Method for improving gene phasing accuracy and gene phasing method using the same

Abstract

This application relates to the field of bioinformatics, specifically to a method for identifying gene phasing errors and a gene phasing method using the same. The identification method includes: 1) performing a first verification on individual variation information based on family data and Mendelian inheritance laws to identify first erroneous variation sites that do not conform to the inheritance laws; 2) setting multiple thresholds to perform a second verification on the individual variation information to identify potential second erroneous variation sites; and 3) performing a third verification on the potential second erroneous variation sites based on the correlation between non-erroneous variation sites and the potential second erroneous variation sites to determine the second erroneous variation sites. The method for identifying gene phasing errors and the gene phasing method using the same proposed in this application can greatly improve the accuracy of gene phasing.

Description

Technical Field

[0001] This application relates to the field of bioinformatics, specifically to a method for identifying gene phasing errors to improve gene phasing accuracy and a gene phasing method using the same. Background Technology

[0002] Gene phasing, also known as haplotype typing, is a method for locating alleles onto paternal or maternal chromosomes. Because most sequencing technologies cannot distinguish the chromosomal origin of sequenced fragments, they can only detect variations in the genome but cannot pinpoint the specific chromosome. Therefore, gene phasing has been developed to differentiate these mutated alleles and determine their relative positions on chromosomes, thereby obtaining haplotypes with variations shared by both parents. Gene phasing is crucial for understanding the complexity of genetic diseases, developing precision medicine, and exploring human genetic diversity.

[0003] In existing technologies, gene phasing results typically need to be evaluated to determine their relative accuracy. Researchers generally use multiple metrics to measure gene phasing results, such as the success rate, switch error rate, and N50 phased block length. These metrics provide an overall assessment of the gene phasing results, allowing for the determination of accuracy and thus confirming the usability of the results. However, in accuracy assessment, the accuracy of the evaluated gene phasing results, and consequently the usability of the results, is only accurate if both the detection of variant sites and the accuracy of the truth set used to assess phasing accuracy are true. In reality, existing variant site detection schemes and the truth sets provided by various species all have a certain error rate. This makes it difficult for traditional methods of evaluating gene phasing results to accurately identify potential phasing errors. Claims of high accuracy in gene phasing results often lead to erroneous inferences of genetic associations and disease susceptibility, affecting subsequent genetic analyses. Summary of the Invention

[0004] This application addresses at least one of the problems of the related technology in the following aspects.

[0005] Therefore, embodiments of this application provide a method for identifying gene phasing errors to improve gene phasing accuracy. This method can accurately identify variant detection errors and gene phasing errors without relying on variant truth sets and phasing truth sets, thereby greatly improving the reliability of gene phasing results.

[0006] The method for identifying gene phasing errors to improve gene phasing accuracy proposed in this application includes the following steps:

[0007] 1) Based on family data, perform a first verification on individual variation information according to a first criterion to identify a first erroneous variation site, wherein the first criterion is Mendelian inheritance law, and the first erroneous variation site includes variation sites in the individual variation information that do not conform to Mendelian inheritance law;

[0008] 2) According to the second criteria, the individual variation information is subjected to a second verification to identify potential second erroneous variant sites, wherein the second criteria include: quality value, sequencing depth and / or allele frequency, and the potential second erroneous variant sites include variant sites in the individual variation information that do not meet the second criteria; and

[0009] 3) According to the third criterion, the potential second error variant sites are subjected to a third verification to determine the second error variant sites, wherein the third criterion includes the correlation between non-error variant sites and the potential second error variant sites, and the second error variant sites include variant sites among the potential second error variant sites that meet the third criterion.

[0010] In some embodiments, the pedigree data includes paternal genotype data and maternal genotype data. For a specific mutation site in the individual, the non-compliance with Mendelian inheritance laws includes:

[0011] Both the paternal and maternal genotypes are wild-type, while the individual described is a mutant.

[0012] Both the paternal and maternal genotypes are homozygous mutants, while the individual in question is a heterozygous mutant.

[0013] The father's genotype is wild-type, the mother's genotype is homozygous mutant, and the individual in question is a homozygous mutant.

[0014] The maternal genotype is wild-type, the paternal genotype is homozygous mutant, and the individual in question is homozygous mutant.

[0015] The paternal genotype is wild-type, the maternal genotype is heterozygous mutant, and the individual is either homozygous mutant or compound heterozygous mutant;

[0016] The individual is either a wild-type mother, a heterozygous mutant father, or a homozygous mutant or a compound heterozygous mutant; or a homozygous mutant father, a heterozygous mutant mother, or a compound heterozygous mutant mother.

[0017] The maternal genotype is homozygous mutant, the paternal genotype is heterozygous mutant, and the individual is a compound heterozygous mutant; or

[0018] Both the paternal and maternal genotypes are heterozygous mutants, while the individual in question is a compound heterozygous mutant.

[0019] In some embodiments, the quality value includes one or more of the following: the base quality value of the variant site, the alignment quality value of the sequencing read containing the variant site, preferably the average alignment quality value and the variant quality value of the sequencing read containing the variant site.

[0020] In some embodiments, the second standard includes:

[0021] The base mass value of the variant site is ≥10;

[0022] The average alignment quality of the sequencing reads containing the variant sites is ≥15, preferably ≥20;

[0023] The mutation quality value of the mutation site is ≥10, preferably ≥20;

[0024] The sequencing depth of the variant sites is ≥5, preferably ≥10; and

[0025] The allele frequency of the variant site is ≥0.2, preferably ≥0.3.

[0026] In some embodiments, the association includes sequencing read detection association and / or linkage disequilibrium (LD) association, and the third criterion includes:

[0027] The potential second error variant site and the non-error variant site were detected in the same sequencing read; and / or

[0028] The LD values ​​of the potential second error variant sites and the non-error variant sites described in the LD reference database.

[0029] In some embodiments, the LD value between the potential second error mutation site and the non-error mutation site is ≥0.7, preferably ≥0.8.

[0030] In some embodiments, the potential second erroneous variant site is adjacent to the non-erroneous variant site. In some embodiments, the potential second erroneous variant site is no more than the length of the sequencing read from the non-erroneous variant site.

[0031] Embodiments of this application also provide a gene phasing method, comprising the following steps:

[0032] 1) The individual variation information is screened according to the method for identifying gene phasing errors as described in any of the above embodiments to identify the first erroneous variation site and the second erroneous variation site;

[0033] 2) Exclude the first and second erroneous variant sites from the individual variant information; and

[0034] 3) Gene phasing is performed based on individual variation information after excluding the first and second erroneous mutation sites.

[0035] In some embodiments, the method further includes: obtaining individual variation information based on the individual's sequencing data.

[0036] In some embodiments, the sequencing data is second-generation sequencing data or third-generation sequencing data.

[0037] Embodiments of this application also provide a gene phasing device, comprising:

[0038] The error mutation site identification module is used to screen the individual variation information according to the method for identifying gene phasing errors as described in any of the above embodiments, so as to identify the first error mutation site and the second error mutation site;

[0039] An erroneous variant site exclusion module is used to exclude the first erroneous variant site and the second erroneous variant site from the individual variant information; and

[0040] The gene phasing module is used to perform gene phasing based on individual variation information after excluding the first and second erroneous mutation sites.

[0041] In some embodiments, the gene phasing device further includes a variant detection module for obtaining variant information of the individual based on the individual's sequencing data.

[0042] Embodiments of this application also provide an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method for identifying gene phasing errors or a gene phasing method according to any of the above embodiments.

[0043] Embodiments of this application also provide a non-transitory computer-readable storage medium, characterized in that the storage medium includes a program that can be executed by a processor to implement a method for identifying gene phasing errors or a gene phasing method according to any of the above embodiments.

[0044] The embodiments of this application achieve the following beneficial effects:

[0045] The method for identifying gene phasing errors and the gene phasing method based on this application proposed in this application can accurately identify variant detection errors and gene phasing errors without relying on variant truth sets and phasing truth sets, thereby greatly improving the reliability of gene phasing results. The method proposed in this application can effectively eliminate variant detection errors generated in sequencing and analysis, as well as false positive sites that are difficult to identify in traditional gene phasing, especially for complex variant sites and local phasing errors, thereby achieving high-accuracy gene phasing. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart of a method for identifying gene phasing errors according to an embodiment of this application;

[0048] Figure 2 This is a schematic diagram of erroneous judgment results generated by existing gene phasing methods;

[0049] Figure 3 Examples of non-Mendelian inheritance laws according to embodiments of this application are shown;

[0050] Figure 4 The results produced by the gene phasing method according to embodiments of this application are shown in part;

[0051] Figure 5A An example is shown of the method proposed in this application for identifying sites that do not conform to Mendelian inheritance laws (i.e., first error mutation sites);

[0052] Figure 5B It shows Figure 5A The variant sites in the true value set corresponding to the shown site regions;

[0053] Figure 6 An example of identifying false positive sites (i.e., second error variant sites) based on sequencing read association is shown in the method proposed in this application.

[0054] Figure 7 An example of identifying false positive sites (i.e., second erroneous variant sites) based on linkage disequilibrium association is shown in the method proposed in the embodiments of this application. Detailed Implementation

[0055] The present invention will now be described in further detail with reference to specific embodiments. The embodiments given are merely illustrative of the invention and are not intended to limit its scope. The embodiments provided below can serve as a guide for further improvements by those skilled in the art and do not constitute a limitation on the invention in any way.

[0056] As used herein, the terms "an" or "a" are intended to mean "one or more" or "one or more types of". The term "comprising" and its variations, such as "including" and "containing", are used to indicate that the addition of other steps or elements is optional and non-exclusive. When used to define compositions and methods, "consistently made of" should mean excluding other elements that have any essential significance to the composition when used for the intended purpose. "Constituted of" should mean excluding trace elements and basic method steps that are more than the other components. Any methods, apparatus, and materials similar to or equivalent to those described herein may be used in the practice of this invention. The following definitions provided herein are intended to aid in understanding certain terms frequently used herein and do not constitute a limitation on the scope of the invention.

[0057] As used herein, the term "sample" refers to any tissue or bodily fluid from an individual suitable for nucleic acid extraction. The individual can be any living or non-living organism, including but not limited to humans or non-human animals. Any human or non-human animal can be selected, including but not limited to mammals, reptiles, birds, amphibians, fish, ungulates, ruminants, bovids (e.g., cattle), equines (e.g., horses), goats and sheep (e.g., sheep, goats), suidae (e.g., pigs), alpacas (e.g., camels, llamas, alpacas), monkeys, apes (e.g., gorillas, chimpanzees), bears (e.g., bears), poultry, dogs, cats, mice, rats, fish, dolphins, whales, and sharks. The individual can be male or female (e.g., women, pregnant women). The individual can be of any age (e.g., embryo, fetus, infant, child, adult).

[0058] In this application, nucleic acids can be isolated from any type of suitable biological sample. The sample can be any sample isolated from or obtained from an individual or a portion thereof. Non-limiting examples of samples include liquids or tissues of an individual, including but not limited to blood or blood products, cord blood, chorionic villi, amniotic fluid, cerebrospinal fluid, cerebrospinal fluid, washings, biopsy samples, intermembranous fluid samples, cells or portions thereof, female genital tract washes, urine, feces, sputum, saliva, nasal mucosa, prostatic fluid, irrigation fluid, semen, lymph, bile, tears, sweat, breast milk, mammary gland fluid, etc., or combinations thereof. In some embodiments, the sample can be blood, and sometimes plasma or serum. Other suitable biological samples are familiar to those skilled in the art. Biological samples can be obtained using techniques well within the general knowledge of clinical practitioners. The methods disclosed herein are applicable to various types of nucleic acids in any type of sample, such as DNA or RNA, including genomic DNA, free nucleic acids, transcripts, mitochondrial DNA, chloroplast DNA, etc.

[0059] In the embodiments of this application, "variation" may involve single-base or multi-base variations, such as single nucleotide polymorphisms (SNPs); small insertions and deletions (collectively referred to as InDels), which generally occur in short, ordered gene segments on the genome, less than 50 bp in length; and larger-scale structural variations (SVs), such as insertions and deletions of segments longer than 50 bp (Big Indels), chromosomal inversions, translocations between or within chromosomes, copy number variations (CNVs), tandem repeats, chimeras, etc.

[0060] In this embodiment, "haplotype" refers to a group of related single nucleotide polymorphisms located in a specific region of a chromosome, which tend to be inherited as a whole by offspring; it is also called a haplotype or haplotype. A haplotype can refer to a pair of loci, a region of a chromosome, or the entire chromosome.

[0061] In the embodiments of this application, "locus" can refer to the location of a gene on a chromosome in a broad sense, or it can refer to the location of a nucleotide (or base pair) of any length on the genome in a narrow sense. Therefore, in some embodiments of this application, "site" can be used interchangeably with "locus".

[0062] As used herein, "pedigree data" includes parental genotype data, namely paternal genotype data and maternal genotype data. In some embodiments, pedigree data may also include other genealogical data.

[0063] As used herein, "Mendelian laws of inheritance" can refer to the law of segregation and the law of independent assortment. These laws can be used to examine individual variant sites based on family data to identify those that do not conform to these laws (i.e., first-error variant sites). In some embodiments, situations that do not conform to Mendelian laws of inheritance may include:

[0064] Both the paternal and maternal genotypes are wild-type, while the individual described is a mutant.

[0065] Both the paternal and maternal genotypes are homozygous mutants, while the individual in question is a heterozygous mutant.

[0066] The father's genotype is wild-type, the mother's genotype is homozygous mutant, and the individual in question is a homozygous mutant.

[0067] The maternal genotype is wild-type, the paternal genotype is homozygous mutant, and the individual in question is homozygous mutant.

[0068] The paternal genotype is wild-type, the maternal genotype is heterozygous mutant, and the individual is either homozygous mutant or compound heterozygous mutant;

[0069] The individual is either a wild-type mother, a heterozygous mutant father, or a homozygous mutant or a compound heterozygous mutant; or a homozygous mutant father, a heterozygous mutant mother, or a compound heterozygous mutant mother.

[0070] The maternal genotype is homozygous mutant, the paternal genotype is heterozygous mutant, and the individual is a compound heterozygous mutant; or

[0071] Both the paternal and maternal genotypes are heterozygous mutants, while the individual in question is a compound heterozygous mutant.

[0072] In this embodiment of the application, "complex heterozygous mutation" can refer to a complex heterozygous mutation composed of heterozygous mutations from both parents. In some embodiments, complex heterozygous mutation refers to a biallelic mutation.

[0073] As used herein, a “sequencing read” refers to a polynucleotide sequence sequenced from any part or all of a nucleic acid molecule. For example, a sequencing read can be a short nucleotide sequence (e.g., 20-150 nucleotides) sequenced from a nucleic acid fragment, a short nucleotide sequence at one or both ends of a nucleic acid fragment, or a sequencing sequence of an entire nucleic acid fragment present in a biological sample (e.g., 20-300 bp, 50-150 bp, or even up to 10 kb). Sequencing reads can be obtained in various ways, such as using sequencing technologies or using probes for microarray capture. In some embodiments, “sequencing” can be second-generation or third-generation sequencing, such as DNBSEQ, Illumina Solexa sequencing by synthesis, Nanopore sequencing, single-molecule real-time fluorescence sequencing (SMRT, PacBio), etc.

[0074] As used herein, the term "sequencing depth" refers to the number of times a locus is covered by sequence reads aligned to that locus. A locus may be as small as a nucleotide, as large as a chromosome arm, or as large as the entire genome. Sequencing depth can be expressed as 10×, 50×, 100×, etc., where "×" indicates the number of times the locus is covered by sequence reads. Sequencing depth can also be applied to multiple loci or the entire genome; in this case, "×" can refer to the average number of times the locus, haploid genome, or entire genome is sequenced.

[0075] As used in this article, the term "quality score (Q-score)" is an integer mapping of the probability of a base identification error, Q = -10 * lgP, where P is the probability of a base identification error. The higher the quality score, the lower the probability of being detected incorrectly.

[0076] As used in this article, the term "alignment quality value (MAPQ or MQ)" is a value output by relevant alignment tools (such as bowtie2, bwa, Minimap, etc.) that reflects the quality of read alignment to the genome. A higher MAPQ indicates better alignment quality. The term "average alignment quality value" refers to the average alignment quality value of all sequencing reads containing a specific variant site.

[0077] As used in this article, the term "quality of variation (QUAL)" refers to the quality value of the variation information output by relevant variation detection tools (such as GATK, Deep Variant, etc.). The higher the QUAL, the more reliable the variation information is.

[0078] As used in this article, the term "allele frequency" can refer to the probability or ratio of a particular allele to occur in a biological population. In this case, the allele frequency can be determined by consulting a specific reference database.

[0079] As used in this article, the term "linkage disequilibrium (LD)" refers to the higher than random frequency of alleles belonging to two or more gene loci appearing simultaneously on a chromosome. LD values ​​can be determined using an LD reference database.

[0080] Existing methods for identifying gene phasing error sites and results presentation

[0081] Current methods for identifying gene phasing error sites require comparing the generated gene phasing results with the truth set, screening for consistent variant sites, analyzing whether the orientation results are completely consistent, whether they contain any phasing error regions, and outputting evaluation metrics, including the phasing success rate (gene phasing proportion), switch error rate, hamming error, and the N50 length of successfully phased large fragments. These metrics provide useful information for the preliminary assessment of gene phasing quality. Figure 2 As shown, the actual phasing result should be two haploids, one red and one blue. However, the actual phasing result may reverse the regions with single or multiple bases, resulting in a switch error, or a partial mutation site from another chromosome may be mixed into a single chromosome. The proportion of mixed sites is called the hamming error.

[0082] It is evident that the success of existing methods for identifying erroneous variant sites relies on two conditions. First, the detected variant sites must be accurate. If false positives are detected and successfully phasing, the phasing error caused by the site detection error will not be identified. In current technologies, the accuracy of variant sites depends on the accuracy of the variant truth set. The second condition is that the phasing truth set used for evaluation must be completely accurate. However, in reality, many species have not yet published mature phasing truth sets, and even published phasing truth sets are not 100% accurate. Furthermore, published versions are generally difficult to gain widespread industry acceptance. Therefore, it is difficult for these two conditions to be met simultaneously in existing identification methods, thus preventing them from revealing localized gene phasing errors or specific types of gene phasing errors.

[0083] A method for identifying gene phasing errors to improve gene phasing accuracy.

[0084] This application proposes a method for identifying gene phasing errors to improve gene phasing accuracy. This method can accurately identify variant detection error sites and phasing error regions without referring to a variant truth set or relying on a phasing truth set, thereby improving the accuracy of gene phasing.

[0085] like Figure 1 As shown, the method for identifying gene phasing errors proposed in this application embodiment may include the following steps:

[0086] S1: Based on family data, perform a first verification of individual variation information according to a first criterion to identify the first erroneous variation site, where the first criterion is Mendelian inheritance law, and the first erroneous variation site includes variation sites in individual variation information that do not conform to Mendelian inheritance law.

[0087] In some embodiments, all variant sites can be screened based on Mendelian inheritance laws and family data, and those variant sites that do not conform to Mendelian inheritance laws can be identified as erroneous variant sites, i.e., variant detection error sites. The method proposed in this application integrates Mendelian inheritance laws, achieving accurate identification of variant detection errors without the need to refer to a true set of variants, thereby avoiding the introduction of these variant detection errors into the final gene phasing, and thus greatly improving the accuracy of gene phasing.

[0088] S2: Perform a second verification on the individual variation information according to the second criteria to identify potential second erroneous variant sites, wherein the second criteria include: quality value, sequencing depth and / or allele frequency, and potential second erroneous variant sites include variant sites in the individual variation information that do not meet the second criteria.

[0089] In some embodiments, the second criterion may include the sequencing quality value of the variant site, such as the base quality value; the alignment quality value of the sequencing read containing the variant site; the variant quality value of the variant site; the sequencing depth of the variant site and / or the allele frequency of the variant site.

[0090] In some specific embodiments, based on the fact that the mutation site is a single base variation, such as a SNP, the second criterion may include: the sequencing quality value of the SNP site, such as the base quality value; the alignment quality value of the sequencing reads containing the SNP site, preferably the average alignment quality value of all sequencing reads containing the SNP site; the mutation quality value of the SNP site; the sequencing depth of the SNP site and / or the allele frequency of the SNP site.

[0091] In some specific embodiments, the second criterion may be set as one or more of the following:

[0092] i. The base mass value of the mutation site is ≥10, for example, it can be ≥20 or ≥30, etc.;

[0093] ii. The average alignment quality of the sequencing reads containing the said variant sites is ≥15, preferably ≥20;

[0094] iii. The mutation quality value of the mutation site is ≥10, preferably ≥20;

[0095] iv. The sequencing depth of the variant site is ≥5, preferably ≥10; and

[0096] v. The allele frequency of the variant site is ≥0.2, preferably ≥0.3, for example ≥0.4 or 0.5.

[0097] In some embodiments, individual variation information can be screened based on a second criterion to identify true positive variants (i.e., non-erroneous variants, TP) and high-risk false positive variants (i.e., potential second erroneous variants, FP). In some embodiments, variants that meet the second criterion are identified as true positive variants (TP), and variants that do not meet the second criterion are identified as high-risk false positive variants (FP). It is understood that true positive variants (i.e., non-erroneous variants, TP) do not include first erroneous variants.

[0098] S3: According to the third criterion, perform a third verification on potential second error variant sites to identify second error variant sites, wherein the third criterion includes the association between non-error variant sites and potential second error variant sites, and the second error variant sites include variant sites among potential second error variant sites that meet the third criterion.

[0099] In some embodiments, the association between true positive variant sites (TP) and high-risk false positive variant sites (FP) included in the third standard may include sequencing read detection association. In some specific embodiments, sequencing read detection association means that a true positive variant site (TP) and a high-risk false positive variant site (FP) appear simultaneously in the same sequencing read. Based on the fact that both are detected in the same sequencing read, they are determined to have a strong association, and the high-risk false positive variant site (FP) is identified as an erroneous variant site.

[0100] In other embodiments, the association between true positive variants (TPs) and high-risk false positive variants (FPs) included in the third criterion may also include linkage disequilibrium (LD) association. In some specific embodiments, an LD threshold can be set for both to determine whether they are associated. For example, based on an LD value ≥ 0.7, such as an LD value ≥ 0.75 or 0.8 or any value in between, the two are determined to be strongly associated, and the high-risk false positive variant (FP) is identified as an erroneous variant.

[0101] In some embodiments, the true positive variant site (TP) and the high-risk false positive variant site (FP) used to determine the association can be adjacent variant sites in the variant information. For example, when determining the association of sequencing reads, the distance between them is sufficient for them to be read by the same sequencing read. Therefore, in this case, the true positive variant site (TP) and the high-risk false positive variant site (FP) used to determine the association of sequencing reads are no more than the length of the sequencing read containing them, specifically no more than the read length of second-generation sequencing or the read length of third-generation sequencing, the read lengths of second-generation sequencing and third-generation sequencing being well known in the art.

[0102] The method for identifying gene phasing errors proposed in this application utilizes Mendelian inheritance principles to accurately identify variant detection errors. Simultaneously, it sets multiple thresholds (variant quality value, alignment quality value, sequencing quality value, sequencing depth, etc.) to screen for high-risk false positive variant sites (FP) and highly reliable true positive variant sites (TP). The association between FP and TP is calculated using whether they are simultaneously detected by a sequencing fragment and linkage disequilibrium values, thereby identifying phasing error regions and avoiding subsequent phasing errors caused by FP sites strongly associated with TP, such as conversion errors or incorrectly cutting long phasing fragments into shorter ones. The method proposed in this application does not rely on known variant and phasing truth sets and can accurately identify variant detection and phasing errors, thereby further improving gene phasing accuracy.

[0103] This application also proposes a gene phasing method, including the following steps:

[0104] 1) Screen individual variation information according to the method for identifying gene phasing errors described in any of the above embodiments to identify the first erroneous variation site and the second erroneous variation site;

[0105] 2) Exclude the first and second erroneous variant sites from the individual variant information; and

[0106] 3) Gene phasing is performed based on individual variation information after excluding the first and second erroneous variant sites.

[0107] In some embodiments, based on the variant information that has screened out the first and second erroneous variant sites, genome phasing of the variant information can be performed using tools and procedures well known in the art. For example, software such as SNPHap, SHAPEIT, WhatsHap, and HapCUT2 can be used for genome phasing. Specifically, this may include linking these reads into longer contigs based on the overlap between sequencing reads, and then classifying heterozygous sites (0 / 1 and 1 / 0) into alleles from the paternal or maternal parent based on whether they can be completely covered by the same sequencing read, thereby achieving genome phasing of variant sites.

[0108] In some embodiments, the gene phasing method may further include: obtaining individual variation information based on individual sequencing data. It is understood that the process of analyzing sequencing data to obtain individual variation information is well known in the art, and may include, for example, sequencing data quality control, read alignment (and optional alignment correction), variant site detection (e.g., SNP calling), and variant site annotation, etc. Furthermore, corresponding quality statistics may be performed after each step.

[0109] In the embodiments of this application, the detection of variant sites, the screening of erroneous variant sites according to the first to third criteria, the exclusion of the screened erroneous variant sites (or the annotation of relevant information), and the subsequent gene phasing can be implemented based on command lines, scripts, software, or self-written scripts known in the art. This application does not limit the specific implementation method.

[0110] The gene phasing method proposed in this application, by comprehensively considering genetic laws and setting multiple judgment thresholds, can accurately identify and eliminate variation detection errors and gene phasing errors introduced by sequencing or data analysis. Furthermore, this method does not require reference to or reliance on a truth set, thus avoiding false positive errors caused by the inaccurate truth sets in existing technologies. This achieves highly accurate gene phasing, especially in the identification of complex genetic variations, false positive sites, and the differentiation of local phasing errors, demonstrating more precise judgment. Therefore, it has great application value in genetic association and disease susceptibility research.

[0111] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a storage medium such as a server, another computer, disk, optical disk, flash drive, or portable hard drive, and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0112] Therefore, another embodiment of this application provides a gene phasing device, comprising: an error mutation site identification module, used to screen the individual variation information according to the method for identifying gene phasing errors described in any of the above embodiments, to identify a first error mutation site and a second error mutation site; an error mutation site exclusion module, used to exclude the first error mutation site and the second error mutation site from the individual variation information; a gene phasing module, used to perform gene phasing based on the individual variation information after excluding the first error mutation site and the second error mutation site; and an optional variation detection module, used to obtain the individual variation information based on the sequencing data of the individual.

[0113] Another embodiment of this application provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method for identifying gene phasing errors or a gene phasing method according to any of the above embodiments.

[0114] Another embodiment of this application provides a non-transitory computer-readable storage medium including a program that can be executed by a processor to implement the method for identifying gene phasing errors or the gene phasing method according to any of the above embodiments.

[0115] Unless otherwise specified, the experimental methods in the following embodiments are conventional methods, performed according to the techniques or conditions described in the literature in this field or according to the product manual. For example, unless otherwise specified, the process parameters in each step of the following embodiments are the default parameters of the corresponding software / script.

[0116] Example

[0117] 1. Test data and reference data

[0118] All test data were downloaded from MGIPCR-free, including:

[0119] The download address for the progeny (male) sequencing data HG002 (Fastq file) is:

[0120] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / data / AshkenazimTrio / HG002_NA24385_son / MGISEQ / PCR-free / NA24385 / MGISEQ2000_PCR-free_NA24385_V100002807_L03_1.fq.gz

[0121] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / data / AshkenazimTrio / HG002_NA24385_son / MGISEQ / PCR-free / NA24385 / MGISEQ2000_PCR-free_NA24385_V100002807_L03_2.fq.gz

[0122] The download address for the paternal sequencing data HG003 (Fastq file) is:

[0123] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / data / AshkenazimTrio / HG003_NA24149_father / MGISEQ / PCR-free / NA24149 / MGISEQ2000_PCR-free_NA24149_V100002807_L01_1.fq.gz

[0124] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / data / AshkenazimTrio / HG003_NA24149_father / MGISEQ / PCR-free / NA24149 / MGISEQ2000_PCR-free_NA24149_V100002807_L01_2.fq.gz

[0125] The download address for the maternal sequencing data HG004 (Fastq file) is:

[0126] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / data / AshkenazimTrio / HG004_NA24143_mother / MGISEQ / PCR-free / NA24143_1 / MGISEQ2000_PCR-free_NA24143_1_V100003043_L03_1.fq.gz

[0127] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / data / AshkenazimTrio / HG004_NA24143_mother / MGISEQ / PCR-free / NA24143_1 / MGISEQ2000_PCR-free_NA24143_1_V100003043_L03_2.fq.gz

[0128] The reference genome is version hg38. Download link:

[0129] ftp: / / ftp.ncbi.nlm.nih.gov / genomes / all / GCA / 000 / 001 / 405 / GCA_000001405.15_GRCh38 / seqs_for_alignment_pipelines.ucsc_ids / GCA_000001405.15_GRCh38_no_alt_analysis_set.fna.gz

[0130] The address for obtaining the phase truth set is:

[0131] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / release / AshkenazimTrio / HG002_NA24385_son / NISTv4.2.1 / GR Ch38 / SupplementaryFiles / HG002_GRCh38_1_22_v4.2.1_benchmark_hifiasm_v11_phasetransfer.vcf.gz

[0132] The address for obtaining the mutation truth set is:

[0133] https: / / ftp-trace.ncbi.nlm.nih.gov / giab / ftp / release / AshkenazimTrio / HG002_NA24385_son / NISTv4.2.1 / GR Ch38 / HG002_GRCh38_1_22_v4.2.1_benchmark.vcf.gz

[0134] 2. Testing Process

[0135] 2.1 Detection of variant sites

[0136] 2.1.1 Data Quality Control

[0137] Download test data HG002, HG003 and HG004. First, use FASTP[1] software to perform quality control on the sequencing data, including filtering out low-quality sequencing fragments and cutting out the barcode fragments of the sequencing pass sequence.

[0138] 2.1.2 Comparison

[0139] The Minimap2[2] software was used to compare the quality-controlled filtered data with the reference genome to obtain the alignment information of each sequencing fragment, including the alignment location and alignment quality.

[0140] 2.1.3 Mutation Detection

[0141] The variant detection tools GATK[3] and DeepVariant[4] were used to analyze variant sites based on the alignment results in order to identify and confirm variant sites and obtain relevant variant information, such as the quality of variants and the initial genotypes that have not been phased (e.g., 0 / 1, 1 / 0, 1 / 1, 1 / 2, etc.).

[0142] 2.2 Identifying the first erroneous variant site

[0143] Based on Mendel's laws of inheritance and using paternal and maternal data, the variations detected in section 2.1.3 were screened to identify and filter out variation sites that did not conform to the laws of inheritance. Figure 3 The situation shown is determined to be inconsistent with the laws of inheritance. Figure 3 In this text, "#N / A" indicates wild type; "0", "1", and "2" all represent different alleles involved in the mutant type. Figure 3 Taking the case shown in the first row as an example, if the father is 1 / 1 and the mother is 1 / 1, then according to the laws of inheritance, the offspring's locus can only be 1 / 1, but in reality it is 1 / 0. This type of locus will be considered a locus that does not conform to the laws of inheritance.

[0144] 2.3 Identifying the second erroneous variant site

[0145] 2.3.1 Identifying potential second error variant sites

[0146] Based on the variant information obtained in section 2.2, which excludes the first erroneous variant sites, several thresholds are set as evaluation criteria for the second erroneous variant sites. These include: the base quality value of the site must be greater than 10, the average alignment quality value of the sequencing fragment mapped to the site must be greater than 20, the quality value of the variant site must be greater than 20, the sequencing depth of the site must be greater than 10, and the allele frequency of the variant site must be greater than 0.3. Only sites that simultaneously meet all of the above conditions are considered high-confidence true positive sites (i.e., non-erroneous variant sites, TP); conversely, sites that do not meet these conditions are considered high-risk false positive sites (i.e., potential second erroneous variant sites, FP).

[0147] 2.3.2 Identifying the second error variant site

[0148] High-risk false positive sites (FPs) are further identified and determined by their association with high-confidence true positive sites (TPs). The determination criteria include two aspects: First, if both a high-risk false positive site and a high-confidence site are detected simultaneously in the same sequencing fragment, a strong association is considered to exist between these two sites; second, referring to an existing linkage disequilibrium (LD) database, the LD values ​​of two adjacent variant sites are checked; if the LD value exceeds 0.8, the two sites are considered to be strongly associated. High-risk false positive sites (FPs) with strong association are identified and confirmed as secondary false variant sites.

[0149] 2.4 Genome Phase

[0150] Based on the variant data excluding the first and second erroneous variant sites, the genome was phased using the HapCUT2[5] software in conjunction with the comparison results with pedigree data. That is, the overlapping regions between sequencing reads were linked into longer fragments. Heterozygous sites (0 / 1 and 1 / 0) were classified into alleles from the father or mother based on whether they could be completely covered by the same sequencing read, thus achieving phasing of the variant sites. Finally, the information of successfully phased fragments and the annotation information of variant sites were output. The successfully phased sites were marked as 0|1 or 1|0, while the annotation information of the unphased sites remained unchanged (e.g., Figure 4 (As shown).

[0151] 2.5 Evaluation of Gene Phased Results

[0152] To verify the reliability of the method in this embodiment, the results of the directional mutation sites are compared with the published phasing truth set and mutation truth set, the consistency between the two is calculated, and relevant evaluation indicators such as the conversion site ratio are output.

[0153] result

[0154] Figures 5A-7 The results obtained by the method proposed in the embodiments of this application are shown in the figure.

[0155] Figure 5A Examples of the methods proposed in this application identifying sites that do not conform to Mendelian inheritance laws (i.e., first error mutation sites) are shown. Figure 5A As shown, a mutation is indicated at position 125797766 on chromosome 3 of the offspring, with a genotype of heterozygous mutation 0 / 1. However, neither the father nor the mother carries this mutation, thus the offspring is deemed not to conform to Mendelian inheritance laws. Figure 5B The true value set of the variants shown indicates that this site is not included, proving that the site is indeed a FP, and the identification method proposed in this application is accurate.

[0156] Figure 6 An example is shown of the method proposed in this application for identifying false positive sites (i.e., second error variant sites) based on sequencing read association. For example... Figure 6 As shown, after threshold screening in section 2.3.1, the left locus was identified as a high-confidence true positive (TP) locus, and the right locus was identified as a high-risk false positive (FP) locus. Since FP and TP loci can be detected by the same sequencing fragment, the right locus FP was determined to be strongly correlated with its neighboring left locus TP, thus identifying the right locus FP as an erroneous variant and excluding it from subsequent phasing. (See corresponding reference.) Figure 6The true value set of the variants shown on the right shows that the right site FP is not included, proving that the site is indeed an FP, and the identification method proposed in this application is accurate.

[0157] Figure 7 An example is shown of the method proposed in this application for identifying false positive sites (i.e., second false variant sites) based on linkage disequilibrium. For example... Figure 7 As shown, after threshold screening in section 2.3.1, the right locus was identified as a high-confidence true positive (TP) locus, and the left locus was identified as a high-risk false positive (FP) locus. FP and TP loci cannot be detected by the same sequencing fragment, but by consulting the LD reference database (https: / / ldlink.nih.gov / ), it was found that their LD values ​​were greater than 0.8. Therefore, the left locus FP was identified as strongly correlated with the right locus TP, thus the left locus FP was identified as an erroneous variant and excluded from subsequent phasing. (Corresponding reference...) Figure 7 The true value set of the variants shown on the right shows that the left site FP is not included, proving that the site is indeed an FP, and the identification method proposed in this application is accurate.

[0158] The above results demonstrate that the method proposed in this application, by comprehensively considering genetic laws and setting multiple judgment thresholds, can accurately identify and eliminate variation detection errors and gene phasing errors introduced by sequencing or data analysis. Furthermore, this method does not require reference to or reliance on a truth set, thus avoiding false positive errors caused by the inaccurate truth sets in existing technologies. This achieves highly accurate gene phasing, especially demonstrating more precise judgment in complex genetic variations, false positive site identification, and differentiation of local phasing errors. Therefore, it has great application value in genetic association and disease susceptibility research.

[0159] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0160] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for identifying gene phasing errors to improve gene phasing accuracy, characterized in that, The method includes the following steps: 1) Based on family data, perform a first verification on individual variation information according to a first criterion to identify a first erroneous variation site, wherein the first criterion is Mendelian inheritance law, and the first erroneous variation site includes variation sites in the individual variation information that do not conform to Mendelian inheritance law; 2) According to the second criteria, the individual variation information is subjected to a second verification to identify potential second erroneous variant sites, wherein the second criteria include: quality value, sequencing depth and / or allele frequency, and the potential second erroneous variant sites include variant sites in the individual variation information that do not meet the second criteria; and 3) According to the third criterion, the potential second error variant sites are subjected to a third verification to determine the second error variant sites, wherein the third criterion includes the correlation between non-error variant sites and the potential second error variant sites, and the second error variant sites include variant sites among the potential second error variant sites that meet the third criterion.

2. The method according to claim 1, characterized in that, The family pedigree data includes paternal genotype data and maternal genotype data. For a specific mutation site in the individual, the following are examples of non-compliance with Mendelian inheritance laws: Both the paternal and maternal genotypes are wild-type, while the individual described is a mutant. Both the paternal and maternal genotypes are homozygous mutants, while the individual in question is a heterozygous mutant. The father's genotype is wild-type, the mother's genotype is homozygous mutant, and the individual in question is a homozygous mutant. The maternal genotype is wild-type, the paternal genotype is homozygous mutant, and the individual in question is homozygous mutant. The paternal genotype is wild-type, the maternal genotype is heterozygous mutant, and the individual is either homozygous mutant or compound heterozygous mutant; The individual is either a wild-type mother, a heterozygous mutant father, or a homozygous mutant or a compound heterozygous mutant; or a homozygous mutant father, a heterozygous mutant mother, or a compound heterozygous mutant mother. The mother's genotype is homozygous mutant, the father's genotype is heterozygous mutant, and the individual is a compound heterozygous mutant; or Both the paternal and maternal genotypes are heterozygous mutants, while the individual in question is a compound heterozygous mutant.

3. The method according to claim 1 or 2, characterized in that, The quality value includes one or more of the following: the base quality value of the variant site, the alignment quality value of the sequencing read containing the variant site, preferably the average alignment quality value and the variant quality value of the sequencing read containing the variant site.

4. The method according to claim 3, characterized in that, The second standard includes: The base mass value of the variant site is ≥10; The average alignment quality of the sequencing reads containing the aforementioned variant sites is ≥15, preferably ≥20; The mutation quality value of the mutation site is ≥10, preferably ≥20; The sequencing depth of the variant sites is ≥5, preferably ≥10; and The allele frequency of the variant site is ≥0.2, preferably ≥0.

3.

5. The method according to claim 4, characterized in that, The associations include sequencing read detection associations and / or linkage disequilibrium (LD) associations, and the third criterion includes: The potential second error variant site and the non-error variant site were detected in the same sequencing read; and / or The LD values ​​of the potential second error variant sites and the non-error variant sites described in the LD reference database. Optionally, the LD value between the potential second error mutation site and the non-error mutation site is ≥0.7, preferably ≥0.

8. Optionally, the potential second erroneous mutation site is adjacent to the non-erroneous mutation site; alternatively, the potential second erroneous mutation site is no more than the length of the sequencing read from the non-erroneous mutation site.

6. A gene phasing method, characterized in that, The method includes the following steps: 1) The method for identifying gene phasing errors according to any one of claims 1 to 5 is used to screen the individual variation information to identify the first erroneous variation site and the second erroneous variation site; 2) Exclude the first and second erroneous variant sites from the individual variant information; and 3) Gene phasing is performed based on individual variation information after excluding the first and second erroneous mutation sites.

7. The gene phasing method according to claim 6, characterized in that, The method further includes: Based on the sequencing data of the individual, the individual's variant information is obtained. Optionally, the sequencing data is second-generation sequencing data or third-generation sequencing data.

8. A gene phasing device, characterized in that, include: An error mutation site identification module is used to screen the individual variation information according to the method for identifying gene phasing errors according to any one of claims 1 to 5, so as to identify the first error mutation site and the second error mutation site; The erroneous variant site exclusion module is used to exclude the first erroneous variant site and the second erroneous variant site from the individual variant information; The gene phasing module is used to perform gene phasing based on individual variation information after excluding the first and second erroneous mutation sites. and An optional variant detection module is used to obtain variant information of the individual based on the individual's sequencing data.

9. An electronic device, comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method for identifying gene phasing errors according to any one of claims 1 to 5 or the gene phasing method according to claim 6 or 7.

10. A non-transitory computer-readable storage medium, characterized in that, The storage medium includes a program that can be executed by a processor to implement the method for identifying gene phasing errors as described in any one of claims 1 to 5 or the gene phasing method as described in claim 6 or 7.

Images