CN117737216A - A method for detecting genomic information based on restriction endonucleases - Google Patents

Abstract

The invention discloses a method for detecting genome information based on restriction enzymes. The method comprises the steps of cutting a genome of a sample by using restriction enzymes to obtain genome DNA fragments with different lengths, enriching the amplified or non-amplified DNA into long fragments, sequencing the enriched long fragment genome DNA fragments on a long length sequencing platform, and finally analyzing the sequenced data by a computer. The method of the invention obviously improves the probability of detecting two alleles at the same time, obviously reduces the allele deletion rate, can better detect heterozygous tumor mutation, and has important significance for early diagnosis and treatment of tumors.

Description

A method for detecting genomic information based on restriction endonucleases

Technical field

The invention belongs to the technical field of genome detection, and specifically relates to a method for detecting genome information based on restriction endonucleases.

Background technique

Cells are the basic building blocks of organisms. In each cell, genetic information is stored in the form of chromosomes. It is generally believed that all cells of each individual have the same genome, so genome research can be conducted at the species or individual level. However, in the following situations, people need to conduct genome exploration from the single-cell scale: (1) Cells are very precious , rare in number, such as human oocytes, embryonic cells and circulating tumor cells; (2) Different cells have their own unique genomes, for example, sperm cells of the same individual have different genomes due to meiotic homologous recombination; (3) Cell lineage tracking, the genome of single cells changes over time, and genome changes can be used to reflect the evolution of cells over time; (4) Single cell genomes are heterogeneous, such as tumors, nerves, immunity, and chimeras. Therefore, single-cell genome sequencing technology came into being.

The single-cell genome only contains about 6 pg of DNA, which is far less than the amount of DNA required for high-throughput sequencing, so uniform amplification is required before sequencing. The development of whole-genome amplification (WGA) technology enables us to amplify enough genomic DNA for sequencing in single cells to study the genetic heterogeneity of cells, including single-nucleotide variations (Single-Nucleotide). Variations, SNVs), copy number variations (Copy-NumberVariations, CNVs) and structural variations (Structural Variations, SVs). A variety of single-cell whole-genome amplification technologies based on Next-Generation Sequencing (NGS) platforms have been developed, such as Degenerate Oligonucleotide-Primed Polymerase Chain Reaction (DOP) -PCR), Multiple Displacement Amplification (MDA), Multiple Annealing and Looping-Based Amplification Cycles (MALBAC), Linear Amplification via Transposon Insertion, LIANTI), primary template-directed amplification (PTA) and complementary strand multiplexed end-tagging amplification of complementary strands (META-CS). Among them, the META-CS method published by Professor Xie Xiaoliang's group uses the complementarity of DNA to eliminate almost all false positives in SNV detection. Only mutation sites supported by both the forward and reverse strands are judged to be SNVs. This has achieved so far Highest precision.

Due to the high sequencing accuracy of the NGS platform, the above-mentioned technologies are very powerful in the detection of CNVs and SNVs, but are limited by the short read length, so their performance in the detection of SVs is poor. SVs include deletions, insertions, duplications and translocations, and are important types of mutations in many heritable diseases, such as cancer. Therefore, studying SVs at single-cell resolution is a crucial issue.

Based on the long-read sequencing platform, namely Third-Generation Sequencing (TGS) platform, our research group developed single-molecule real-time sequencing of long fragments amplified by transposon insertion. amplified through transposon insertion, SMOOTH-seq), which uses low-concentration Tn5 transposase to randomly fragment single-cell genomic DNA to achieve relatively uniform genome amplification. In addition to CNVs and SNVs, SMOOTH-seq can also effectively detect SVs. However, in diploid cells, the simultaneous coverage of both alleles is very limited, causing SMOOTH-seq to have a high false negative rate when detecting heterozygous SNPs (hetSNPs).

Allelic loss is an important problem faced by single-cell whole-genome amplification technology. When diploid cells have heterozygous mutations, if one of the two alleles cannot be amplified and detected, it will lead to allele deletion, which is the main cause of false negative SNVs. Previous single-cell whole-genome methods either relied on random primer amplification or Tn5 random fragmentation. These random interruption or amplification methods are not conducive to simultaneous capture of alleles. For example, for a pair of alleles A and B, if the genome coverage is n%, that is, there is an n% probability of capturing gene A or gene B, then the probability of capturing both alleles at the same time is n%× n%, that is, n2‰. Therefore the probability of capturing both alleles at the same time is very low.

Since the two alleles in a diploid genome usually have the same restriction endonuclease recognition site, the homologous DNA fragments produced by restriction endonuclease digestion are usually of the same length and are much smaller than those produced by Tn5 random transposition or random primer amplification. DNA fragments are more likely to be amplified simultaneously. Based on this, the present invention developed a single-cell long-read whole-genome sequencing technology Refresh-seq (Restriction fragments ligation-based genome amplification and third-generation sequencing) based on restriction endonuclease cutting and ligation strategies, which significantly improved two et al. The probability of simultaneous detection of all genes. Allele deletion causes false negatives in prenatal diagnosis. Due to allele deletion, only one allele may be detected and the other allele may not be detected. For abnormal embryos (heterozygous mutations) caused by a single gene ), if only normal allele is detected and the disease-causing mutant allele is not detected due to allele deletion, the embryo will be misjudged as a normal embryo. This method significantly reduces the allele deletion rate, can reduce such erroneous judgments, and can select healthy embryos to promote eugenics and postnatal care. In addition, tumor cells have a high mutation load, and mutations are usually heterozygous. Previous methods tend to underestimate the mutation status of tumor genomes due to allelic loss, but this method can better detect heterozygous tumors. Mutations are of great significance for the early diagnosis and treatment of tumors.

Contents of the invention

The object of the present invention is to provide a method for detecting genomic information based on restriction endonucleases.

A method for detecting genomic information based on restriction endonucleases, including the following steps:

(1) Use restriction endonucleases to cut the genome of the sample to obtain genomic DNA fragments of different lengths; at this time, alleles of homologous chromosomes are usually cut into DNA fragments of the same length;

This invention simulates enzyme digestion fragments of the genome of the target species and infers the distribution of genome fragments after enzyme digestion, thereby selecting appropriate restriction endonucleases (Figure 3); cell lysis is performed in a small volume system to release genomic DNA;

The restriction endonuclease is a restriction endonuclease that recognizes a 4-10 bp specific sequence. Preferably, it is a restriction endonuclease that recognizes a 6 bp or 8 bp specific sequence. More preferably, the restriction endonuclease The enzymes are EcoRI , SacI and AsiSI .

For the human genome, the length of restriction endonuclease fragments with a 6 bp recognition sequence is mostly distributed between 1 and 8 kb, while the length of restriction enzyme fragments with an 8 bp recognition sequence is mostly distributed between 1 and 8 kb. Between 15 kb-16 Mb (Figure 3). Therefore, if you want to obtain higher coverage, choose a restriction endonuclease with a 6 bp recognition sequence, such as Eco R I and Sac I. If you want a better enrichment effect, choose an endonuclease with an 8 bp recognition sequence, such as Asi S I. When hoping to obtain higher coverage, the digested fragments need to be distributed as concentrated as possible, that is, the DNA fragments obtained by cutting have similar lengths and are concentrated between 1-3 kb. In this case, better amplification uniformity can be achieved, and both Genome coverage and detection rates of both alleles.

In the above cell lysis step, the cells can be derived from any one of humans, animals, plants and microorganisms;

(2) Enrich long genomic DNA fragments from amplified or non-amplified genomic samples;

(3) Sequence the enriched long genomic DNA fragments on a sequencing platform;

(4) Computer analysis is performed on the data obtained by sequencing, by attaching the long genomic DNA fragments to the genomic region, and obtaining the sequence information of the sample in the genomic region through comparison and calculation. The sequence information includes genetic and epigenetic information.

The genomic samples are free DNA, DNA released from cells in culture medium (such as embryos or eggs), one or more cells or nuclei, viruses, mitochondria, chloroplasts, and other sample genomes.

The restriction endonuclease selected in step (1) is an endonuclease selected based on simulating the digestion fragments of the genome of the target species and inferring the distribution of genome fragments after digestion.

Preferably, step (2) performs end repair on the genomic DNA fragments, adds A, connects adapters, performs PCR amplification, and enriches long genomic DNA fragments after amplification. The adapters used can be either non-barcoded adapters or barcoded adapters.

Each PCR tube is carried out separately during subsequent purification and library construction using the adapter without barcode, and adapters at the 5' end and 3' end are provided during PCR amplification; use the adapter with barcode (that is, in The 5' end adapter is attached when the connector is connected). After the adapter is connected, the sample tubes with different barcodes are mixed and purified, and then amplified in one tube, and the 3' end adapter is attached through amplification.

The long genomic DNA fragment described in step (2) refers to a fragment with a length greater than 700 nucleotide pairs, preferably a fragment with a length greater than 1000 nucleotide pairs.

The amplification in step (2) is a polymerase chain reaction, and polymerase chain reaction and fragment screening are used to enrich long genomic DNA fragments, wherein the fragment screening is gel running fragment screening or magnetic bead fragment screening.

For large input samples, perform genomic fragment screening directly after restriction enzyme cleavage. The enzyme digestion method is used to make the fixed region have a fixed fragment size, and then through fragment screening, fragments of a specific size are enriched, so that the fixed region can be enriched, so that the sequencing area is concentrated in the genome region of a specific size, that is to say, The sequencing depth of genomic regions of a specific fragment length is increased, and the sequencing depth of genomic regions of other lengths is reduced or not detected. Therefore, allelic information in these regions can be detected more sensitively.

For a small amount of starting samples and single cell samples, enzyme digestion is followed by ligation, amplification, and gel fragment screening or magnetic bead fragment screening. Adapter ligation and PCR amplification also play a role in fragment screening, that is, the adapter ligation efficiency of overly long DNA fragments is reduced, and PCR amplification preferentially amplifies short fragments, so overly long fragments are filtered out. Small fragments are filtered out through gel fragment screening or magnetic bead fragment screening, so the sample fragments amplified by ligated adapters have a stronger ability to screen fragments. The final fragment length of this library is mainly distributed between 1-3 kb. In addition, since the amplification efficiency of allelic regions tends to be consistent, PCR can further enrich alleles, increasing the probability of detecting two alleles at the same time.

The sequencing platform described in step (3) is a long-read sequencing platform. Optionally, the sequencing platform is a Nanopore sequencing platform or PacBio sequencing platform and other long-read sequencing platforms subsequently developed.

In the library construction process of NGS sequencing, sequencing quality problems caused by amplification errors introduced by PCR and the read length limit of sequencing sequences (usually less than 500 bp) make it difficult for NGS technology to meet the higher requirements of some modern biological problems: such as long DNA Determination of repeated fragments, determination of DNA/RNA methylation modification issues, determination of structural variations, etc. The emergence of long-read sequencing technology has made up for the shortcomings of NGS. Currently, there are two main long-read sequencing platforms: Single-molecule real-time sequencing (SMRT) from Pacific Biosciences (PacBio) and single nanopore sequencing from Oxford Nanopore Technologies (ONT). PacBio is an SMRT based on zero-mode waveguide (ZMW) characteristics. ZMW is a nanophotonic containment structure consisting of circular holes in an aluminum coating placed on a transparent silicon dioxide substrate. The ZMW pores have a diameter of approximately 70 nm and a depth of approximately 100 nm. Due to the small aperture of ZMW, when light passes through the ZMW aperture, the light field exhibits exponential attenuation. Within the illuminated ZMW wells, the activity of DNA polymerase containing a single nucleotide is easily detected. PacBio SMRT sequencing technology uses topological circular DNA molecules as template libraries (called SMRTbell). SMRTbell consists of the two ends of the inserted double-stranded DNA fragment connected by a hairpin structure. It is a closed single-stranded circular DNA. . The length of the inserted DNA fragments can vary from 1 to greater than hundreds of thousands of bases, allowing the generation of long sequencing reads. After the SMRTbell is assembled, it is bound by DNA polymerase and loaded onto the SMRT Cell, which contains up to 8 million ZMW. In each ZMW, a single polymerase is anchored to the bottom, which can bind to either of the SMRTbell's hairpin adapters and initiate replication. In sequencing-by-synthesis, the polymerase uses SMRTbell as a template for processing, and combines four fluorescently labeled deoxynucleoside triphosphates that produce different emission spectra into the nascent chain. The excitation light located at the bottom of the small hole can excite the fluorescent label on the nucleotide substrate, and then the fluorescence signal is recorded through the monitoring system to obtain base information. During the entire sequencing process, DNA molecules do not need to undergo PCR amplification, and each DNA molecule can be sequenced individually. Currently, PacBio sequencing has two commonly used sequencing modes, continuous long reads (CLR) and rolling circle sequencing (Circular Consensus Sequencing, CCS). ONT is a sequencing technology based on electrical signals, and the core of this technology is protein nanopores. The basic working principle of nanopores is: a nanoscale pore will be formed between two electrolyte chambers, and there is an impermeable membrane between the electrolyte chambers. The protein nanopores are embedded in the synthetic membrane, with approximately hundreds of them. to thousands of nanopores that are immersed in an electrophysiological solution (synthetic membranes have very high electrical resistance, and the nature of protein nanopores is to form channels in the membrane). When a voltage is applied to the electrolyte chamber, a steady-state ionic current is generated through the pores. The passage of large molecules through a pore causes transient changes in the ion flux through the pore, so monitoring the current through the pore enables molecular sensing. These current fluctuations convey many properties of the sample, including biomolecule size, concentration and structure. By controlling the size of the pore, its surface properties, applied voltage, and solution conditions, one can tailor different nanopores to detect different types of biomolecules. And because nanopore sensing does not require biomolecule modification, labeling, or surface immobilization, the technology can be used to detect a wide range of molecules and complexes. ONT technology uses linear DNA molecules that are typically one to hundreds of kilobases in length, but some can reach several megabases. ONT sequencing first connects double-stranded DNA molecules to sequencing adapters preloaded with motor proteins, which unwind the double-stranded DNA and, together with an electric current, drive the negatively charged DNA through the pore at a controlled rate. As DNA passes through the nanopore, it causes characteristic disruptions in the electrical current, and these disruptions are analyzed in real time to determine the sequence of bases in the DNA strand. Currently long-read data can be generated on any of three standard ONT platforms: MinION, GridION, and PromethION. Another type of read generated by the ONT sequencing platform is the ONT ultra-long read. These reads, first generated by Josh Quick, are typically greater than 100 kb in length, but can also be several megabases long.

Genome structural variations (SVs) mainly include large-segment DNA deletions, insertions, segment duplications and other variation types in the genome. Studies have shown that SV is related to a variety of complex genetic diseases such as cancer, autism, and neurodevelopmental disorders, and has continued to receive attention in the fields of medicine and genetics in recent years. Due to the limitation of read length, NGS is greatly limited in the detection of SV. The advancement and popularization of long-read genome sequencing technology has enabled a large number of structural variations to be continuously discovered and studied, and some structural variations with strong pathogenicity have been gradually verified. This method is based on a long-read sequencing platform, which can efficiently detect SV, and can perform high-precision typing of haploid SV across the entire chromosome.

Based on the long-length sequencing platform, the linkage information of SNP or other variations can be better detected. Due to the short read length of NGS sequencing, most reads only have at most one variant information, while long-length sequencing can detect multiple variants on the same read, so it can be used to study the linkage of SNP, SV and other variants. Linkage information is crucial to the judgment of diseases. For example, for recessive genetic diseases, assume that a gene has mutations at two different sites. If the two mutations are located on the same chromosome, that is, linked, then this chromosome The gene on the cell loses its function, but there is a normal copy on the other chromosome, so the cell does not have a mutant phenotype; if the two mutations are located on different chromosomes, that is, both alleles have occurred. Mutations will cause disease. In addition, linkage information helps determine whether the mutated gene for a genetic disease comes from the father or the mother. Genomic imprinting refers to the different clinical phenotypes caused by differences in the genetic affinity (ie, paternal or maternal origin) of the disease-causing genes. Some genes are transcriptionally active only when they come from the father, while genes from the mother are not expressed. On the contrary, some genes are transcriptionally active. Genes are transcriptionally active only if they come from the mother; genes from the father are not expressed. At this time, distinguishing whether the mutated gene comes from the father or the mother can determine whether the disease-causing gene will be expressed, and thus determine the health status of the embryo.

Long-read sequencing has the potential to directly obtain DNA/RNA modifications (without antibodies or chemical treatments), which is of great significance. Modifications will change the efficiency of nucleotide matching. SMRT sequencing calculates modifications with single nucleotide accuracy by detecting the difference in the time that different fluorescently labeled dNTPs/NTPs bind to target nucleotides. At the same time, the modification will also change the electrical signal of the nucleotide, and Nanopore sequencing calculates the modification carried by detecting the electrical signal of the nucleotide passing through the nanopore. Based on this principle, when this method is applied to a large number of starting samples, since amplification is not required, epigenetic modification information can be retained and can be directly read through long-read sequencing. Therefore, this method can detect and compare allele epigenetic modifications in a large number of samples. This is something that NGS sequencing cannot achieve. And the linkage between different types of genomic variations and epigenetic states can be explored.

The fragment information in step (4) includes one or more of the following: 1) fragment length information; 2) fragment abundance information; 3) hybrid single nucleotide polymorphism information; 4) genome structural variation information, The genome structural variation information includes one or more of insertion, deletion, duplication, inversion, and translocation; 5) Repeated sequence information, the repeated sequence information includes short interspersed elements, long interspersed elements, and long terminal repeats One or more of elements, DNA repetitive elements, simple repeats, satellite foci, and other repetitive elements; 6) Genome copy number variation information; 7) Allele information; 8) Linkage relationships of allele information; 9) Table Epigenetic information includes DNA methylation and DNA hydroxymethylation.

The method of the present invention can detect genome information as low as a single cell at the same time, with high sensitivity, high probability of detecting two alleles at the same time, and can analyze as little as a single cell or cell nucleus. The present invention names this method as Restriction fragmentsligation-based genome amplification and third-generation sequencing (Refresh-seq), hereafter referred to as Refresh-seq. The connection of labeled adapters is called Refresh-seq (multiplexed).

Terms used in this invention:

Restriction endonuclease: an endonuclease that can recognize and cut specific double-stranded DNA sequences in organisms, including type I restriction endonuclease and type II restriction endonuclease, type I restriction enzyme Sexual endonucleases can catalyze both methylation of host DNA and hydrolysis of unmethylated DNA; whereas type II restriction endonucleases can only catalyze the hydrolysis of unmethylated DNA. Restriction enzymes are generally composed of the first letter of the microorganism's genus name and the first two letters of the species name, and the fourth letter indicates the strain. For example, the restriction endonuclease extracted from Bacillus amylolique faciens H is called Bam H. Several enzymes with different specificities that recognize different base sequences obtained from the same strain of bacteria can be compiled into different numbers, such as Hind II, Hind III, Hpa I, Hpa II, etc.

Homologous chromosomes: Two chromosomes with the same length and centromere position seen in the metaphase of mitosis of a cell, or paired chromosomes seen during meiosis. One homologous chromosome comes from the father and the other Come from the parent body; they are generally the same shape, size and structure.

Allele: A gene located at the same position on a pair of homologous chromosomes that controls different forms of the same trait.

Allele information: The allele information involved in this patent includes all variant types at alleles on homologous chromosomes, including SNPs, SVs, and repeated sequence information (short interspersed elements, long interspersed elements) on alleles. base elements, long terminal repeat elements, DNA repeat elements, simple repeats, satellite foci), epigenetic information, etc.

Epigenetic information: Epigenetic modification refers to the regulation of gene expression, which changes the DNA and proteins on the chromosome through chemical modification, thereby affecting the expression of genes. This modification can affect gene transcription, splicing, stability, translation, nucleosome assembly, and chromatin structure at multiple levels, thereby affecting the physiological and pathological processes of cells, as well as the phenotype of offspring. Common epigenetic modifications include DNA methylation, histone modifications, non-coding RNA, RNA modifications and chromatin remodeling.

DNA methylation: DNA methylation refers to adding a methyl group to the DNA molecule, thereby changing the chemical properties and structure of the DNA and affecting gene expression. Usually occurs on CpG dinucleotides and can inhibit gene expression. This modification is catalyzed by DNA methyltransferases (DNMTs). There are three main DNMTs in humans, namely DNMT1, DNMT3A and DNMT3B. Among them, DNMT1 is mainly responsible for maintaining methylation patterns, while DNMT3A and DNMT3B are responsible for new methylation.

DNA hydroxymethylation: DNA hydroxymethylation is the oxidation of 5-methylcytosine (5mC) in DNA methylation under the catalysis of TET family enzymes to form 5-hydroxymethylcytosine (5hmC) . 5hmC has very important biological functions. 5hmC not only participates in chromosome reprogramming and transcriptional regulation of gene expression, but also plays an important role in the DNA demethylation process. And research shows that 5hmC is closely related to the occurrence of tumors.

Linker: A sequence that allows two DNA molecules or the two ends of one DNA molecule to be paired after being digested by enzymes and then covalently connected by ligase.

Magnetic bead fragment screening: Magnetic beads can interact and adsorb with DNA under certain conditions. In higher concentrations of PEG and NaCl solutions, PEG takes away the water in the hydration layer outside the DNA molecule, causing the hydration layer to be destroyed and the DNA The molecules aggregate and precipitate, and the negatively charged phosphate groups are exposed. The sodium ions and the carboxyl groups on the surface of the magnetic beads form a "salt bridge", or also called an "electric bridge", causing the DNA to be adsorbed to the surface of the magnetic beads. The longer the DNA, the higher the surface. The more exposed negatively charged phosphate groups, the stronger the negative charge of the entire molecule, and the easier it is to adsorb to the magnetic beads. It only requires a lower concentration of PEG and NaCl to recover; the shorter the DNA, the higher the concentration required. The concentration of PEG and NaCl will destroy the hydration layer on the surface more thoroughly, exposing enough negatively charged phosphate groups, which can be adsorbed by the magnetic beads and recycled back; therefore, by controlling the concentrations of PEG and NaCl and The amount of magnetic beads used can screen DNA fragments of different lengths.

Beneficial effects of the present invention: For the first time, the present invention combines restriction endonuclease cleavage and ligation strategies with the third-generation single-cell genome sequencing platform, and develops the long-read genome sequencing technology Refresh-seq. Compared with SMOOTH-seq based on the random cutting principle, Refresh-seq increases genome coverage and uniformity. It improves the probability of simultaneous detection of both alleles of diploid cells, allowing considerable detection probabilities even at very shallow sequencing depths, and therefore has great potential for medical applications such as preimplantation genetics diagnosis. The method can be adjusted with different restriction enzymes to meet different needs. Generally speaking, Refresh-seq can obtain relatively high genome coverage when cutting with restriction endonucleases (such as Eco R I and Sac I) with 6 bp recognition sequences, and is the first choice for single-cell whole-genome sequencing; using 8 bp recognition sequences When the sequence is cut with a restriction endonuclease (such as Asi S I), Refresh-seq can enrich reads to specific genomic regions (Figure 1) under the premise of the same sequencing volume, thereby simplifying genome sequencing. Refresh-seq is based on a third-generation sequencing platform and can effectively detect structural variations and repetitive elements. Refresh-seq also has limitations. Due to the efficiency of the ligation reaction, the amplicon length is only 2-3 kb, which is much shorter than the full length of SMOOTH-seq, which is about 6 kb. Therefore Refresh-seq cannot capture very long insertion events due to its limited amplicon length range. The library construction of Refresh-seq can be completed within one day. The library construction cost of the single-tube version is 20 yuan/cell, and the library construction cost of the multi-tube version is 12 yuan/cell. The present invention successfully applied Refresh-seq technology to study the meiosis of single germ cells in male and female B6D2F1 mice. At 0.1-0.3× depth sequencing, the average coverage of sperm, PG oocytes and PB2 was approximately 5%, and the average coverage of oocytes and PB2 was 7.7%. This is consistent with the coverage of MALBAC-amplified sperm and oocytes. The inventors obtained high-resolution genetic maps of male and female meiotic recombination at low sequencing depth and revealed the differences between females and males. Because Refresh-seq has the characteristics of high uniformity and low allele loss rate, it has good application prospects in aneuploid sperm and egg cell screening. It also has advantages in detecting SVs in highly repetitive or low-complexity genomic regions due to its longer read length compared to NGS platforms. The inventors used Refresh-seq data to successfully conduct full-chromosome hetSV typing on sperm cells and female haploid germ cells, and analyzed the repetitive element characteristics of these SVs.

Description of drawings

Figure 1 is a flow chart of database construction in Example 1.

Figure 2 shows the test effect of using PCR to enrich long genomic DNA fragments.

Figure 3 is a simulation diagram of enzyme digestion fragments;

In the figure, a- Eco R I simulates the distribution of enzyme digestion fragments; b- Sac I simulates the distribution of enzyme digestion fragments; c- Asi SI simulates the distribution of enzyme digestion fragments.

Figure 4 shows the test chart of Refresh-seq (multiplexed) cross-contamination.

Figure 5 shows the situation where two alleles are detected simultaneously.

In the figure, a-the sequencing amount of each HG002 cell and the proportion of heterozygous SNP; b-quantification of the proportion of hybrid SNP by the three methods at 0.25× sequencing depth; c-the calculation of the area covering more than 5 reads. Gene deletion rate.

Figure 6 shows the performance of Refresh-seq using different endonucleases in different cell lines;

Figure a- shows the sequencing volume and genome coverage of HG001 cells amplified by Refresh-seq ( Eco R I/ Sac I) and Refresh-seq (multiplexed) ( Eco R I/ Sac I/ Asi S I), in which SMOOTH-seq Data are from HG002 cell line; b-shows sequencing volume and heterozygous SNP calls of HG001 cells amplified by Refresh-seq ( Eco R I/ Sac I) and Refresh-seq (multiplexed) ( Eco R I/ Sac I/ Asi S I) rate, where SMOOTH-seq data comes from HG002 cell line; c-shows Refresh-seq ( Eco R I/ Sac I), Refresh-seq (multiplexed) ( Eco R I/ Sac I/ Asi S I) and SMOOTH-seq amplified Sequencing volume and genome coverage of HG002 cells; d-shows Refresh-seq ( Eco R I/ Sac I), Refresh-seq (multiplexed) ( Eco R I/ Sac I/ Asi S I) and SMOOTH-seq amplified HG002 cells Sequencing volume and hybrid SNP detection rate; e- Refresh-seq and Refresh-seq (multiplexed) sequencing of HG002 cells using SMOOTH-seq and different restriction enzymes ( Eco R I/ Sac I/ Asi S I) depth.

Figure 7 shows the application of Refresh-seq in sperm;

In the figure a - Schematic diagram of the meiosis process of hybrid mouse sperm and Refresh-seq of single sperm. Mature sperm of B6D2F1 (B6×DBA F1 heterozygote) that have undergone meiotic homologous recombination are obtained. After flow sorting, each Refresh-seq is performed on a single sperm; b-displays the sequencing data volume and genome coverage of each sperm. Sperm with a genome coverage greater than 1% are selected for subsequent analysis. The demarcation is marked with a red dotted line; c-displays each sperm that has passed the quality control. Sequencing data volume and genome coverage of sperm cells, fitting linear regression at 95% confidence interval; d-Refresh-seq amplified average read length distribution of single sperm; e-distribution of the number of hetSNPs covered in each sperm; f- Diploid cells are identified by the discontinuity score of each sperm (the highest frequency autosomes are excluded at this time). The discontinuity score of diploid cells is much higher than that of haploid sperm. The red dotted line marks an inflection point beyond Cells at this inflection point are marked as potential diploid cells; g-use the number of reads on the X and Y chromosomes to distinguish X sperm from Y sperm.

Figure 8 shows the identification of aneuploid sperm by Refresh-seq;

In the figure, a-the discontinuity score of all chromosomes in each sperm, diploid cells are marked D1-D12, and aneuploid sperm cells are marked A1-A7; b-h-the discontinuity score of specific chromosomes in each sperm score. Diploid cells have higher discontinuity scores on most chromosomes, and aneuploid sperm have higher discontinuity scores only on chromosomes with abnormal segregation; i-7 aneuploid sperm cells The proportions of hetSNPs in the 19 autosomal chromosomes were compared to the gold standard. Blue dots indicate chromatid losses, red dots indicate chromatid gains. Dot size indicates deviation from the mean ratio. Verified aneuploidy. Chromosomes highlighted with rectangles, sperm A7 is more likely to be an unevenly amplified sample (technical error) than true aneuploidy; j - ratio of two alleles covered in sperm cells. Heatmap showing heterozygosity rates for 19 autosomes from 12 diploid cells (2N), 7 aneuploid sperm cells (1N ± m), and several haploid sperm cells (1N).

Figure 9 Identification of structural variants and chromosome typing for Refresh-seq;

In the figure, a-distribution of true positive structural variants per sperm; b-length distribution of identified true positive SVs, the local peak of SV length is represented by an orange dotted line; c-SVs (deletions and insertions) detected by Refresh-seq ) and the percentage of true positives for SVs with different numbers of supporting cells; d-accuracy of genome-wide typing of SVs at the chromosome scale; e-recall rate of correctly typed SVs; f-different missing events of typing The proportion of type elements; the proportion of different types of elements for g-typed insertion events.

Figure 10 shows Refresh-seq used for egg cells and polar bodies;

In the figure a - schematic diagram of sampling from hybrid female mice, B6D2F1 MII oocytes that have undergone meiotic homologous recombination are fertilized with DBA male mice or parthenogenetic activation induces PB2 extrusion to obtain haploid PB2, parthenogenetically activated oocytes and Diploid PB1, MII, fertilized eggs and separated by capillary; b-number and ploidy of different types of cells; c-displays sequencing data and genome coverage of each cell; d-crossover of haploid female germ cells Number distribution; e - Resolution of crossover assay in female haploid cells; f - Crossover position density map of all chromosomes in male and female mice, showing crossover density from centromere to telomere.

Detailed ways

In order to facilitate an understanding of the invention, the invention will be described more fully below. However, the invention may be embodied in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that a thorough understanding of the present disclosure will be provided.

Example 1

Based on the human genome sequence and the selected restriction endonuclease recognition sequence, the fragment length is simulated. As shown in Figure 3, most of the fragments simulated by Eco R I and Sac I are between 1-3 kb in length, so they are suitable for whole-genome amplification and sequencing. The Asi S I recognition sequence is 8 bp and is more sparsely distributed on the genome. Therefore, when using Asi S I for Refresh-seq library construction, it can simplify genome sequencing and perform deep sequencing of specific regions with the same amount of data.

The specific steps of Refresh-seq are shown in Figure 1: This example uses two human cell lines (HG002 and HG001). After the cells are washed three times with PBS containing 0.1% BSA, the single cells are separated using an oral pipette or flow cytometer. cells, into eight-row PCR tubes containing 2.5 μL of lysis buffer. Digest histones at 50°C for 3 hours, and inactivate the protease at 70°C for 30 minutes; the cell lysis solution is 10 mM Tris-EDTA (1M Tris + 0.1M EDTA), 1 mg/mL Qiagenprotease, 0.3% triton X-100, 20mM KCL, and 15mM DTT.

After single cell lysis, add 0.5 μL 10× enzyme digestion buffer, 1.9 μL water and 0.1 μL restriction endonuclease to digest the single cell gDNA. The reaction program is adjusted according to the restriction enzyme used. For EcoR I (NEWENGLAND BioLabs, cat# R3101L) and Sac I (NEW ENGLAND BioLabs, cat# R3156S), digest at 37°C for 15 minutes, then inactivate the enzymes at 65°C for 20 minutes. For AsiS I (NEW ENGLAND BioLabs, cat# R0630S), digest at 37°C for 1 hour, then inactivate the endonuclease at 80°C for 20 minutes. Then perform end repair and add A (Kapa Biosystems, Kapa HyperPrep kit, cat# KK8504), connect the dsDNA adapter (NEBNext Singleplex Oligos for Illumina) to the molecule with A added at the end, and then add USER enzyme (uracil-specific excision reagent) , NEW ENGLAND BioLabs, cat# M5505L) to cut the ring connector into a "Y" shaped connector. Each sample was purified with 1 × AMPure

ACACGACGCTCTTCCGATCT) and Barcode-P3 (ATCG-[24 bp P3-barcode 1-24]-GACTGGAGTTCAGACGTGTGCT) amplification. The PCR program was 98°C for 45 s, 98°C for 15 s, then 98°C for 15 s, 65°C for 30 s, and 72°C for 5 min, for 20 cycles. Afterwards, purify twice with 0.7 × AMPure XP (twice with 0.65 × AMPure XP for haploid cells). Purified amplicons were quantified using the Equalbit 1 × dsDNA HS Assay Kit. Mix the samples according to the sequencing requirements and then run the sequence on the computer.

For Refresh-seq (multiplexed), this example uses a barcoded adapter to increase throughput. First anneal the paired single-stranded oligonucleotides into a "Y"-shaped linker, NEB same-A: 5' phosphorylation (GATCGGAAGAGCACACGTCTGAACTCCAGTC and Barcoded-B: ACACT

CTTTCCCTACACGAC-[24 bp adapter-barcode 31-46]-GCTCTTCCGATC*T) was dissolved in water into a 100 μM mother solution, mixed 1:1 and then cooled anneal to obtain barcoded adapters with a concentration of 50 μM. The genomic DNA ends interrupted by restriction endonucleases were repaired and A was added and then ligated with barcoded adapters. The cells ligated with different barcoded adapters were pooled together. After 1×AMPure XP purification, Common-P5 (ACACTCTTTCCCTACACGAC) and Barcode-P3 ( ATCG-[24 bp P3-barcode 1-24]-GACTGGAGTTCAGACGTGTGCT) was amplified. Afterwards, purify twice with 0.7×AMPure XP. Purified amplicons were quantified using the Equalbit 1×dsDNA HS Assay Kit. Mix the samples according to sequencing requirements and run the sequence on the machine (Figure 2).

Table 1. Primer sequences involved in Refresh-seq

* stands for thiophosphoric acid.

After using ONT nanopore sequencing technology to sequence the Refresh-seq library and obtaining the original sequencing data, the inventor's basic processing of the data is to align the reads to the reference genome, which includes the following steps:

The raw data generated by ONT sequencing is converted to fastq format. According to the structure of the Refresh-seq double-end barcode library, this example uses nanoplexer v0.1 to perform two consecutive barcode disassemblies on each single cell, and Cutadaptv3.4 is used to remove adapter sequences at both ends of the reads and reads less than 500 bp in length. These reads were then aligned to the human reference genome hg38 or the mouse reference genome mm10 through minimap2 v2.24. In this example, samtoolsv1.14 is used to filter reads with mapping quality less than 30 and remove PCR duplicates.

Cross-contamination assessment: In order to evaluate the cross-contamination situation of Refresh-seq (multiplexed), this example uses hg38 and mm10 human and mouse mixed genome positioning strategies. The mixed genome is indexed by minimap2, and the parameter is '-I 10G'. This example calculates the number and proportion of reads mapped to the mm10 and hg38 genomes for each single cell. The reference genome species with the main comparison (greater than 90%) is judged to be the single-celled species. If the proportion of reads aligned to the genome of a minor species is greater than 10%, it is judged to be a cross-contaminated cell. The results are shown in Figure 4: The cells in the quality control results were not judged to be mixed human and mouse cells, indicating that the cross-contamination of Refresh-seq (multiplexed) is very small.

SNP site heterozygosity analysis: This example uses whatshap v.1.5 to calculate the likelihood of all three genotypes (0/0, 0/1, 1/1) at a given heterozygous SNP site and combine them Output to a VCF file along with genotype predictions. Run the command "whatshap genotype -reference ref.fasta -o genotyped.vcf variants.vcfreads.bam". The variants.vcf file is the HG002 or HG001 SNP benchmark set downloaded from GIAB.

The results are shown in Figure 5. Compared with SMOOTH-seq based on the principle of Tn5 random cutting genomic fragment amplification, Refresh-seq based on Eco RI has better amplification uniformity, higher whole-genome coverage, and higher double allele detection rate at single nucleotide diversity sites. At a sequencing depth of ~0.25×, Refresh-seq detected ~1.64% of heterozygous SNPs, which is 5 times that of SMOOTH-seq (~0.33%). Among heterozygous SNP sites covered by more than 5 reads, the average double allele capture rate of Refresh-seq was 62%, significantly higher than the 10% capture rate of SMOOTH-seq.

Figure 6 shows that Refresh-seq and Refresh-seq (multiplexed) have consistent performance on HG001 cells and HG002 cells. Refresh-seq and Refresh-seq (multiplexed) using Eco R I and Sac I had higher genome coverage than SMOOTH-seq, and more heterozygous SNPs were detected. Refresh-seq (multiplexed) using Asi S I achieves deeper sequencing depth with the same sequencing volume.

The above experiments confirmed the universality and advantages of Refresh-seq technology. Refresh-seq has better genome coverage and the probability of detecting two alleles at the same time. And Refresh-seq using long recognition sequence endonucleases can achieve read enrichment, thereby simplifying genome sequencing.

Example 2 Refresh-seq technology applied to single sperm sequencing

In this example, Eco R I is used to perform Refresh-seq. The specific library construction steps are consistent with Example 1. The difference is that the purification after library amplification is 0.65×AMPure XP purification twice. 676 sperm cells were amplified using Refresh-seq (single-tube version), and 152 sperm cells were amplified using Refresh-seq (multiplexed). Since there is no difference in the detection of crossover events between Refresh-seq and Refresh-seq (multiplexed), different versions of Refresh-seq are not distinguished subsequently.

The experimental results are shown in Figure 7. Refresh-seq can obtain sufficient genome coverage at low sequencing volume. Under ~0.1-0.3× deep sequencing, 700 sperm out of 828 sperm passed quality control, with genome coverage higher than 1% (Fig. 7b). The genome coverage increased nearly linearly with the increase of sequencing data, and at 0.1-1 Gb sequencing volume, the average coverage was approximately 5% (Figure 7c). The average read length was 1.9 kb (Fig. 7d), and the average number of reads per sperm was 143,914. On average, ~250,000 hetSNPs were detected per sperm (Fig. 7e), and the accuracy of SNP detection exceeded 98.9%. By defining discontinuity scores (i.e., the frequency with which continuous SNPs transition between paternal and maternal origin), Refresh-seq can efficiently screen out contaminated diploid cells and make accurate X and Y sperm determinations. Among the 700 sperm cells that passed the quality control, there were 688 haploid sperm cells and 12 contaminated diploid cells (Figure 7f). The diploid cells were labeled D1 to D12 (Fig. 7f), and the authenticity of these 12 diploid cells was verified using the distribution plot. Then, X spermatids and Y spermatids were distinguished based on the number and proportion of reads mapping to the X and Y chromosomes (Fig. 7g). A total of 344 X sperm cells and 329 Y sperm cells were identified, of which 8 sperm cells were indistinguishable (increase or decrease of sex chromosomes). The ratio of X sperm cells to Y sperm cells was close to 1:1, consistent with Mendel's law of segregation.

Example 3 Refresh-seq technology is applied to the identification of aneuploidy

Since Refresh-seq has a stronger ability to detect two alleles at the same time, a primary screening for aneuploidy is first performed by calculating the discontinuity score of each chromosome. The heterozygosity of the SNP site is then used to confirm the chromosome addition event. If a chromosome has an increase in copy number, two alleles of the same SNP site will often be detected at the same time in a chromosome of the heterozygous offspring. In the case of genes, based on the sudden increase in the number of two allele events detected simultaneously within the 1 Mb interval, the event of chromosome gain is confirmed; for the case of chromosome deletion, the number of SNP sites that can be detected is compared with normal. A sudden decrease occurs, so chromosomal deletion events can be determined by an increase or decrease in the number of SNPs detected at 1 Mb.

The experimental results are shown in Figure 8. Refresh-seq technology can use a variety of methods to identify aneuploidy. The genome of haploid sperm contains only one set of two sets of chromosomes from the parents. When a chromosome addition event occurs, the chromosome will have two different sets of genes from the parents at the same time. Ideally, each SNP site The genotypes of both the paternal and maternal parents can be detected at the same time. However, due to the ubiquitous phenomenon of allele loss, most SNP sites can often only detect one genotype. The genotypes of SNP sites in the chromosome-increasing interval are There will be random alternation of male and female parents, and there will be a difference in the frequency of alternating parental genotypes in the haploid interval, that is, the discontinuity score will increase significantly. Therefore, the first method screened out single sperm cells A1, A2, A3, A4, and A6 with increased chromosome occurrence by calculating the discontinuity score of each chromosome (Fig. 8a–h). In the case of random amplification, the increase of chromosomes means that more DNA fragments can be captured, and the loss of chromosomes means that the number of DNA fragments that can be captured is reduced. In the sequencing data, the positions where the chromosomes are increased have more sequencing reads. Coverage, more SNPs can be detected, while the positions of chromosome deletions have less read coverage, and fewer SNPs can be detected. Therefore, the second method can know the events of chromosome increase and decrease through the deviation of SNP number compared with the mean number of SNP numbers of all other chromosomes (Figure 8i). Consistent with the principle of method one, method three uses Refresh-seq to have the advantage of capturing two alleles at the same time. More double-genotype SNP sites can be detected in chromosomes with increased chromosome occurrence, that is, the heterozygosity increases, and The loss of chromosomes manifests itself as a decrease in heterozygosity (Fig. 8j). The aneuploid chromosomes found by these three methods verify each other and can be verified through chromosome distribution maps and CNV. Finally, 6 sperm with autosomal aneuploidy were found. Among them, A1, A3, and A6 had chromosome gains, A5 had chromosome deletions, and A4 and A6 had both gains and losses in chromosomes (chr3). Sperm A7 is more likely to be an unevenly amplified sample (technical error) than a true aneuploidy.

Example 4 Refresh-seq technology is applied to the identification of sperm structural variations

In this example, Refresh-seq uses cuteSV, a highly sensitive and fast bioinformatics software suitable for third-generation sequencing data, to detect structural variants (SVs) in long-read data generated by Nanopore. Parameters were set to default parameters specific to Nanopore and the minimum number of supported reads was set to 1 to achieve single-cell single-molecule resolution. In the analysis of the accuracy of multi-cell support for structural variation detection, SURVIVOR is first used to merge the SVs of all cells, and the accuracy of SVs supported by different cells is calculated according to the formula accuracy = true positive/(true positive + false positive). The reference set uses third-generation Nanopore sequencing data initiated from a large number of single sperm cells. In the haplotype typing of SV, we first establish a 0/1 matrix based on the parental genotypes of SV in single sperm in the reference set, where 0 represents the C57 maternal type and 1 represents the DBA paternal type. Judgment and reference set SV Whether it is consistent or not requires that the length of the SV is similar to the length of the SV in the reference set and the position needs to be within ±100 bp. If consistent, the genotype is marked with the same genotype as the SV in the reference set, and vice versa if inconsistent. The generated matrix first uses the tool ` hapiFrameSelection` in the R package Hapi to filter SVs supported by less than 5 sperm, and then selects the 100 cells with the most SVs as the precursor frame for subsequent typing. In order to improve the accuracy of typing, it is necessary to perform HMM calibration on the precursor framework. If more than half of the cells at each position support that an error has occurred, its genotype will be flipped. At this point, the basic framework is formed, and the missing genotypes are iteratively filled using the `imputationFun1` function with reference to other cells. Then use the `hapiPhase` function for primary haplotype typing, and use `hapiAssemble` after calibration with `hapiBlockMPR` to assemble high-resolution and high-consistency haplotypes.

The results are shown in Figure 9. Refresh-seq was able to identify structural variations in sperm, with an average of 973 structural variation events detected per cell (Figure 9a). In the length distribution 0 of all detected structural variants, two peaks appeared around 180 bp and 6 kb-7 kb, corresponding to the B1 element (equivalent to Alu in humans) and the LINE1 element respectively (Figure 9b). The accuracy of structural variation events detected by Refresh-seq can reach 80% when supported by more than three cells (Figure 9c), while the accuracy of chromosome-scale haploid typing is as high as 98% (Figure 9d). Genomic elements were successfully annotated for these typed structural variants (Fig. 9e–g).

Example 5 Refresh-seq technology is applied to the identification of egg cells and polar bodies

In this example, egg cells and polar bodies were obtained through fertilization and parthenogenetic activation, and EcoRI was used to perform Refresh-seq. The specific library construction steps were consistent with Example 1. A total of 185 second polar bodies, 87 parthenogenetically activated egg cells, 132 second polar bodies, 33 second meiotic cells and 26 zygotic cells were collected. Among them, the second polar body and parthenogenetically activated eggs are haploid cells, and an average of 14 crossover events were detected.

The experimental results are shown in Figure 10. In addition to being applied to male sperm cells, Refresh-seq also achieved good results in female germ cells. Refresh-seq can also obtain sufficient genome coverage under low sequencing volume. Diploid cells have higher genome coverage under the same sequencing volume, and the genome coverage increases with the increase in sequencing data. This shows that coverage saturation has not been reached (Figure 10c). Second polar bodies and parthenogenetically activated eggs were able to detect an average of 14 crossover events (Fig. 10d), with single-cell crossover events ranging from 6 to 25. The median crossover resolution is 283kb, so Refresh-seq can also obtain high-resolution crossover data in female haploid germ cells with shallow sequencing. Crossover distribution density plot shows that female mice have relatively fewer crossovers near centromeres and more crossovers near subtelomeres, relative to males. The degree of enrichment near subtelomeres will be lighter (Fig. 10f).

The above-mentioned embodiments only express several implementation modes of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the scope of protection of the patent of the present invention should be determined by the appended claims.

Claims (11)

1. A method for detecting genomic information based on restriction endonucleases, characterized by comprising the following steps: (1) Use restriction endonucleases to cut the genome of the sample to obtain genomic DNA fragments of different lengths; (2) Enrich long genomic DNA fragments from amplified or non-amplified genomic samples; (3) Sequence the enriched long genomic DNA fragments on a long-read sequencing platform; (4) Computer analysis is performed on the data obtained by sequencing, by attaching the long genomic DNA fragments to the genomic region, and obtaining the sequence information of the sample in the genomic region through comparison and calculation. 2. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the restriction endonuclease is a restriction endonuclease that recognizes a 4-10 bp specific sequence, preferably, In order to identify restriction endonucleases with specific sequences of 6 bp and 8 bp, it is more preferable to select Eco R I and Sac I when the goal is to obtain higher coverage, and to select when the goal is to achieve better enrichment effects. Asi S I; When the goal is to obtain higher coverage, the resulting DNA fragments are of similar length and concentrated between 1-3 kb. 3. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the genomic sample is free DNA, DNA released by cells in the culture medium, one or more cells or nuclei, Viruses, mitochondria or chloroplasts. 4. The method for detecting genomic information based on restriction enzymes according to claim 1, characterized in that step (2) performs end repair on the genomic DNA fragments, adds A, connects adapters, performs PCR amplification, and amplifies the genomic information. After enrichment, long genomic DNA fragments are enriched. 5. The method for detecting genomic information based on restriction enzymes according to claim 4, characterized in that the adapter used in the amplification in step (2) is an adapter without a barcode or an adapter with a barcode; use During the subsequent purification and library construction process of the adapter without barcode, each PCR tube is carried out separately, and adapters at the 5' end and 3' end are attached during PCR amplification; using the adapter with barcode, after the adapters are connected, Sample tubes with different barcodes are mixed and purified, amplified in one tube, and 3'-end adapters are attached through amplification. 6. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the sequencing platform described in step (3) is a long-read sequencing platform, and optionally, the sequencing platform It is Nanopore sequencing platform or PacBio sequencing platform. 7. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the restriction endonuclease selected in step (1) is inferred based on simulation of restriction endonuclease fragments of the genome of the target species. The distribution of genomic fragments after digestion, thereby selecting the endonuclease. 8. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the long genomic DNA fragment in step (2) refers to a fragment with a length greater than 700 nucleotide pairs, preferably a length of Fragments larger than 1000 nucleotide pairs. 9. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the amplification in step (2) is a polymerase chain reaction, and polymerase chain reaction and fragment screening are used to enrich the genome information. Collect long genomic DNA fragments, and the fragment screening is gel-running fragment screening or magnetic bead fragment screening. 10. The method for detecting genomic information based on restriction endonucleases according to claim 1, characterized in that the sequence information in step (4) includes one or more of the following: 1) fragment length information; 2) Fragment abundance information; 3) Hybrid single nucleotide polymorphism information; 4) Genome structural variation information, which includes insertions, deletions, duplications, inversions, and translocations; 5) Repeated sequence information, The repeated sequence information includes short interspersed elements, long interspersed elements, long terminal repeat elements, DNA repeat elements, simple repeats, and satellite foci; 6) genome copy number variation information; 7) allele information; 8) alleles Linkage relationship of genetic information; 9) Epigenetic information, which includes DNA methylation and DNA hydroxymethylation. 11. The method for detecting genomic information based on restriction endonucleases according to claim 10, characterized in that the allele information is the variation type at alleles on homologous chromosomes, including alleles. SNP, SV, repeated sequence information, epigenetic information; the repeated sequence information includes short scattered elements, long scattered elements, long terminal repeat elements, DNA repeat elements, simple repeats, and satellite foci; the epigenetic information Including DNA methylation and DNA hydroxymethylation.