A method, system, and computer device for identifying single nucleotide polymorphisms
By combining short-read and long-read sequencing technologies to generate a composite variant dataset, the problem of low accuracy and sensitivity when using short-read or long-read sequencing technologies alone is solved, and more efficient single nucleotide polymorphism identification is achieved.
Patent Information
- Application Number
- CN202410463974.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-17
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2044-04-17
AI Technical Summary
Existing methods for identifying single nucleotide polymorphisms mainly rely on either short-read sequencing or long-read sequencing technologies, which leads to problems such as low detection accuracy or low detection sensitivity.
By combining short-read sequencing and long-read sequencing technologies, short-read variant datasets and long-read variant datasets are generated. Using short-read genotyping and long-read genotyping methods, a combined long-short-read variant dataset is generated to achieve accurate analysis of the sample data to be tested.
It improves the accuracy and sensitivity of single nucleotide polymorphism identification, balancing sequencing costs and detection efficiency.
Smart Images

Figure CN118447924B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of single nucleotide polymorphism (SNP) identification technology, specifically to a method, system, and computer equipment for SNP identification. Background Technology
[0002] Single nucleotide polymorphisms (SNPs) mainly refer to DNA sequence polymorphisms caused by variations in a single nucleotide at the genomic level. Single nucleotide polymorphism analysis refers to the genetic markers formed by variations in a single nucleotide at the genome. These markers are numerous and exhibit rich polymorphisms.
[0003] In related technologies, single nucleotide polymorphism (SNP) identification methods mainly employ long-read sequencing or short-read sequencing technologies. Long-read sequencing-based SNP identification methods are susceptible to the high error rate of long reads when sequencing sample depth is low, resulting in a high error rate for detected SNPs. While short-read sequencing technology offers higher accuracy and lower cost, its short read length (only a few hundred bases) makes it difficult to accurately align to repetitive regions of the genome. Therefore, short-read sequencing-based SNP identification methods have lower sensitivity for detecting SNPs in complex repetitive regions.
[0004] In summary, for detecting single nucleotide polymorphisms at the population level, it is necessary to balance sequencing cost and detection efficiency. Current methods for identifying single nucleotide polymorphisms mainly focus on using either short-read sequencing technology or long-read sequencing technology for analysis and identification, which suffers from low detection accuracy or low detection sensitivity. Summary of the Invention
[0005] In view of this, the present invention provides a method, apparatus and computer equipment for identifying single nucleotide polymorphisms, in order to solve the problems of low accuracy and low sensitivity of current single nucleotide polymorphism identification methods, which mainly focus on using only one sequencing technology, either short-read sequencing technology or long-read sequencing technology, for analysis and identification.
[0006] According to the first aspect, this embodiment provides a method for identifying single nucleotide polymorphisms, the method comprising:
[0007] Obtain the test sample data of the single nucleotide polymorphism identification object. The test sample data includes: test long read sample data and test short read sample data.
[0008] Using a short read variant detection method, a short read variant dataset is generated to identify single nucleotide polymorphisms in short read sample data.
[0009] Using long-read variant detection, a long-read variant dataset is generated to identify single nucleotide polymorphisms in short-read sample data.
[0010] Using the long read genotyping method, long read variant datasets are genotyped to obtain long read genotyping datasets;
[0011] Integrate short-read variant datasets and long-read genotyping datasets to generate combined short-read and long-read variant data for identifying test samples.
[0012] By implementing the above methods, and comprehensively utilizing short-read variant detection, long-read variant detection, and long-read genotyping, the generated long-read variant dataset and long-read genotyping dataset are integrated. This enables precise analysis of the sample data to be tested, thereby improving the accuracy of identifying single nucleotide polymorphisms in the sample data.
[0013] In one optional implementation, acquiring the test sample data of the single nucleotide polymorphism identification object includes:
[0014] Biological data generation software was used to generate test long-read sample data for single nucleotide polymorphism identification objects;
[0015] The initial short-read sample data was processed according to the short-read quality standard using biological data quality control software to generate short-read sample data to be tested for single nucleotide polymorphism identification.
[0016] By implementing the above methods, relatively accurate sample data for single nucleotide polymorphism identification can be obtained.
[0017] In one optional implementation, a short-read variant dataset for identifying single nucleotide polymorphisms in the sample data is generated using a short-read variant detection method, including:
[0018] The short read length sample data to be tested were compared using short read length comparison software, sorted using biological data sorting software, detected using biological data repetitive sequence detection software, and calibrated using base mass fraction calibration software.
[0019] Based on the calibrated short read length sample data to be tested, short read length gvcf files are generated using short read length text generation software.
[0020] The short read sample merging software was used to merge multiple short read sample data of the short read gvcf file containing the test data.
[0021] Genotyping of the merged short-read gvcf files was performed using short-read genotyping software;
[0022] Sample features were extracted from the short read sample data after genotyping. The sample features included: quality score divided by coverage depth, alignment quality rank sum test value, read position rank sum test value, strand specificity test value, alignment quality root mean square feature, strand specificity ratio test value, and coverage depth.
[0023] The quality score is divided by the coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, strand specificity test value feature, alignment quality root mean square feature, strand specificity ratio test value feature, and coverage depth feature, and then input into the single nucleotide polymorphism sensitivity calibration model to predict the sensitivity of the short read sample data to be tested after genotyping.
[0024] The sensitivity of the short-read sample data to be tested after genotyping is filtered to generate a short-read variant dataset with a sensitivity greater than or equal to a first preset threshold.
[0025] By implementing the above methods, short-read variation detection of the short-read sample data to be tested can improve the sensitivity and accuracy of identifying single nucleotide polymorphisms.
[0026] In one optional implementation, a long-read variant detection method is used to generate a long-read variant dataset for identifying single nucleotide polymorphisms in the long-read sample data to be tested, including:
[0027] Long read alignment software was used to compare the long read sample data to be tested using a high-quality sequencing method.
[0028] Generate long-read GVCF files using long-text generation software;
[0029] Long-read merging software was used to merge long-read GVCF files, and the merged long-read GVCF files were then genotyped.
[0030] After filtering the genotyping data of the long read samples to be tested, a long read variant dataset with a gene quality score greater than or equal to the second preset threshold is generated.
[0031] By implementing the above methods, long-read variation detection can be performed on the long-read sample data to be tested, enabling accurate analysis of the long-read sample data to be tested, thereby improving the accuracy of identifying single nucleotide polymorphisms in the long-read sample data to be tested.
[0032] In one optional implementation, long-read genotyping is used to perform long-read genotyping on the long-read variant dataset to obtain a long-read genotyping dataset, including:
[0033] Long read alignment software was used to align long read variant datasets using a low-quality sequencing method.
[0034] Genotyping of long-read variant datasets was performed using long-read genotyping software, and the first likelihood value of genotyping for long-read variant datasets was calculated.
[0035] Based on the long read variant dataset after genotyping, long read VCF files are generated using long read text generation software.
[0036] Based on Hardy Weinberg detection information, allele information, genotyping information, and genotyping missing information, a single filter can be used to extract data from multiple test samples contained in long-read VCF files.
[0037] Based on low-quality long read length information, long read length haplotype phasing software is used to obtain short read length variant datasets for haplotype phasing information.
[0038] Based on haplotype phasing information, the second likelihood value of genotyping for long-read variant datasets was calculated using long-read genotyping software;
[0039] Based on Hardy Weinberg detection information, allele information, and genotyping information, long-read variant datasets are filtered a second time.
[0040] The first and second likelihood values of genotyping are input into the genotyping correction software to correct the long-read variant dataset after secondary filtering, so as to generate a long-read genotyping dataset.
[0041] By implementing the above methods, the accuracy and sensitivity of single nucleotide polymorphism (SNP) identification can be significantly improved by utilizing long-read genotyping.
[0042] In one alternative implementation, short-read variant datasets and long-read genotyping datasets are integrated to generate a combined long-short-read variant dataset for identifying test sample data, including:
[0043] Biological sample comparison software compares short-read variant datasets and long-read genotyping datasets to obtain single nucleotide polymorphism shared datasets between the two datasets;
[0044] In the single nucleotide polymorphism shared dataset, calculate the short-read Hardy Weinberg detection values corresponding to the short-read variant dataset;
[0045] In the single nucleotide polymorphism shared dataset, calculate the long-read Hardy Weinberg detection values corresponding to the long-read variant dataset;
[0046] Calculate the identification ratio between the long-read Hardy Weinberg detection value and the short-read Hardy Weinberg detection value;
[0047] If the identification ratio is greater than the third preset threshold, the long read genotyping dataset will be used as the long and short read composite variant dataset.
[0048] If the identification ratio is less than or equal to the third preset threshold, the short read length variant dataset will be used as the long-short read length composite variant dataset.
[0049] By implementing the above methods, the complementary advantages between short-read sequencing and long-read sequencing improve the accuracy and sensitivity of single nucleotide polymorphism detection.
[0050] According to the second aspect, this embodiment provides a single nucleotide polymorphism identification system, the apparatus comprising:
[0051] The test sample data acquisition module is used to acquire test sample data of the single nucleotide polymorphism identification object;
[0052] The short read variation data detection module is used to generate a short read variation dataset for identifying single nucleotide polymorphisms in the sample data to be tested, using the short read variation detection method.
[0053] The long read variant data detection module is used to generate a long read variant dataset for identifying single nucleotide polymorphisms in the sample data to be tested using the long read variant detection method.
[0054] The long read data genotyping module is used to perform long read genotyping on long read variant datasets using the long read genotyping method, and obtain long read genotyping datasets;
[0055] The long and short read variant integration module is used to integrate the short read variant dataset and the long read genotyping dataset to generate a long and short read composite variant dataset for identifying the data of the sample to be tested.
[0056] According to a third aspect, this embodiment provides a computer device, including:
[0057] The memory and processor are interconnected and communicate with each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the single nucleotide polymorphism identification method in the first aspect or any embodiment of the first aspect.
[0058] According to the fourth aspect, this embodiment provides a computer-readable storage medium storing computer instructions for causing a computer to perform the single nucleotide polymorphism identification method of the first aspect or any embodiment of the first aspect.
[0059] According to the fifth aspect, this embodiment provides a computer program product, including computer instructions for causing a computer to perform the single nucleotide polymorphism identification method of the first aspect or any embodiment of the first aspect. Attached Figure Description
[0060] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0061] Figure 1 This is a structural block diagram of a single nucleotide polymorphism identification system according to an embodiment of the present invention;
[0062] Figure 2 This is a flowchart illustrating a method for identifying single nucleotide polymorphisms according to an embodiment of the present invention;
[0063] Figure 3 This is a flowchart illustrating another method for identifying single nucleotide polymorphisms according to an embodiment of the present invention;
[0064] Figure 4 This is a flowchart illustrating another method for identifying single nucleotide polymorphisms according to an embodiment of the present invention;
[0065] Figure 5 This is a flowchart illustrating another method for identifying single nucleotide polymorphisms according to an embodiment of the present invention;
[0066] Figure 6 This is a flowchart illustrating another method for identifying single nucleotide polymorphisms according to an embodiment of the present invention;
[0067] Figure 7 This is a flowchart illustrating another method for identifying single nucleotide polymorphisms according to an embodiment of the present invention;
[0068] Figure 8 This is a schematic diagram illustrating the evaluation of the sensitivity and accuracy of detectors for identifying single nucleotide polymorphisms.
[0069] Figure 9 This is a structural block diagram of another single nucleotide polymorphism identification device according to an embodiment of the present invention;
[0070] Figure 10 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation
[0071] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0072] To address the issues of low detection accuracy or sensitivity when using either short-read or long-read sequencing technologies alone to identify biological samples, this disclosure provides a method for identifying single nucleotide polymorphisms (SNPs). This method combines the advantages of both short-read and long-read sequencing technologies, achieving accurate identification of SNPs in biological samples while balancing sequencing costs and detection efficiency.
[0073] The single nucleotide polymorphism (SNP) identification method provided in this disclosure is applied to a single nucleotide polymorphism (SNP) identification system. For example... Figure 1 As shown, this single nucleotide polymorphism (SNP) identification system mainly includes: a short-read variant data detection module 11, a long-read variant data detection module 12, a long-read data genotyping module 13, and a long-short-read variant integration module 14. Specifically, the short-read variant data detection module 11 is the Illumina short-read variant detection module, the long-read variant data detection module 12 is the PacBioHiFi long-read variant detection module, and the long-read data genotyping module 13 is the PacBioZMW representative long-read genotyping module.
[0074] Among them, Figure 1 In the module 11, which is the Illumina short-read variant detection module, the following functions are included in sequence: read length alignment (BWA), variant detection (GATK HaplotypeCaller), population joint detection (GenotypeGVCF), candidate variant filtering (variable quality calibration failure, Hardy Weinberg detection over-heterozygote), haplotype phasing variant filtering (Hardy Weinberg detection p-value < 1E-6 missing rate > 0.05), read length-based haplotype phasing (WhatsHap), haplotype phasing variant filtering (Unphased by LRS and MAC = 1), and population-based haplotype phasing (Shapeit4). Among them, read length alignment (BWA) is further divided into confounding check VerifyBamID2 and sex check X / A coverage ratio.
[0075] Among them, Figure 1In the middle, the long read variation data detection module 12 includes read length alignment (pbmm2), variation detection (DeepVariant), population joint detection (DeepVariant), and candidate variation filtering (allele score <40 more alleles).
[0076] Among them, Figure 1 In the PacBio ZMW long-read variant genotyping module 13, the long-read data genotyping module includes read length alignment (pbmm2), sex check (X / A coverage ratio), candidate variant genotyping (WhatsHap), candidate variant filtering (missing rate >0.05 Hardy Weinberg detection for overly heterozygous genotype quality score <50), candidate variant regenotyping (WhatsHap), and genotype correction (Beagle4).
[0077] Among them, Figure 1 In the middle, the long and short read integration module includes the generation of long and short read integration variant sets (using the short read variant set as the basis, adding variants unique to the long read variant set, and determining the genotype of shared variants based on the ratio of Hardy Weinberg test p-values between the long read variant set and the short read variant set) and variant filtering (Merfine).
[0078] This disclosure provides a method for identifying single nucleotide polymorphisms. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.
[0079] This embodiment provides a method for identifying single nucleotide polymorphisms, which can be executed by computer equipment. Figure 2 This is a flowchart of a single nucleotide polymorphism identification method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:
[0080] Step S201: Obtain the test sample data of the single nucleotide polymorphism identification object. The test sample data includes the test long read sample data and the test short read sample data.
[0081] For example, the single nucleotide polymorphism (SNP) identification target can be common human DNA or human RNA biological samples, and the sample data to be tested can be multiple DNA or RNA biological samples from the human body. This sample data mainly includes, but is not limited to, DNA or RNA molecules extracted from materials such as cells, cell lines, blood, saliva, hair, tears, urine, feces, organoids, ground organ tissue samples, and ground tumor tissue samples. In this embodiment, the sample data to be tested includes long-read sample data and short-read sample data. The long-read sample data is preprocessed to meet certain accuracy requirements using HiFi-reads and ZMW to represent long-read data, while the short-read sample data is preprocessed to meet certain accuracy requirements using Illumina short-read data. Therefore, the long-read sample data mentioned above represents HiFi-reads and ZMW to represent long-read data, and the short-read sample data represents Illumina short-read data.
[0082] In one alternative implementation, such as Figure 3 As shown, step S201 above, obtaining the test sample data of the single nucleotide polymorphism identification object, includes:
[0083] Step S301: Use biological data generation software to generate test long read sample data of the single nucleotide polymorphism identification object.
[0084] For example, the biological data generation software can be CCS software. CCS software is used to generate PacBio HiFi-reads and ZMW representative long-read data based on representative read lengths. For instance, the `--all` option in CCS software outputs all representative long reads of ZMWs, regardless of sequencing repeat counts and read quality, selecting reads with an accuracy ≥ 0.99 from the successfully corrected representative reads as HiFi long-read data. In this embodiment, the biological data generation software CCS generates HiFi-reads and ZMW representative long reads that meet certain accuracy requirements, where HiFi-reads are high-quality long-read sequencing data, and ZMW representative long reads are low-quality long-read sequencing data.
[0085] By performing step S301 above, HiFi long reads and ZMW long reads can be generated for downstream analysis, which improves accuracy compared to obtaining raw long read sequencing data.
[0086] Step S302: The initial short-read sample data is processed according to the short-read quality standard using biological data quality control software to generate short-read sample data to be tested for single nucleotide polymorphism identification.
[0087] For example, biological data quality control software can be FastP software. Initial short-read sample data processed by FastP software according to the short-read quality standard (Illumina short-read quality standard) is processed, i.e., Illumina short reads are processed. This processing includes excising adapter sequences, excising low-quality sliding window fragments, and filtering low-quality reads.
[0088] By performing step S302, low-quality short reads can be filtered out, and the accuracy is improved compared to the initial short read sample data acquisition. In this embodiment, steps S301-S302 are not in any particular order.
[0089] Step S202: Using the short read variation detection method, a short read variation dataset is generated to identify the single nucleotide polymorphism of the short read sample data to be tested.
[0090] Specifically, step S202 mainly involves... Figure 1 The Illumina short read mutation detection module is used in the dataset, therefore, the short read mutation dataset can be the Illumina short read mutation dataset.
[0091] Because short-read sequencing technology is highly accurate and inexpensive, this embodiment introduces a short-read variant detection method to generate a short-read variant dataset, in order to ensure the accuracy of the sample data and also to balance sequencing costs.
[0092] This embodiment provides another method for identifying single nucleotide polymorphisms. Figure 4 This is a flowchart of another method for identifying single nucleotide polymorphisms according to embodiments of this disclosure, such as... Figure 4 As shown, step S202 above utilizes a short-read variant detection method to generate a short-read variant dataset for identifying single nucleotide polymorphisms in the short-read sample data to be tested. This process includes the following steps:
[0093] Step S401: The short read length sample data to be tested is compared using short read length comparison software, sorted using biological data sorting software, detected using biological data repetitive sequence detection software, and calibrated using base mass fraction calibration software.
[0094] For example, the short read alignment software can be BWA-MEM software, the biological data sorting software can be SAMtools software, the biological data repetitive sequence detection software can be GATKMarkDuplicates software, and the base quality fraction calibration software can be GATKBaseRecalibrator software, ApplyBQSR software, eRecalibrator software, and ApplyBQSR software. For instance, the first step uses BWA-MEM software to align Illumina short reads to the reference genome to obtain a BAM file; the second step uses SAMtools software to sort the BAM file; the third step uses GATKMarkDuplicates software to detect repetitive sequences generated by polymerase chain reaction (PCR) in the sorted BAM file; and the fourth step uses GATKBaseRecalibrator software and ApplyBQSR software to calibrate the base quality fraction of the short read sample data after repetitive sequence detection.
[0095] Among them, the BWA-MEM software can obtain the accurate genomic location of short reads by comparing them with Illumina short reads, which can be used to compare the differences between short reads and genomic sequences, thereby identifying single nucleotide polymorphisms.
[0096] The SAMtools software sorts BAM files so that they can be arranged according to genomic order, facilitating subsequent analysis.
[0097] The purpose of GATKMarkDuplicates software is to detect duplicates and filter out short duplicate reads in subsequent analysis, preventing reads from being counted as duplicates when identifying single nucleotide polymorphisms, thereby improving the accuracy of single nucleotide polymorphism identification.
[0098] The GATKBaseRecalibrator and ApplyBQSR software can calibrate the quality of read lengths based on alignment information, thereby improving the accuracy of identifying single nucleotide polymorphisms.
[0099] In this embodiment, the software mentioned above are all relatively mature software developed in biology. In this embodiment, multiple different software are used in combination to process the short-read sample data to be tested. Compared with using only one software, the measurement accuracy and precision of the short-read sample data to be tested can be fully guaranteed.
[0100] Step S402: Based on the calibrated short read length sample data to be tested, generate a short read length gvcf file using short read length text generation software.
[0101] For example, the short read length text generation software can be GATKhaplotypeCallerGVCF software. For instance, by inputting the short read length sample data to be tested, which contains multiple sample numbers, into GATKhaplotypeCallerGVCF software, gvcf files for each sample of the short read length sample data to be tested can be obtained.
[0102] In this embodiment, the short read GVCF file is used to obtain the potential single nucleotide polymorphism (SNP) information of each sample in the short read sample data to be tested. Subsequently, the potential SNP information of all samples can be merged, and the information of all samples can be comprehensively utilized to identify SNPs at the population level, thereby improving the accuracy of SNP identification.
[0103] Step S403: Use short read sample merging software to merge multiple short read sample data to be tested contained in the short read gvcf file.
[0104] For example, the short-read sample merging software can be GATK's GenomicsDBImport software. For instance, GATK's GenomicsDBImport software can be used to generate a gvcf file for each sample. Merging gvcf files using GATK's GenomicsDBImport software aims to combine the latent single nucleotide polymorphism (SNP) information from all samples in the short-read sample data to be tested. By comprehensively utilizing the information from all samples, SNPs can be identified at the population level, thereby improving the accuracy of SNP identification.
[0105] Step S404: Genotyping the merged short-read gvcf file using short-read genotyping software.
[0106] For example, the short-read genotyping software can be GATKGenotypeGVCFs software, which can be used to perform joint genotyping of all samples.
[0107] The GATKGenotypeGVCF software is designed to comprehensively utilize information from all samples in the short-read sample data to be tested, thereby facilitating the identification of single nucleotide polymorphisms at the population level.
[0108] Step S405: Extract sample features from the short read sample data after genotyping. The sample features include: quality score divided by coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, strand specificity test value feature, alignment quality root mean square feature, strand specificity ratio test value feature, and coverage depth feature.
[0109] For example, the quality score divided by the coverage depth feature is abbreviated as QD, the alignment quality rank sum test value is abbreviated as MQRankSum, the read length position rank sum test value is abbreviated as ReadPosRankSum, the chain specificity Fisher test value is abbreviated as FS, the alignment quality root mean square test value is abbreviated as MQ, the chain specificity ratio test value is abbreviated as SOR, and the coverage depth is abbreviated as DP.
[0110] The sample features extracted in this embodiment can be used to distinguish between correct and incorrect single nucleotide polymorphisms (SNPs). Specifically, QD reflects the quality score after correcting for coverage depth; correct SNPs have a higher QD score than incorrect SNPs. MQRankSum reflects the alignment quality of the read at the SNP site; correct SNPs have a higher MQRankSum score than incorrect SNPs. MQ reflects the alignment quality of the read at the SNP site; correct SNPs have a higher MQ score than incorrect SNPs. SOR reflects the chain specificity of the read at the SNP site; correct SNPs have a higher SOR score than incorrect SNPs. DP reflects the coverage depth of the SNP site; correct SNPs have a higher DP score than incorrect SNPs.
[0111] In this embodiment, the feature parameters can be set by setting parameters during the feature extraction process.
[0112] Step S406: Divide the quality score by the coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, strand specificity test value feature, alignment quality root mean square feature, strand specificity ratio test value feature, and coverage depth feature, and input them into the single nucleotide polymorphism sensitivity calibration model to predict the sensitivity of the short read sample data to be tested after genotyping.
[0113] For example, the single nucleotide polymorphism (SNP) sensitivity calibration model here can be a model generated by GATKVariantRecalibrator software and ApplyVQSR software. During model training, the maximum number of Gaussian clusters can be set, and this parameter value can be 6. In actual prediction, all the above feature values are simultaneously input into the SNP sensitivity calibration model, and the predicted results of the sensitivity of the short-read sample data after genotyping are output.
[0114] The single nucleotide polymorphism sensitivity calibration model in this embodiment mainly uses the Gaussian mixture model. By inputting all the feature parameters into this model, it is possible to effectively distinguish between correct and incorrect single nucleotide polymorphisms.
[0115] Step S407: Filter the sensitivity of the short-read sample data to be tested after genotyping, and generate a short-read variant dataset with sensitivity greater than or equal to the first preset threshold.
[0116] For example, this step is equivalent to the filtering step in a short-read variant detection method. For instance, in this filtering step, single nucleotide polymorphisms (SNPs) with a quality calibration sensitivity level greater than 99.5% are filtered out. Then, SNPs with a Hardy Weinberg detection p-value less than 10⁻⁶ and indicating an over-heterozygous direction are filtered out, generating the Illumina short-read variant set. Therefore, the first preset threshold can be a sensitivity level of 99.5%. In practical applications, this first preset threshold can be flexibly set according to the actual application scenario. In this embodiment, by executing step S407, erroneous SNPs can be effectively filtered out.
[0117] Step S203: Using the long read variant detection method, a long read variant dataset is generated to identify single nucleotide polymorphisms in the sample data to be tested.
[0118] Specifically, step S202 mainly involves... Figure 1 The long read mutation detection module in PacBioHiFi is used for this purpose; therefore, the long read mutation dataset can be the PacBioHiFi long read dataset.
[0119] Since single nucleotide polymorphism identification methods in long-read sequencing are susceptible to high error rates when sequencing sample depth is low, this embodiment introduces the PacBioHiFi long-read dataset, a high-quality long-read sequencing dataset, for long-read variant detection.
[0120] In this embodiment, step S203 allows for the identification of single nucleotide polymorphisms using long read information from all samples in the population, thereby obtaining more accurate potential single nucleotide polymorphism sites.
[0121] This embodiment provides another method for identifying single nucleotide polymorphisms. Figure 5 This is a flowchart of another method for identifying single nucleotide polymorphisms according to embodiments of this disclosure, such as... Figure 5 As shown, step S203 above utilizes a long-read variant detection method to generate a long-read variant dataset for identifying single nucleotide polymorphisms in the long-read sample data to be tested. This process includes the following steps:
[0122] Step S501: Use long read alignment software to compare the long read sample data to be tested in a high-quality sequencing manner.
[0123] For example, the long read alignment software can be pbmm2 software, and the long read sample data to be tested here are preprocessed HiFi reads. In this embodiment, the high-quality sequencing method is PacBioHiFi long read method. For example, pbmm2 software is used to align the PacBioHiFi long read data to the reference genome, and the long read alignment parameters (presentCCS) are set in pbmm2 software.
[0124] In this embodiment, the pbmm2 software can be used to compare HiFi long reads to obtain the accurate genomic location of short reads, which can be used to compare the differences between long reads and genomic sequences to identify single nucleotide polymorphisms.
[0125] Step S502: Use long-read long text generation software to generate a long-read long gvcf file.
[0126] For example, the long-read text generation software can be DeepVariant software. For instance, DeepVariant software can be used to detect candidate variants and generate gvcf files for each sample in the long-read sample data to be tested, with the minimum supported read length set to 1.
[0127] In this embodiment, the purpose of generating long-read gvcf files is to obtain the potential single nucleotide polymorphism (SNP) information of each sample in the long-read sample data to be tested. Subsequently, the potential SNP information of all samples can be merged, and the information of all samples can be comprehensively utilized to identify SNPs at the population level, thereby improving the accuracy of identifying SNPs.
[0128] Step S503: Use long read merging software to merge long read gvcf files and perform genotyping on the merged long read gvcf files.
[0129] For example, the long read merging software can be GLnexus software, such as using GLnexus software to merge the gvcf files of all samples and perform joint genotyping.
[0130] In this embodiment, the use of GLnexus software to merge long read GVCF files is to combine the potential single nucleotide polymorphism (SNP) information of all samples, comprehensively utilize the information of all samples, identify SNPs at the population level, and thus improve the accuracy of SNP identification.
[0131] Step S504: Filter the long read sample data to be tested after genotyping, and generate a long read variant dataset with a gene quality score greater than or equal to the second preset threshold.
[0132] For example, based on the long-read sample data to be tested after genotyping, single nucleotide polymorphisms with an allele quality score of less than 40 are filtered out, and bi-allelic single nucleotide polymorphisms are selected as candidate variants for subsequent genotyping. Therefore, the second preset threshold can be 40. A gene quality score greater than or equal to 40 means that the processing is qualified, and a long-read variant dataset that meets the quality requirements is finally generated.
[0133] Therefore, the single nucleotide polymorphism identification method in this embodiment, by executing the above steps S501-S504, can achieve accurate analysis of the long read sample data to be tested, thereby improving the accuracy of identifying the single nucleotide polymorphism of the long read sample data to be tested.
[0134] Step S204: Using the long read genotyping method, perform long read genotyping on the long read variant dataset to obtain the long read genotyping dataset.
[0135] Specifically, step S204 mainly involves... Figure 1 The PacBioHiFi long-read genotyping module is used in this embodiment, therefore, this long-read genotyping dataset can be ZMW long-read genotyping data. In this embodiment, the long-read genotyping dataset mainly represents low-quality sequencing datasets. If long-read sequencing is used alone during the entire sequencing process in the traditional way, not only will the excessive amount of high-quality long-read data increase costs, but it will also be difficult to guarantee accuracy. If short-read sequencing is used alone during the entire sequencing process in the traditional way, it is difficult to detect single nucleotide polymorphisms in complex repetitive regions with high sensitivity. Therefore, in this embodiment, the long-read genotyping dataset is used to fully utilize sequencing information and ultimately improve sensitivity and accuracy.
[0136] This embodiment provides another method for identifying single nucleotide polymorphisms. Figure 6 This is a flowchart of another method for identifying single nucleotide polymorphisms according to embodiments of this disclosure, such as... Figure 6 As shown, step S204 above uses long-read genotyping to perform long-read genotyping on the long-read variant dataset to obtain a long-read genotyping dataset. This process includes the following steps:
[0137] Step S601: Use long read alignment software to align the long read variant dataset using a low-quality sequencing method.
[0138] For example, the long read alignment software can be pbmm2 software, and the low-quality sequencing method here refers to long read data represented by PacBioZMW. For instance, pbmm2 software is used to align the long read data represented by PacBioZMW to a reference genome, and the long read alignment parameter is set to `presentSUBREAD`. In this embodiment, based on the low-quality sequencing method, genotyping using ZMW long reads can fully utilize sequencing information, thereby enhancing the sensitivity and accuracy of single nucleotide polymorphism (SNP) identification.
[0139] Step S602: Use long-read genotyping software to genotype the long-read variant dataset and calculate the first likelihood value of the genotyping of the long-read variant dataset.
[0140] For example, long-read genotyping software can be WhatsHap or Longshot. For instance, WhatsHap can be used to genotype candidate variants and calculate the likelihood value of the genotype. This embodiment can also use Longshot software for genotyping candidate variants.
[0141] In this embodiment, calculating the first likelihood value of genotyping can obtain the probability that a sample has a single nucleotide polymorphism. This first likelihood value of genotyping can be used as input information for BEAGLE4 to correct the genotype, further improving the accuracy of identifying single nucleotide polymorphisms.
[0142] Step S603: Based on the long read variant dataset after genotyping, generate long read VCF files using long read text generation software.
[0143] For example, long-read text generation software can be BCFtoolsmerge software, such as using BCFtoolsmerge to merge long-read VCF files of all test samples.
[0144] In this embodiment, a long read VCF file is generated to merge the potential single nucleotide polymorphism (SNP) information of all samples, comprehensively utilize the information of all samples, identify SNPs at the population level, and thus improve the accuracy of SNP identification.
[0145] Step S604: Based on Hardy Weinberg detection information, allele information, genotyping information, and genotyping missing information, filter long-read VCF files containing multiple numbers of test sample data in one step.
[0146] For example, variant sites in the test sample data are filtered out according to the following criteria: for example, in long read VCF files, test sample data with a Hardy Weinberg test p-value less than 10⁻⁶ and an over-heterozygous direction are filtered out; the number of alleles is 0; the median genotype score is less than 50; the genotype deletion rate is greater than 5%; and the Whatshap genotype and Longshot genotype identity is less than 90%.
[0147] In this embodiment, step S604 is executed to achieve the purpose of filtering out variations in the data of the sample to be tested.
[0148] Step S605: Based on low-quality long read information, use long read genotyping software to obtain short read variant datasets for haplotype phasing information.
[0149] For example, low-quality long read information is ZMW long read information.
[0150] The long-read genotyping software is still Whatshap software. For example, using ZMW long-read information, Whatshap software is used to perform haplotype phasing on the Illumina short-read variant set.
[0151] In this embodiment, because the Illumina short read variant set is more accurate than the HiFi long read variant set, step S605 uses the Illumina variant set to obtain sample haplotype phasing. This haplotype phasing can improve the accuracy of calculating genotyping likelihood values, thereby further improving the accuracy and sensitivity of single nucleotide polymorphism identification.
[0152] Step S606: Based on haplotype phasing information, calculate the second likelihood value of the genotyping of the long-read variant dataset using long-read genotyping software.
[0153] For example, the long-read genotyping software can be the WhatsHap software mentioned above. For instance, using haplotype phasing information from the Illumina short-read variant set, WhatsHap software can be used to calculate the genotype likelihood value again. This genotype likelihood value is the second likelihood value for genotyping mentioned above. Here, in this embodiment, the calculation of the second likelihood value for genotyping considers the sample haplotype orientation information. Haplotype phasing can improve the accuracy of calculating the genotyping likelihood value, thereby further improving the accuracy and sensitivity of single nucleotide polymorphism (SNP) identification.
[0154] Step S607: Based on Hardy Weinberg detection information, allele information, and genotyping information, the long read variant dataset is filtered a second time.
[0155] For example, candidate variant sites are filtered out according to the following criteria: Hardy Weinberg test p-value less than 10⁻⁶ and in the direction of excessive heterozygosity; allele count of 0; median genotype score less than 50.
[0156] In this embodiment, secondary filtration can further improve the accuracy and sensitivity of single nucleotide polymorphism identification.
[0157] Step S608: Input the first likelihood value and the second likelihood value of genotyping into the genotyping correction software to correct the long read variant dataset after secondary filtering, so as to generate a long read genotyping dataset.
[0158] For example, the genotyping correction software can be BEAGLE4 software. For instance, using BEAGLE4 software, the first likelihood value and the second likelihood value of the genotyping described above are input into the genotyping correction software. The linkage disequilibrium information of single nucleotide polymorphisms in the population is used to correct the genotype, filter out single nucleotide polymorphisms with an allele count of 0, and generate the PacBio long read genotyping dataset.
[0159] In this embodiment, based on long read variant detection, while ensuring sequencing sensitivity, sequencing cost and sequencing accuracy also need to be considered. Therefore, based on low-quality sequencing, using ZMW long reads for genotyping can make full use of sequencing information, thereby enhancing the sensitivity and accuracy of single nucleotide polymorphism identification.
[0160] Step S205: Integrate the short read variant dataset and the long read genotyping dataset to generate a long and short read composite variant dataset for identifying the test sample data.
[0161] Current methods for identifying single nucleotide polymorphisms (SNPs) mainly focus on analyzing and identifying them using either short-read or long-read sequencing technologies alone. There is a lack of analytical workflows that can integrate short-read and long-read sequencing data to identify SNPs at the population level, so as to fully leverage the complementary advantages of the two sequencing data types and improve the accuracy and sensitivity of SNP detection.
[0162] In this field, long-read sequencing technology is typically used alone to identify single nucleotide polymorphisms (SNPs) throughout the entire process, while neglecting to further genotypify long-read genes using low-quality sequencing methods. This leads to increased sequencing costs and reduced measurement accuracy. However, this embodiment integrates short-read variant detection and long-read variant detection on the premise of long-read gene genotyping, thereby fully leveraging the advantages of both long-read and short-read sequencing methods. Ultimately, this improves the accuracy and sensitivity of SNP detection and also achieves a balance in sequencing costs.
[0163] For example, this embodiment uses Figure 1 The long and short read variant set integration module integrates the short read variant dataset and the long read genotyping dataset.
[0164] This embodiment provides another method for identifying single nucleotide polymorphisms. Figure 7 This is a flowchart of another method for identifying single nucleotide polymorphisms according to embodiments of this disclosure, such as... Figure 7 As shown, step S205 above integrates the short-read variant dataset and the long-read genotyping dataset to generate a long-short-read composite variant dataset for identifying the sample data to be tested. This process includes the following steps:
[0165] Step S701: In the single nucleotide polymorphism shared dataset, calculate the short-read Hardy Weinberg detection value corresponding to the short-read variant dataset.
[0166] For example, the short read mutation dataset is the Illumina short read mutation dataset, and the corresponding short read Hardy Weinberg detection value is P1.
[0167] Step S702: In the single nucleotide polymorphism shared dataset, calculate the long read Hardy Weinberg detection value corresponding to the long read variant dataset.
[0168] For example, the long read mutation dataset is the PacBio long read mutation dataset, and the corresponding short read Hardy Weinberg detection value for this long read mutation dataset is P2.
[0169] Step S703: Calculate the identification ratio between the long-read Hardy Weinberg detection value and the short-read Hardy Weinberg detection value.
[0170] For example, the identification ratio can be expressed as W, where W = P1 / P2.
[0171] Step S704: If the identification ratio is greater than the third preset threshold, then the long read genotyping dataset is used as the long and short read composite variant dataset.
[0172] For example, the third preset threshold can be 1000, and the specific value of the third preset threshold can be set according to the actual scenario. If W>1000, it means that the long read genotyping dataset can be used as a long and short read composite variant dataset.
[0173] Step S705: If the identification ratio is less than or equal to the third preset threshold, then the short read length variant dataset is taken as the long and short read length composite variant dataset.
[0174] For example, if W≦1000, it means that the long read genotyping dataset can be used as a long and short read composite variant dataset.
[0175] like Figure 8 The diagram shown illustrates the evaluation of the detector sensitivity and accuracy of the single nucleotide polymorphism (SNP) identification method in this embodiment.
[0176] exist Figure 8 In this study, the accuracy and sensitivity of single nucleotide polymorphism detection were evaluated using the GIAB baseline variant sets (HG002, HG005) and multiple high-quality baseline sample datasets (1KCP).
[0177] 1. In the GIAB and 1KCP benchmark variant sets, the final long-short read integrated variant set showed a significant improvement in recall compared to the Illumina short read variant set, while maintaining similar precision. The recall of HG002 increased from 93.89% to 97.01%, and the precision increased from 99.30% to 99.28%; the recall of HG005 increased from 93.91% to 97.21%, and the precision increased from 99.03% to 99.04%; the average recall of 1KCP increased from 96.91% to 97.86%, while the average precision remained unchanged at 99.60%.
[0178] 2. Evaluation of different genomic regions in the GIAB variant set revealed that the improved recall rate of the long and short read integration variant set was more pronounced in difficult genomic regions, with the recall rate of HG002 increasing from 78.18% to 92.42% and the recall rate of HG005 increasing from 77.30% to 93.29%.
[0179] Therefore, the single nucleotide polymorphism (SNP) identification method in this embodiment can significantly improve the sensitivity and accuracy of SNP identification.
[0180] This embodiment also provides a single nucleotide polymorphism identification system, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.
[0181] This disclosure provides a single nucleotide polymorphism identification system, such as... Figure 9 As shown, the device includes:
[0182] The test sample data acquisition module 91 is used to acquire test sample data of the single nucleotide polymorphism identification object;
[0183] The short read variation data detection module 92 is used to generate a short read variation dataset for identifying single nucleotide polymorphisms in the sample data to be tested by using the short read variation detection method.
[0184] The long read variant data detection module 93 is used to generate a long read variant dataset for identifying single nucleotide polymorphisms in the sample data to be tested by using the long read variant detection method.
[0185] The long read data genotyping module 94 is used to perform long read genotyping on long read variant datasets using the long read genotyping method to obtain long read genotyping datasets;
[0186] The long and short read variation integration module 95 is used to integrate the short read variation dataset and the long read genotyping dataset to generate a long and short read composite variation dataset for identifying the data of the sample to be tested.
[0187] In one optional implementation, the sample data acquisition module 91 includes:
[0188] The long-read sample data acquisition submodule is used to generate long-read sample data of single nucleotide polymorphism identification objects using biological data generation software;
[0189] The short-read sample data acquisition submodule is used to process the initial short-read sample data according to the short-read quality standard using biological data quality control software, and generate short-read sample data to be tested for single nucleotide polymorphism identification objects.
[0190] In one optional implementation, the short read variation data detection module 92 includes:
[0191] The short read sample data processing submodule is used to sequentially compare the short read sample data to be tested using short read alignment software, sort the short read sample data to be tested using biological data sorting software, detect the short read sample data to be tested using biological data repetitive sequence detection software, and calibrate the short read sample data to be tested using base mass fraction calibration software.
[0192] The short read length data file generation submodule is used to generate short read length gvcf files based on the calibrated short read length sample data to be tested using short read length text generation software.
[0193] The short read length sample data merging submodule is used to merge multiple short read length sample data to be tested contained in a short read length gvcf file using short read length sample merging software.
[0194] The short-read data genotyping submodule is used to perform genotyping on the merged short-read gvcf files using short-read genotyping software.
[0195] The short read sample data extraction submodule is used to extract sample features from the short read sample data to be tested after genotyping. The sample features include: quality score divided by coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, chain specificity test value feature, alignment quality root mean square feature, chain specificity ratio test value feature, and coverage depth feature.
[0196] The short-read sample data prediction submodule is used to divide the quality score by the coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, strand specificity test value feature, alignment quality root mean square feature, strand specificity ratio test value feature, and coverage depth feature and input them into the single nucleotide polymorphism sensitivity calibration model to predict the sensitivity of the short-read sample data to be tested after genotyping.
[0197] The short-read sample data generation submodule is used to filter the sensitivity of the short-read sample data to be tested after genotyping, and generate a short-read variant dataset with a sensitivity greater than or equal to a first preset threshold.
[0198] In one optional implementation, the long read variation data detection module 93 includes:
[0199] The long read data alignment submodule is used to compare the long read sample data to be tested using long read alignment software in a high-quality sequencing manner.
[0200] The long read file generation submodule is used to generate long read gvcf files using long read text generation software;
[0201] The long read file merging submodule is used to merge the long read gvcf files using long read merging software and perform genotyping on the merged long read gvcf files;
[0202] The long read data filtering submodule is used to filter the long read sample data to be tested after genotyping, and generate a long read variant dataset with a gene quality score greater than or equal to the second preset threshold.
[0203] In one optional implementation, the short-read data genotyping submodule includes:
[0204] The long read data alignment submodule is used to align long read variant datasets using long read alignment software in a low-quality sequencing manner.
[0205] The long-read genotyping submodule is used to perform genotyping on long-read variant datasets using long-read genotyping software and to calculate the first likelihood value of the genotyping of long-read variant datasets.
[0206] The long read file generation submodule is used to generate long read VCF files based on the long read variant dataset after genotyping using long read text generation software.
[0207] The long read data filtering submodule is used to filter long read VCF files containing multiple numbers of test sample data based on Hardy Weinberg detection information, allele information, genotyping information, and genotyping missing information.
[0208] The haplotype phasing acquisition submodule is used to obtain short-read variant datasets based on low-quality long-read information and to perform haplotype phasing information using long-read genotyping software.
[0209] The empirical likelihood calculation submodule is used to calculate the second likelihood value of the genotyping of long-read variant datasets based on haplotype phasing information using long-read genotyping software;
[0210] The long read data secondary filtering submodule is used to filter long read variant datasets based on Hardy Weinberg detection information, allele information, and genotyping information.
[0211] The long read data correction submodule is used to input the first and second likelihood values of genotyping into the genotyping correction software to correct the long read variant dataset after secondary filtering, so as to generate a long read genotyping dataset.
[0212] In one optional implementation, the long and short read length variation integration module 95 includes:
[0213] The single nucleotide polymorphism information acquisition submodule is used by biological sample comparison software to compare short-read variant datasets and long-read genotyping datasets to obtain the single nucleotide polymorphism shared dataset between the short-read variant dataset and the long-read genotyping dataset;
[0214] The short-read Hardy Weinberg detection value calculation submodule calculates the short-read Hardy Weinberg detection value corresponding to the short-read variant dataset in the single nucleotide polymorphism shared dataset;
[0215] The long-read Hardy Weinberg detection value calculation submodule is used to calculate the long-read Hardy Weinberg detection value corresponding to the long-read variant dataset in the single nucleotide polymorphism shared dataset;
[0216] The identification ratio calculation submodule is used to calculate the identification ratio between the long-read Hardy Weinberg detection value and the short-read Hardy Weinberg detection value.
[0217] The first determination submodule for long and short read composite variant data is used to use the long read genotyping dataset as the long and short read composite variant dataset if the identification ratio is greater than the third preset threshold.
[0218] The second determination submodule for long and short read length composite variant data is used to treat the short read length variant dataset as the long and short read length composite variant dataset if the identification ratio is less than or equal to a third preset threshold.
[0219] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.
[0220] The single nucleotide polymorphism identification system in this embodiment is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.
[0221] This invention also provides a computer device having the above-described features. Figure 10 The single nucleotide polymorphism identification system shown.
[0222] Please see Figure 10 , Figure 10 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 10 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 10 Take a processor 10 as an example.
[0223] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.
[0224] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.
[0225] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0226] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.
[0227] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.
[0228] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.
[0229] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.
[0230] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.
Claims
1. A method for identifying single nucleotide polymorphisms, characterized in that, The method includes: Acquire test sample data for single nucleotide polymorphism identification objects, wherein the test sample data includes: test long read sample data and test short read sample data; Using a short read variant detection method, a short read variant dataset is generated to identify the single nucleotide polymorphisms in the short read sample data to be tested; Using a long-read variant detection method, a long-read variant dataset is generated to identify the single nucleotide polymorphisms in the short-read sample data to be tested; The long read genotyping dataset is obtained by performing long read genotyping on the long read variant dataset using the long read genotyping method. The short read variant dataset and the long read genotyping dataset are integrated to generate a long and short read composite variant dataset for identifying the test sample data; The process of integrating the short-read variant dataset and the long-read genotyping dataset to generate a long-short-read composite variant dataset for identifying the test sample data includes: Biological sample comparison software compares the short-read variant dataset and the long-read genotyping dataset to obtain a single nucleotide polymorphism shared dataset between the short-read variant dataset and the long-read genotyping dataset; In the single nucleotide polymorphism shared dataset, calculate the short-read Hardy Weinberg detection value corresponding to the short-read variant dataset; In the single nucleotide polymorphism shared dataset, calculate the long-read Hardy Weinberg detection value corresponding to the long-read variant dataset; Calculate the identification ratio between the long-read Hardy Weinberg detection value and the short-read Hardy Weinberg detection value; If the identification ratio is greater than the third preset threshold, then the long read genotyping dataset is used as the long and short read composite variant dataset. If the identification ratio is less than or equal to the third preset threshold, then the short read length variant dataset is taken as the long-short read length composite variant dataset.
2. The method for identifying single nucleotide polymorphisms according to claim 1, characterized in that, The acquisition of test sample data for single nucleotide polymorphism identification includes: Biological data generation software was used to generate the target long-read sample data for single nucleotide polymorphism identification. The initial short-read sample data was processed according to the short-read quality standard using biological data quality control software to generate the short-read sample data to be tested for single nucleotide polymorphism identification.
3. The method for identifying single nucleotide polymorphisms according to claim 1 or 2, characterized in that, Using a short-read variant detection method, a short-read variant dataset is generated to identify single nucleotide polymorphisms in the test sample data, including: The short read sample data to be tested is sequentially compared using short read length comparison software, sorted using biological data sorting software, detected using biological data repetitive sequence detection software, and calibrated using base mass fraction calibration software. Based on the calibrated short read length sample data to be tested, a short read length gvcf file is generated using short read length text generation software. The short read sample merging software is used to merge multiple short read sample data to be tested contained in the short read gvcf file. Genotyping of the merged short-read gvcf files was performed using short-read genotyping software. Sample features are extracted from the short read sample data after genotyping. The sample features include: quality score divided by coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, strand specificity test value feature, alignment quality root mean square feature, strand specificity ratio test value feature, and coverage depth feature. The quality score is divided by the coverage depth feature, alignment quality rank sum test value feature, read position rank sum test value feature, strand specificity test value feature, alignment quality root mean square feature, strand specificity ratio test value feature, and coverage depth feature, and input into the single nucleotide polymorphism sensitivity calibration model to predict the sensitivity of the short read sample data to be tested after the genotyping. The sensitivity of the short-read sample data to be tested after filtering genotyping is used to generate the short-read variant dataset with a sensitivity greater than or equal to a first preset threshold.
4. The method for identifying single nucleotide polymorphisms according to claim 1, characterized in that, Using long-read variant detection methods, a long-read variant dataset is generated to identify single nucleotide polymorphisms in the long-read sample data to be tested, including: The long-read sample data to be tested were compared using long-read alignment software in a high-quality sequencing manner. Generate long-read GVCF files using long-text generation software; The long read gvcf files were merged using long read merging software, and the merged long read gvcf files were then genotyped. After filtering the genotyping data of the long read samples to be tested, a long read variant dataset with a gene quality score greater than or equal to a second preset threshold is generated.
5. The method for identifying single nucleotide polymorphisms according to claim 1 or 4, characterized in that, Using long-read genotyping, the long-read variant dataset is genotyped to obtain a long-read genotyping dataset, including: The long-read variant dataset was aligned using long-read alignment software in a low-quality sequencing manner. The long-read variant dataset was genotyped using long-read genotyping software, and the first likelihood value of the genotyping of the long-read variant dataset was calculated. Based on the long read variant dataset after genotyping, long read VCF files are generated using long read text generation software. Based on Hardy Weinberg detection information, allele information, genotyping information, and genotyping missing information, the long read VCF file contains multiple numbers of the test sample data in one filter. Based on low-quality long read information, haplotype phasing information is obtained from the short read variant dataset using long read genotyping software. Based on the haplotype phasing information, the second likelihood value of the genotyping of the long read variant dataset is calculated using the long read genotyping software; Based on Hardy Weinberg detection information, allele information, and genotyping information, the long read variant dataset is filtered a second time. The first likelihood value and the second likelihood value of the genotyping are input into the genotyping correction software to correct the long read variant dataset after secondary filtering, so as to generate the long read genotyping dataset.
6. A single nucleotide polymorphism identification system, characterized in that, The system includes: The test sample data acquisition module is used to acquire test sample data of the single nucleotide polymorphism identification object. The test sample data includes: test long read sample data and test short read sample data. The short read variation data detection module is used to generate a short read variation dataset for identifying the single nucleotide polymorphism of the short read sample data to be tested using a short read variation detection method. The long read variant data detection module is used to generate a long read variant dataset for identifying the single nucleotide polymorphisms of the short read sample data to be tested using the long read variant detection method. The long read data genotyping module is used to perform long read genotyping on the long read variant dataset using the long read genotyping method to obtain a long read genotyping dataset; The long and short read variation integration module is used to integrate the short read variation dataset and the long read genotyping dataset to generate a long and short read composite variation dataset for identifying the test sample data; The long and short read length variation integration module includes: The single nucleotide polymorphism information acquisition submodule is used by biological sample comparison software to compare short-read variant datasets and long-read genotyping datasets to obtain the single nucleotide polymorphism shared dataset between the short-read variant dataset and the long-read genotyping dataset; The short-read Hardy Weinberg detection value calculation submodule calculates the short-read Hardy Weinberg detection value corresponding to the short-read variant dataset in the single nucleotide polymorphism shared dataset; The long-read Hardy Weinberg detection value calculation submodule is used to calculate the long-read Hardy Weinberg detection value corresponding to the long-read variant dataset in the single nucleotide polymorphism shared dataset; The identification ratio calculation submodule is used to calculate the identification ratio between the long-read Hardy Weinberg detection value and the short-read Hardy Weinberg detection value. The first determination submodule for long and short read composite variant data is used to use the long read genotyping dataset as the long and short read composite variant dataset if the identification ratio is greater than the third preset threshold. The second determination submodule for long and short read length composite variant data is used to treat the short read length variant dataset as the long and short read length composite variant dataset if the identification ratio is less than or equal to a third preset threshold.
7. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the single nucleotide polymorphism identification method according to any one of claims 1 to 5.
8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the single nucleotide polymorphism identification method according to any one of claims 1 to 5.
9. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the single nucleotide polymorphism identification method according to any one of claims 1 to 5.
Citation Information
Patent Citations
Methods, devices, and media for detecting single nucleotide variations and insertions / deletions
CN114496077A
Method, device, and storage medium for detecting mutation on the basis of sequencing data
WO2018214010A1
