of the first embodiment, we can consider various SNP points as a normal distribution, so a mixed normal model can be established based on the following assumptions to correct the deviation:

1) 10,000 mimetic sites were generated, in proportion to the reported putative AAaa, AAab, ABaa and ABab sites in pregnant women plasma of 7:1:1:1. AAaa has a homozygous site and a heterozygous site due to sequencing errors. The sequencing error rate e is 0.26% based on the previously reported sequencing error rate of Hiseq2000.

2) Set the standard fetal DNA concentration to 0.5% to 25%, and the interval 0.5% to take multiple standard fetal DNA concentrations.

3) According to the description of the construction process of the first embodiment, the SNP site in the plasma can be regarded as a normal distribution with a variance equal to the mean, which can be expressed as follows:

a) For AAaa loci:

x _i ~ N (D-0.0026*D, (D-0.0026*D)^0.5)

y _i ~ N (0.0026 * (D, 0.0026 * D) ^ 0.5)

b) For AAab sites:

x _i ~N(D*(1-fe/2),(D*(1-fe/2))^0.5)

y _i ~ N (D*fe/2, (D*fe/2)^0.5)

c) For the ABaa locus:

x _i ~ N (D*(1/2-fe/2), (D*(1/2-fe/2))^0.5)

y _i ~ N (D*(1/2+fe/2), (D*(1/2+fe/2))^0.5)

d) For ABab sites:

x _i ~N(D/2,(D/2)^0.5)

y _i ~N(D/2,(D/2)^0.5)

According to the above assumption, the simulated site data is generated by the R language, the range of the AAab region of the first embodiment is adjusted, and the predicted fetal nucleic acid concentration is calculated according to the method of the second embodiment according to the simulated site falling into the adjusted AAab region. A set of values of 0.5 to 25% of the standard fetal concentration and the corresponding predicted fetal concentration at a sequencing depth can be obtained, and then a fitting equation is obtained. Table 2 shows the fitted equations at different depths produced for ease of use. When obtaining a calculated fetal DNA concentration from maternal plasma, first assess whether its sequencing depth is greater than 65X or whether the fetal DNA concentration is greater than 10%, if any of them are negative or both are negative, The predicted value can be corrected by the fitting equation to obtain more accurate results.

Table 2

Sequencing depth	formula
		50X	f=10.981f 0 3 -8.401f 0 2 +3.1292f 0 -0.1883
60X	f=14.449f 0 3 -9.757f 0 2 +3.2212f 0 -0.1759
		70X	f=18.57f 0 3 -11.595f 0 2 +3.4261f 0 -0.1745
80X	f=18.693f 0 3 -11.293 f 0 2 +3.279f 0 -0.1566
		90X	f=20.076f 0 3 -11.749f 0 2 +3.2816f 0 -0.1494
100X	f=19.126f 0 3 -11.025f 0 2 +3.098f 0 -0.1337
		110X	f=19.81f 0 3 -11.159f 0 2 +3.0725f 0 -0.1279
120X	f=20.61f 0 3 -11.38f 0 2 +3.0554f 0 -0.1226
		130X	f=19.808f 0 3 -10.82f 0 2 +2.9285f 0 -0.1128
140X	f=20.752f 0 3 -10.892f 0 2 +2.8731f 0 -0.1061

150X	f=16.71f 0 3 -9.1447f 0 2 +2.623f 0 -0.0937
		160X	f=16.878f 0 3 -9.1543f 0 2 +2.6011f 0 -0.0904
170X	f=15.433f 0 3 -8.3874f 0 2 +2.4715f 0 -0.0831
		180X	f=17f 0 3 -8.9749f 0 2 +2.5224f 0 -0.0828
190X	f=14.627f 0 3 -7.8187f 0 2 +2.3464f 0 -0.0743
		200X	f=13.3f 0 3 -7.2048f 0 2 +2.2491f 0 -0.0688

Embodiment 4

The plasma sample of the pregnant woman of Example 2 was calculated to have a fetal nucleic acid content of 5.2%, which was corrected by the equation f=19.126f ₀ ³ -11.025f ₀ ² +3.098f ₀ -0.1337 in the deviation correction model of the third embodiment. The corrected fetal nucleic acid content was 2.6%.

Embodiment 5

Sequencing data of maternal and fetal nucleic acids were mixed to simulate multiple maternal plasma samples of different fetal nucleic acid contents, and the accuracy of the calculated fetal nucleic acid content was corrected using the bias correction model, and the sequencing depth (data amount) was set to 150X. The overall results are shown in Figure 3, where Figure 3a is the result before the correction, Figure 3b is the corrected result, it can be seen that when the real fetal nucleic acid content is less than 10%, the calculated fetal nucleic acid content is compared with the actual value. Obvious deviation, if the true value is about 2%, the calculated value is greater than 5%. After correction, the calculated value does not deviate significantly from the true value. It is indicated that at low fetal nucleic acid levels, the correction facilitates obtaining accurate fetal nucleic acid concentrations.

The depth of sequencing and the number of SNP loci in the AAab region are the most important factors affecting the accuracy of the calculated nucleic acid content, and the accuracy at different depths is also examined here. As shown in Fig. 3, the fetal nucleic acid concentration can be accurately obtained at a fetal nucleic acid content of about 3%, and there are relatively many effective SNP sites. The test takes the change of 40~200X accuracy at different depths. The change of accuracy is shown in Fig. 4. When the sequencing depth is 65X, the absolute deviation e ₁ (e ₁ %=|f-f0|) is less than 1%.

In order to analyze the minimum number of SNP sites required to calculate the concentration, the present invention randomly samples the SNP site to calculate the minimum number of sites required for SNPs in the AAab region and then derives the actual number of SNP sites required. As shown in Fig. 5, whether it is 4.8%, 10% or 19.8%, the relative error e ₂ (e ₂ %) between the calculated value f ₀ and the true value f when the SNP site in the AAab region exceeds 35 =

Less than 10%. The relative error is not more than 10%, the SNP position in the AAab region is at least 35, and the AAab-type SNP locus accounts for no more than 10% of all types of SNP sites in maternal plasma samples, which can accurately detect different fetuses. The minimum number of all types of SNP sites required for nucleic acid concentration is shown in Table 3.

table 3

Embodiment 6

Eighteen maternal plasma samples were used to test the feasibility of the method for determining the fetal nucleic acid content of a pregnant woman sample. Four of the 18 samples were plasma of different pregnant women of the same pregnant woman. The results are shown in Table 4, in which the specificity = the number of true AAab sites falling within the AAab region / the total number of sites within the AAab region, sensitivity = the total number of true AAab sites / AAab sites falling within the AAab region The number of true AAab sites and the total number of AAab sites can be obtained by the traditional experimental typing method. It can be seen from the data in Table 4 that the method of the present invention is accurate and feasible. For further comparative analysis, the example also uses the Y chromosome depth to calculate the 9 male fetal fetal nucleic acid concentrations in the 18 plasma samples, and compares them with the male fetal fetal nucleic acid concentration values calculated by the method of the present invention. As shown in 6, there is a strong correlation (r = 0.94; p < 0.0001). The use of the Y chromosome depth to calculate the male fetal nucleic acid concentration is a known method, and can be referred to [Struble C A, Syngelaki A, Oliphant A, et al. Fetal fraction estimate in twin pregnancies using directed cell-free DNA analysis [J]. Fetal diagnosis and therapy, 2013, 35(3): 161-165.] performed, including obtaining the depth of the SNP site of the Y chromosome in the dbSNP database in the sample, filtering out the data at the site of the female fetus, and taking the remaining The median of the depth of the lower site is multiplied by 2 (since the Y chromosome is only one) and divided by the median of the autosomal depth to obtain the fetal nucleic acid concentration calculated using the depth of the Y chromosome.

Table 4

Claims (31)

1. A device for constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a combination of a genotype of the SNP site in a first source nucleic acid and a genotype in a second source nucleic acid, the characteristics thereof Wherein the device comprises

   a first region-second region building unit for constructing a first region and a second region, the first region being a distribution region of a first combined genotype SNP site, and the second region being a second combined genotype The distribution area of the SNP locus,

   The first area and the second area are divided based on a difference between the first relationship and the second relationship,

   The first relationship is a relationship between a depth of a second highest base of the first combined genotype SNP site and a depth of the highest base,

   The second relationship is a relationship between a depth of a second highest base of the second combined genotype SNP site and a depth of the highest base,

   The first combined genotype is ABaa and ABab, and the second combined genotype is AAaa and AAab;

   a third region-fourth region building unit for constructing a third region and a fourth region from the second region, the third region being a first predetermined ratio of the second combined genotype SNP sites a distribution region of the AAaa SNP site, wherein the fourth region is a distribution region of the AAab SNP site in the second combined genotype SNP site,

   The third area and the fourth area are divided based on a difference between the third relationship and the fourth relationship,

   The third relationship is a relationship between a depth of a second highest base of the AAaa SNP site of the first predetermined ratio of the second combined genotype SNP site and a depth of the highest base thereof,

   The fourth relationship is a relationship between a depth of a sub-high base of the AAab SNP site in the second combined genotype SNP site and a depth of the highest base; wherein

   AA and AB respectively represent homozygous and heterozygous SNP sites from the first source nucleic acid, and aa and ab respectively represent homozygous and heterozygous identical SNP sites from the second source nucleic acid,

   A or a represents the highest base of the SNP site, and B or b represents the next highest base of the same SNP site.
2. The device of claim 1 further comprising

   Closing a fourth region building block for constructing a closed fourth region from the fourth region, the closed fourth region being a second predetermined ratio of AAab SNP sites in the second combined genotype SNP site Distribution area,

   The closed fourth region is based on a normal distribution based on the sub-high base and the highest base depth of the AAab SNP site, and sets the second predetermined ratio to be constructed from the fourth region of.
3. The device of claim 1 or 2, wherein said first predetermined ratio is not less than 95%.
4. The apparatus of claim 1 wherein said first relationship is x/2 ≥ y ≥ x / 3 said second relationship is 0 < y < x / 3 and said third relationship is 0 < y < (x + y) *e+m*δ, the fourth relationship is (x+y)*e+m*δ<y<3/x, the first The difference between the relationship and the second relationship is y=x/3, and the difference between the third relationship and the fourth relationship is y=(x+y)*e+m*δ, where

   y is the depth of the next highest base of the SNP site,

   x is the depth of the highest base of the SNP site,

   e is the sequencing error rate,

   δ is the standard deviation, δ=((x+y)*e)^0.5,

   m depends on the first predetermined ratio, and the relationship between m*δ and the first predetermined ratio is a ratio relationship of the standard deviation and the probability in the standard normal distribution.
5. The device of claim 4 wherein e ≤ 1%.
6. The apparatus of claim 4 wherein m = 3 when said first predetermined ratio is 99.9%.
7. The device of claim 2 wherein said second predetermined ratio is no less than 95%.
8. The apparatus of claim 2 wherein said closed fourth region is y = x / 3, y = (x + y) * e + m * δ, y = D0 - n * δ - x and y = D ₀ +n*δ-x formed by the region, wherein

   y is the depth of the next highest base of the SNP site,

   x is the depth of the highest base of the SNP site,

   e is the sequencing error rate, e ≤ 1%,

   δ is the standard deviation, δ=((x+y)*e)^0.5,

   D ₀ is the average depth of the SNP site,

   m depends on the first predetermined ratio, and the relationship between m*δ and the first predetermined ratio is a ratio relationship between the standard deviation and the probability in the standard normal distribution,

   n Depends on the second predetermined ratio, the relationship between n*δ and the second predetermined ratio is a ratio relationship of the standard deviation and the probability in the standard normal distribution.
9. The apparatus of claim 8 wherein m = n = 3 when said first predetermined ratio and said second predetermined ratio are both 99.9%.
10. The device of claim 8 wherein D ₀ ≥ 100X.
11. A method of constructing a distribution region of different combined genotype SNP sites, wherein the combined genotype is a combination of a genotype of the SNP site in a first source nucleic acid and a genotype in a second source nucleic acid, Characteristically, the method comprises

    Constructing the first area and the second area based on the difference between the first relationship and the second relationship,

    The first region is a distribution region of a first combined genotype SNP site,

    The second region is a distribution region of a second combined genotype SNP site,

    The first relationship is a relationship between a depth of a second highest base of the first combined genotype SNP site and a depth of the highest base,

    The second relationship is a relationship between a depth of a second highest base of the second combined genotype SNP site and a depth of the highest base,

    The first combined genotype is ABaa and ABab, and the second combined genotype is AAaa and AAab; and the third region and the fourth region are constructed from the second region based on the difference between the third relationship and the fourth relationship ,

    The third region is a distribution region of a first predetermined ratio of AAaa SNP sites in the second combined genotype SNP site,

    The fourth region is a distribution region of the AAab SNP site in the second combined genotype SNP site,

    The third relationship is a relationship between a depth of a second highest base of the AAaa SNP site of the first predetermined ratio of the second combined genotype SNP site and a depth of the highest base thereof,

    The fourth relationship is a relationship between a depth of a sub-high base of the AAab SNP site in the second combined genotype SNP site and a depth of the highest base; wherein

    AA and AB respectively represent homozygous and heterozygous SNP sites from the first source nucleic acid, and aa and ab respectively represent homozygous and heterozygous identical SNP sites from the second source nucleic acid,

    A or a represents the highest base of the SNP site, and B or b represents the next highest base of the same SNP site.
12. The method of claim 11 further comprising

    Constructing a closed fourth region from the fourth region, the closed fourth region being a distribution region of a second predetermined ratio of AAab SNP sites in the second combined genotype SNP site, including

    The depths of the next highest base and the highest base based on the AAab SNP site are subject to a normal distribution, and the second predetermined ratio is set to be constructed from the fourth region.
13. The method of claim 11 or 12, wherein said first predetermined ratio is not less than 95%.
14. The method of claim 11 wherein said first relationship is x/2 ≥ y ≥ x / 3, said second relationship is 0 < y < x / 3, and said third relationship is 0 < y < (x+y)*e+m*δ, the fourth relationship is (x+y)*e+m*δ<y<3/x, and the difference between the first relationship and the second relationship is y= x/3, the difference between the third relationship and the fourth relationship is y=(x+y)*e+m*δ, wherein

    y is the depth of the next highest base of the SNP site,

    x is the depth of the highest base of the SNP site,

    e is the sequencing error rate,

    δ is the standard deviation, δ=((x+y)*e)^0.5,

    m depends on the first predetermined ratio, and the relationship between m*δ and the first predetermined ratio is a ratio relationship of the standard deviation and the probability in the standard normal distribution.
15. The method of claim 14 wherein e ≤ 1%.
16. The method of claim 14 wherein m = 3 when said first predetermined ratio is 99.9%.
17. The method of claim 12 wherein said second predetermined ratio is no less than 95%.
18. The method of claim 12 wherein said closed fourth region is y = x / 3, y = (x + y) * e + m * δ, y = D _{0 -} n * δ - x and y = a region formed by D ₀ +n*δ-x, wherein

    y is the depth of the next highest base of the SNP site,

    x is the depth of the highest base of the SNP site,

    e is the sequencing error rate, e ≤ 1%,

    δ is the standard deviation, δ=((x+y)*e)^0.5,

    D ₀ is the average depth of the SNP site,

    m depends on the first predetermined ratio, and the relationship between m*δ and the first predetermined ratio is a ratio relationship between the standard deviation and the probability in the standard normal distribution,

    n Depends on the second predetermined ratio, the relationship between n*δ and the second predetermined ratio is a ratio relationship of the standard deviation and the probability in the standard normal distribution.
19. The method of claim 18 wherein m = n = 3 when said first predetermined ratio and said second predetermined ratio are both 99.9%.
20. The method of claim 18, wherein D ₀ ≥ 100X.
21. A method for distinguishing different combined genotype SNP sites, wherein the combined genotype is a combination of a genotype of the SNP site in a first source nucleic acid and a genotype in a second source nucleic acid, characterized in that The method includes,

    Performing sequence determination on at least a portion of the nucleic acid in the mixed nucleic acid sample, the sequencing data comprising a plurality of reads, the mixed nucleic acid sample comprising the first source nucleic acid and the second source nucleic acid;

    Comparing the sequencing data with a reference sequence to obtain a comparison result;

    Identifying a SNP site based on the alignment result;

    Determining, according to the comparison result, a distribution area in which the SNP site is located, the distribution area being constructed according to any one of claims 11-20;

    The combined genotype of the SNP site is determined based on the distribution region in which the SNP site is located.
22. A method for determining fetal nucleic acid content in a sample of a pregnant woman, characterized in that

    Obtaining a sequencing result, the obtaining of the sequencing result comprises performing sequence determination on at least a portion of the nucleic acid in the pregnant woman sample, the sequencing result consisting of a plurality of readings, the pregnant female sample comprising the parent nucleic acid and the fetal nucleic acid;

    Comparing the sequencing result with a reference sequence to obtain a comparison result;

    Identifying a SNP site based on the alignment result;

    Determining a distribution area in which the SNP site is located, the distribution area being constructed according to any of claims 11-20;

    The fetal nucleic acid content in the maternal sample is determined based on the fourth region in the distribution region or the SNP site in the closed fourth region.
23. The method of claim 22 wherein said maternal sample is derived from at least one of maternal peripheral blood and maternal urine.
24. The method of claim 22, wherein said sequencing result comprises an amount of data of not less than 65X.
25. The method of claim 22, wherein said fetal nucleic acid content is a median of 2*y ₄ /(x ₄ +y ₄ ),

    y ₄ is the depth of the next highest base of each SNP site in the fourth region or the closed fourth region,

    x ₄ is the depth of the highest base of each of the corresponding SNP sites in the fourth region or the closed fourth region, wherein

    The depth of the base is the number of supported reads it has obtained.
26. The method of any one of claims 22-25, further comprising: correcting said fetal nucleic acid content using a bias correction model when the sequencing result comprises less than 65X of data and/or said fetal nucleic acid content is less than 10%, A corrected fetal nucleic acid content is obtained.
27. The method of claim 26, wherein when the sequencing result comprises less than 65X of data and/or said fetal nucleic acid content is less than 10%, the first predetermined ratio is adjusted to increase prior to correction using the bias correction model The fourth area or the fourth area is closed.
28. The method of claim 26, wherein the establishing of the deviation correction model comprises

    K simulation sites were obtained, K=K ₁ +K ₂ +K ₃ +K ₄ , K ₁ is the number of combined genotype AAaa mimic sites, K ₂ is the number of mimic sites of the combined genotype AAab, K ₃ In order to combine the number of genotype ABaa mimic sites, K ₄ is the number of combinatorial genotype ABab mimic sites, K ₂ /K ≥ 0.5%, K ₂ ≥ 35;

    Setting different standard fetal nucleic acid contents f, based on the assumption, using the simulated sites in the fourth region or the closed fourth region to calculate the corresponding fetal nucleic acid content f ₀ ,

    Performing polynomial regression on the obtained plurality of sets (f, f ₀ ) to establish the deviation correction model;

    The assumptions include,

    The depths of the highest base and the second highest base of the combined genotype AAaa mimic sites are respectively obeyed by N (D-e*D, D-e*D and N(e*D, e*D),

    The depths of the highest base and the second highest base of the combined genotype AAab mimic sites are respectively obeyed by N(D*(1-f/2), D*(1-f/2)) and N(D*f/2 , (D*f/2)),

    The depths of the highest base and the second highest base of the combined genotype ABaa analog site are respectively obeyed by N(D*(1/2-f/2), D*(1/2-f/2)) and N(D *(1/2+f/2), D*(1/2+f/2)),

    The depths of the highest base and the second highest base of the combined genotype ABab mimetic site are respectively obeyed by N(D/2, D/2) and

    

    among them,

    e is the sequencing error rate, e = K ₁ / K ≤ ₁ %,

    D is the average sequencing depth of the analog site,

    f is the standard fetal nucleic acid content, 0.5% ≤ f ≤ 25%.
29. The method of claim 28 wherein said deviation correction model comprises

    When D=50X, f=10.981f ₀ ³ -8.401f ₀ ² +3.1292f ₀ -0.1883,

    When D=60X, f=14.449f ₀ ³ -9.757f ₀ ² +3.2212f ₀ -0.1759,

    When D=70X, f=18.57f ₀ ³ -11.595f ₀ ² +3.4261f ₀ -0.1745,

    When D=80X, f=18.693f ₀ ³ -11.293f ₀ ² +3.279f ₀ -0.1566,

    When D=90X, f=20.076f ₀ ³ -11.749f ₀ ² +3.2816f ₀ -0.1494,

    When D=100X, f=19.126f ₀ ³ -11.025f ₀ ² +3.098f ₀ -0.1337,

    When D=110X, f=19.81f ₀ ³ -11.159f ₀ ² +3.0725f ₀ -0.1279,

    When D=120X, f=20.61f ₀ ³ -11.38f ₀ ² +3.0554f ₀ -0.1226,

    When D=130X, f=19.808f ₀ ³ -10.82f ₀ ² +2.9285f ₀ -0.1128,

    When D=140X, f=20.752f ₀ ³ -10.892f ₀ ² +2.8731f ₀ -0.1061,

    When D=150X, f=16.71f ₀ ³ -9.1447f ₀ ² +2.623f ₀ -0.0937,

    When D=160X, f=16.878f ₀ ³ -9.1543f ₀ ² +2.6011f ₀ -0.0904,

    When D=170X, f=15.433f ₀ ³ -8.3874f ₀ ² +2.4715f ₀ -0.0831,

    When D=180X, f=17f ₀ ³ -8.9749f ₀ ² +2.5224f ₀ -0.0828,

    When D=190X, f=14.627f ₀ ³ -7.8187f ₀ ² +2.3464f ₀ -0.0743,

    When D=200X, f=13.3f ₀ ³ -7.2048f ₀ ² +2.2491f ₀ -0.0688.
30. A device for determining fetal nucleic acid content in a pregnant woman sample, characterized in that

    a data input unit for inputting data;

    a data output unit for outputting data;

    a storage unit for storing data, including an executable program;

    And a processor coupled to the data input unit, the data output unit, and the storage unit for executing the executable program, the execution of the executable program comprising completing the method of any one of claims 22-29.
31. A computer readable medium for storing a program for execution by a computer, the execution of the program comprising performing the method of any of claims 22-29.

Images