Somatic cell variation detection method and device based on whole exon sequencing data of normal blood

Abstract

The invention relates to the technical field of somatic variation detection, and discloses a somatic variation detection method and device based on whole exon sequencing data of normal blood, and the method comprises the following steps: obtaining variation sites of blood whole exon sequencing data from normal blood; sequentially carrying out filtering, haploid typing and haplotype classification pretreatment on the variation sites; and inputting the preprocessed variation sites into the somatic variation prediction model to obtain somatic variation in normal blood. According to the method, variation sites in whole exon sequencing data from a normal blood sample are adopted, are sequentially subjected to filtering, haploid typing and haplotype classification preprocessing and then are input into the somatic cell variation prediction model, so that the somatic cell variation detection result is obtained, and the somatic cell variation detection accuracy can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of somatic mutation detection, and particularly relates to a somatic mutation detection method and device based on whole-exome sequencing data of normal blood. BACKGROUND

[0002] Somatic mutations occur spontaneously and accumulate in normal cells. Although most of these mutations are neutral, some mutations can provide selective advantage, promote cell division and may lead to clonal expansion. When this event occurs in hematopoietic stem cells (HSCs), it is called clonal hematopoiesis (CH). Clonal hematopoiesis is not only closely related to hematological tumors, but also related to the risk of a series of non-hematological diseases such as cardiovascular diseases. Therefore, accurately identifying somatic mutations from the blood of healthy individuals is crucial to reveal their role in the transformation from health to disease.

[0003] At present, although the single sample analysis strategy based on biological test software Mutect2 is commonly used, it often faces the challenge of false positive results. The existing machine learning method cannot realize accurate detection of somatic mutations from whole-exome sequencing data of normal blood, and the standard model of the existing machine learning is based on whole-genome sequencing data, especially for brain tissue, and does not support detection of somatic mutations of whole-exome sequencing (WES) data of normal blood. SUMMARY

[0004] Therefore, the present application provides a somatic mutation detection method based on whole-exome sequencing data of normal blood to solve the problem that the existing machine learning method cannot accurately detect somatic mutations from whole-exome sequencing data of normal blood.

[0005] In a first aspect, the present application provides a somatic mutation detection method based on whole-exome sequencing data of normal blood, which comprises:

[0006] Obtaining the variant sites of the whole-exome sequencing data from normal blood;

[0007] Filtering, haplotype typing, and haplotype classification preprocessing are sequentially performed on the variant sites;

[0008] The preprocessed variant sites are input into a somatic mutation prediction model to obtain the somatic mutation detection results in normal blood.

[0009] In an optional embodiment, the filtering preprocessing of the variant sites of the whole-exome sequencing data of normal blood comprises:

[0010] Based on the variant sites, mutations marked as preset identities by biological software are reserved;

[0011] Mutations with less than a preset sequencing depth, less than or equal to a preset variant read length, and read length with supporting mutations only in the positive strand or the reverse strand are removed;

[0012] Mutations with a variant allele frequency less than a first preset value, or greater than or equal to a second preset value are removed;

[0013] Mutations located in repeat regions, including simple repeats, RepeatMasker annotation regions, and repeat region annotations obtained from the UCSC Genome Browser database, are removed;

[0014] Mutations located in cluster regions with a haplotype number enrichment greater than or equal to 3 are removed;

[0015] Mutations located in germline copy number variation regions are removed.

[0016] In an optional embodiment, the pre-processing of haplotype typing of the variant sites includes:

[0017] The haplotype number of the variant sites is determined based on the read length haplotype typing technology; the haplotype number includes 2, 3, and greater than 3, wherein 2 indicates that the variant site is a germline cell mutation of heterozygous type, 3 indicates a somatic mutation site, and greater than 3 indicates a variant site located in a low mapping region, a copy number variation region, or a sequencing error.

[0018] In an optional embodiment, the pre-processing of haplotype classification of the variant sites includes:

[0019] A preset number of haplotype numbers are randomly selected as 3 for the variant sites;

[0020] The variant sites are checked using the Integrated Genome Browser;

[0021] The sites in the messy region are marked as repeat regions, the sites in the clear region are marked as somatic regions, and the remaining sites are determined as uncertain regions.

[0022] In an optional embodiment, the variant sites with a haplotype number of 3 are further classified, including:

[0023] The genotype categories of all the variant sites are reassigned by using a polynomial logistic regression model including 31 read length features, the genotype categories including a germline category, a somatic category and a repeat category; wherein, the read length features include two features reflecting mapping quality, namely, mapq_p feature and mapq_difference feature; three features recording the number of mismatches in each read, namely, major_mismatches_mean feature, minor_mismatches_mean feature and mismatches_p feature; six features reflecting base quality, namely, baseq_p feature, baseq_t feature, ref_baseq1b_p feature, ref_baseq1b_t feature, alt_baseq1b_p feature and alt_baseq1b_t feature; and the remaining features including read depth, variant allele frequency, genotyping likelihood, strand bias, read pair bias, sequencing cycle bias, read mapping position bias, allele query position bias, local mappability score, proportion of clipped reads, multi-allele test, GC content and three nucleotide base contexts of the variant.

[0024] In an optional implementation, before the step of obtaining the variant sites, the method further includes:

[0025] The training set corresponding to the variant sites of the whole exome sequencing data of the normal blood used for training the somatic variant prediction model is obtained, including:

[0026] The variant training set is sequentially filtered, haplotyped and haplotype classified for preprocessing.

[0027] According to a second aspect, the embodiments of the present disclosure provide a somatic variant detection device based on whole exome sequencing data of normal blood, including:

[0028] The obtaining module is configured to obtain variant sites of whole exome sequencing data from normal blood;

[0029] The processing module is configured to sequentially filter, haplotype and haplotype classify the variant sites for preprocessing;

[0030] The detection module is configured to input the preprocessed variant sites into the somatic variant prediction model to obtain a somatic variant detection result in the normal blood.

[0031] According to a third aspect, the embodiments of the present disclosure provide a computer device, including:

[0032] The memory and the processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the somatic variation detection method in the first aspect or any of the implementation manners of the first aspect.

[0033] According to a fourth aspect, the present disclosure provides a computer readable storage medium, which stores computer instructions, and the computer instructions are used to make a computer execute the somatic variation detection method in the first aspect or any of the implementation manners of the first aspect.

[0034] According to a fifth aspect, the present disclosure provides a computer program product, which includes computer instructions, and the computer instructions are used to make a computer execute the somatic variation detection method in the first aspect or any of the implementation manners of the first aspect.

[0035] The technical scheme has the following advantages:

[0036] The present application relates to somatic variation detection technical field, disclose a kind of based on normal blood's whole exon sequencing data's somatic variation detection method and device, wherein method, comprising: obtaining the variation site of whole exon sequencing data from normal blood;Variation site is filtered, variation site haplotype typing, variation site haplotype classification preprocessing in turn;The variation site after preprocessing is input into somatic variation prediction model, obtains the somatic variation detection result of normal blood.The present application uses variation site in whole exon sequencing data from normal blood sample, and after being filtered, haplotype typing, haplotype classification preprocessing in turn, input into somatic variation prediction model to obtain somatic variation detection result, can improve the accuracy of somatic variation detection. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0038] Figure 1 It is a flowchart of somatic variation detection method based on whole exon sequencing data of normal blood according to the embodiment of the present application;

[0039] Figure 2 It is another flowchart of somatic variation detection method based on whole exon sequencing data of normal blood according to the embodiment of the present application;

[0040] Figure 3This is a schematic flowchart of another somatic cell variation detection method based on whole exon sequencing data of normal blood according to an embodiment of the present invention;

[0041] Figure 4 This is a flowchart illustrating another method for detecting somatic cell variations based on whole-exome sequencing data of normal blood according to an embodiment of the present invention.

[0042] Figure 5 This is a comparative schematic diagram of a somatic cell variation detection method based on whole exon sequencing data of normal blood according to an embodiment of the present invention;

[0043] Figure 6 This is a structural block diagram of a somatic cell variation detection device based on whole exon sequencing data of normal blood according to an embodiment of the present invention;

[0044] Figure 7 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0045] 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.

[0046] According to an embodiment of the present invention, an embodiment of a method for detecting somatic cell variations based on whole exon sequencing data of normal blood is provided. 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. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0047] This embodiment provides a method for detecting somatic cell variations based on whole-exome sequencing data from normal blood, which can be used on mobile terminals such as mobile phones and tablets. Figure 1 This is a flowchart of a method for detecting somatic cell variations according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps:

[0048] Step S101: Obtain variant sites from whole-exome sequencing data in normal blood.

[0049] Specifically, this disclosure describes the analysis of normal blood WES data from 200,000 individuals in the UK Biobank using the Mutect2 biological software to generate candidate somatic variants, namely the blood exome sequencing variant data mentioned above. The Mutect2 software was run in single-sample mode with recommended settings, including the use of the gnomAD dataset, Panel of Normals (PoN), and statistical models specifically designed to detect chain and orientation bias. Furthermore, the PoN was constructed based on 100 randomly selected young individuals (under 40 years of age) to filter out technology-related false positive mutations.

[0050] In this embodiment, the variant sites in normal blood are derived from whole-exome sequencing data (blood WES data), which can improve the accuracy of detecting somatic cell variants from normal blood WES data.

[0051] Step S102 involves sequentially filtering, haplotype typing, and haplotype classification preprocessing of the variant sites.

[0052] Specifically, the following conditions are applied for filtering in this disclosure: (1) Mutations marked as 'PASS' by FilterMutectCalls are retained. (2) Mutations with sequencing depth <30, number of reads supporting the variant ≤3, and reads supporting the variant only in the positive or reverse strand are removed. (3) Mutations with a variant allele frequency (VAF) <0.02 or VAF ≥0.4 are removed. (4) Mutations located in repetitive regions are removed, including simple repeats, RepeatMasker annotated regions, and repetitive region annotations obtained from the UCSC Genome Browser database. (5) Mutations located in cluster regions enriched with ≥3 haplotypes are removed. (6) Mutations located in regions of reproductive copy number variations (CNVs) are removed.

[0053] Further, haplotype phasing is performed on the filtered variant sites: this process uses read-length-based phasing techniques to determine the number of haplotypes at candidate mutation sites. These sites can be divided into three categories: a haplotype count of 2 indicates that the site is a heterozygous germline mutation, a haplotype count of 3 indicates that the site is a somatic mutation, and a haplotype count greater than 3 indicates that the site is located in a low-mapping region, a copy number variation region, or a sequencing error.

[0054] Furthermore, the classification of variant sites with a haplotype number of 3 was further refined: In this embodiment, 500 variant sites with a haplotype number of 3 were randomly selected and examined using the Integrated Genome Viewer (IGV). Sites in disordered regions were labeled 'repeat' (123 sites), sites in clear regions were labeled 'somatic' (355 sites), and the remaining sites were determined to be indeterminate. To refine the classification, a multinomial logistic regression model containing 31 read length features was used to redistribute the genotype categories of all variant sites, including 'germline', 'somatic', and 'repeat'. The read length features include two features reflecting mapping quality ('mapq_p' and 'mapq_difference'); three features used to record the number of mismatches in each read ('major_mismatches_mean', 'minor_mismatches_mean', 'mismatches_p'); and six features reflecting base quality ('baseq_p', 'baseq_t', 'ref_baseq1b_p', 'ref_baseq1b_t', 'alt_baseq1b_p', 'alt_baseq1b_t'). The remaining features include read length depth, variant allele frequency (VAF), genotyping likelihood, strand bias, read pair bias, sequencing cycle bias, read length mapping position bias, allele query position bias, local mappability score, proportion of spliced ​​reads, multi-allelic test, GC content, and three nucleotide background of variants.

[0055] The embodiments disclosed herein filter variant sites, thereby improving the accuracy of detecting somatic cell variants from normal blood using somatic cell variant prediction models.

[0056] Step S103: Input the preprocessed variant sites into the somatic cell variant prediction model to obtain the somatic cell variant detection results in normal blood.

[0057] Specifically, the variant sites are preprocessed by haplotype typing and haplotype classification and then input into the somatic cell variant prediction model for training, thereby obtaining the somatic cell variant detection results of normal blood.

[0058] This embodiment of the disclosure uses variant sites from whole-exome sequencing data of normal blood from MosaicForecast-WES data. After filtering, haplotype typing, and haplotype classification preprocessing, the data are input into a somatic variant prediction model, which can improve the accuracy of detecting somatic variants from normal blood.

[0059] This embodiment provides a method for detecting somatic cell variations based on whole-exome sequencing data from normal blood, which can be used on mobile terminals such as mobile phones and tablets. Figure 2 This is a flowchart of a method for detecting somatic cell variations according to an embodiment of the present invention, such as... Figure 2 As shown, step S102 involves filtering and preprocessing the variant sites. This process includes the following steps:

[0060] Step S201: Based on the mutation site, retain the mutations that have been marked with a preset identifier by the biological software.

[0061] Specifically, for candidate somatic cell mutations, mutations marked as "preset identifier PASS" by the biological software FilterMutectCalls are retained.

[0062] Step S202: Remove mutations that are less than the preset sequencing depth, less than or equal to the preset variant read length, and reads that support the variant only in the positive or negative strand.

[0063] Specifically, reads shorter than a preset sequencing depth of 30, reads shorter than or equal to 3 supporting the variant, and reads supporting the variant that are only present in the positive or negative strand are removed. The preset sequencing depth and preset variant read length can be determined based on actual conditions.

[0064] Step S203: Remove mutations whose allele frequencies are less than a first preset value, or greater than or equal to a second preset value.

[0065] Specifically, mutations with a Variant allele frequency (VAF) less than a first preset value of 0.02 or greater than a second preset value of 0.4 are removed.

[0066] Step S204: Remove mutations located in repeating regions, including simple repeats, RepeatMasker annotated regions, and repeating region annotations obtained from the UCSC Genome Browser database.

[0067] Specifically, mutations located in repetitive regions are removed, including simple repetitions, RepeatMasker annotated regions, and repetitive region annotations obtained from the UCSC Genome Browser database.

[0068] Step S205: Remove mutations located in cluster regions enriched with a number of haplotypes greater than or equal to 3.

[0069] Specifically, mutations located in cluster regions enriched with a number of haplotypes greater than or equal to 3 are removed.

[0070] Step S206: Remove mutations located in the region of reproductive copy number variation.

[0071] Specifically, mutations located within the region of reproductive copy number variations (CNVs) are removed.

[0072] This disclosure filters variant sites in whole-exome sequencing data of normal blood, thereby improving the accuracy of detecting somatic variants from normal blood using somatic variant prediction models.

[0073] In one alternative implementation, in Figure 2 In addition to the preprocessing for haplotype typing of variant sites, the following steps are also included:

[0074] Step S207: Determine the number of haplotypes of the variant site based on read length haplotype typing technology; the number of haplotypes includes: 2, 3 and greater than 3, where 2 represents a variant site of heterozygous germ cell mutation, 3 represents a somatic cell mutation site, and greater than 3 represents a variant site located in a low-mapping region, copy number variation region or sequencing error.

[0075] Specifically, haplotype phasing is performed on the filtered variant sites. This process uses read-length-based phasing techniques to determine the number of haplotypes at candidate mutation sites. These sites can be divided into three categories: a haplotype count of 2 indicates that the site is a heterozygous germline mutation; a haplotype count of 3 indicates that the site is a somatic mutation; and a haplotype count greater than 3 indicates that the site is located in a low-mapping region, a copy number variation region, or a sequencing error.

[0076] In one alternative implementation, in Figure 2 In the process, the preprocessing for haplotype classification of variant sites includes:

[0077] Step S208: Randomly select a preset number of variant sites with a haplotype number of 3.

[0078] Specifically, in this embodiment of the disclosure, 500 variant sites with a haplotype number of 3 were randomly selected and these sites were examined using an Integrated Genome Viewer (IGV).

[0079] Step S209: Use a comprehensive genome browser to examine variant sites.

[0080] Specifically, these sites are examined using the Integrated Genome Browser (IGV).

[0081] In step S210, sites in the disordered region are marked as repeating regions, sites in the clear region are marked as somatic regions, and the remaining sites excluding the disordered and clear regions are determined as uncertain regions.

[0082] Specifically, sites in messy areas were marked as 'repeat' (123 sites), sites in clear areas were marked as 'somatic' (355 sites), and the remaining sites were determined to be uncertain.

[0083] In one specific implementation, Figure 2 In this study, sites with a haplotype number of 3 were further subdivided, including:

[0084] Step S211: A multinomial logistic regression model containing 31 read length features is used to redistribute the genotype categories of all variant sites. These genotype categories include germline, somatic, and repeat categories. The read length features include two features reflecting mapping quality: mapq_p and mapq_difference. Three features are used to record the number of mismatches in each read segment: major_mismatches_mean, minor_mismatches_mean, and mismatc. hes_p features; six features reflect base quality, namely baseq_p features, baseq_t features, ref_baseq1b_p features, ref_baseq1b_t features, alt_baseq1b_p features, and alt_baseq1b_t features. The remaining features include read depth, frequency of variant alleles, genotyping likelihood, strand bias, read pair bias, sequencing cycle bias, read mapping position bias, allele query position bias, local mappability score, proportion of spliced ​​reads, multi-allele test, GC content, and three nucleotide bases of variant background.

[0085] After preprocessing the whole exome sequencing data of normal blood through the above steps S201-S211, filtering the variant sites, classifying the variant sites into haplotypes, and classifying the variant sites into haplotypes, a trained somatic variant prediction model is obtained. This model can use the above 31 read length features as covariates and the refined genotype classification as response variables.

[0086] like Figure 3 The diagram shown is a schematic of a somatic cell variation detection method based on whole-exome sequencing data from normal blood.

[0087] This embodiment provides a method for detecting somatic cell variations based on whole-exome sequencing data from normal blood, which can be used on mobile terminals such as mobile phones and tablets. Figure 4 This is a flowchart of a method for detecting somatic cell variations according to an embodiment of the present invention, such as... Figure 4As shown, prior to step S101, which involves obtaining variant sites from normal blood for whole-exome sequencing data, the following steps are also included:

[0088] S100, Obtain the training set corresponding to the mutation sites used to train the somatic mutation prediction model, including:

[0089] The training set corresponding to the variant sites is preprocessed sequentially by filtering, haplotype typing, and haplotype classification. The filtering, haplotype typing, and haplotype classification preprocessing steps are described in steps S201-S211 above and will not be repeated here.

[0090] Specifically, in this embodiment, 100,000 UK Biobank samples with WES data were randomly divided into two equal subsets: a training set and a test set, each with 50,000 samples (50% of the total). Using the method described above, we trained a somatic mutation model, MosaicForecast-WES, on the segregable variant sites in the training set. Subsequently, this embodiment evaluated the model's accuracy and power using the segregable variant sites in the test set and compared its performance with the MosaicForecast default model and Mutect2. In the analysis of this embodiment, variant sites with a haplotype number of 3 were considered "true" somatic mutations. This embodiment calculated the accuracy of somatic mutations predicted by MosaicForecast-WES and the MosaicForecast default model, as well as the accuracy of all segregable variant sites identified by Mutect2. The efficacy of different approaches was determined by comparing the MosaicForecast-WES and MosaicForecast default models with all Mutect2-classifiable variant sites detected in the DNMT3A gene, a gene known to be associated with clonal hematopoiesis.

[0091] like Figure 5 As shown, the results indicate that the embodiments of this disclosure have significantly higher accuracy (approximately 89%) while maintaining a high level of efficacy, compared to Mutect2 (approximately 50%) and the default model MosaicForecast (63%).

[0092] In summary, the somatic variation detection method based on whole-exome sequencing data of normal blood in this embodiment includes: obtaining variant sites from whole-exome sequencing data of normal blood; performing preprocessing on the variant sites, including filtering, haplotype typing, and haplotype classification; and inputting the preprocessed variant sites into a somatic variation prediction model to obtain the somatic variation detection results of normal blood. Ultimately, this invention not only supports the detection of somatic variations in whole-exome sequence (WES) data of normal blood but also achieves accurate detection of somatic variations.

[0093] This embodiment also provides a somatic cell variation detection device based on whole-exome sequencing data of normal blood. This device is used to implement the above embodiments and preferred embodiments, and 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 device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0094] This embodiment provides a somatic cell variation detection device based on whole-exome sequencing data from normal blood, such as... Figure 6 As shown, it includes:

[0095] Acquisition module 61 is used to acquire variant sites from whole-exome sequencing data in normal blood.

[0096] The processing module 62 performs filtering, haplotype typing, and haplotype classification preprocessing on the variant sites in sequence.

[0097] The detection module 63 is used to input the preprocessed variant sites into the somatic cell variant prediction model to obtain the somatic cell variant detection results of normal blood.

[0098] In one alternative implementation, the processing module 62 includes:

[0099] The mutation identification and labeling submodule is used to retain mutations that have been marked with preset labels by biological software based on the mutation site;

[0100] The first removal submodule is used to remove mutations that are less than the preset sequencing depth, less than or equal to the preset variant read length, and have supporting variant reads only in the positive or negative strand.

[0101] The second removal submodule is used to remove mutations whose mutation allele frequencies are less than the first preset value, or greater than or equal to the second preset value.

[0102] The third removal submodule is used to remove mutations located in repetitive regions, including simple repetitions, RepeatMasker annotated regions, and repetitive region annotations obtained from the UCSC Genome Browser database;

[0103] The fourth removal submodule is used to remove mutations located in cluster regions enriched with a number of haplotypes greater than or equal to 3;

[0104] The fifth removal submodule is used to remove mutations located in regions of reproductive copy number variation.

[0105] In one alternative implementation, the processing module 62 includes:

[0106] The mutation site determination submodule is used to determine the number of haplotypes of variant sites based on read length haplotype typing technology. The number of haplotypes includes 2, 3 and greater than 3, where 2 represents a variant site that is a heterozygous germ cell mutation, 3 represents a somatic cell mutation site, and greater than 3 represents a variant site located in a low-mapping region, copy number variation region or sequencing error.

[0107] In one alternative implementation, the processing module 60 includes:

[0108] The variant site determination submodule is used to randomly select a preset number of variant sites with a haplotype number of 3;

[0109] The variant site inspection submodule is used to inspect variant sites using a comprehensive genome browser;

[0110] The site region labeling submodule is used to label sites in messy regions as repeating regions, sites in clear regions as somatic regions, and other sites as uncertain regions.

[0111] In one alternative implementation, the variant sites with a haplotype number of 3 are further refined into categories, including:

[0112] A multinomial logistic regression model with 31 read length features was used to redistribute the genotype categories of all variant sites, including germline, somatic, and repeat categories. The read length features include two features reflecting mapping quality: `mapq_p` and `mapq_difference`. Three features are used to record the number of mismatches in each read segment: `major_mismatches_mean`, `minor_mismatches_mean`, and `mismatches_p`. Six features reflect base quality: baseq_p, baseq_t, ref_baseq1b_p, ref_baseq1b_t, alt_baseq1b_p, and alt_baseq1b_t. The remaining features include read depth, variant allele frequency, genotyping likelihood, strand bias, read pair bias, sequencing cycle bias, read mapping position bias, allele query position bias, local mappability score, proportion of spliced ​​reads, multi-allele test, GC content, and three nucleotide bases of variants.

[0113] In one optional implementation, it further includes: a mutation training set acquisition module, used to acquire a training set corresponding to the mutation sites used for training the somatic cell mutation prediction model;

[0114] The module for obtaining the mutation training set includes:

[0115] The preprocessing submodule is used to perform filtering, haplotype typing, and haplotype classification preprocessing on the training set corresponding to the variant sites in sequence.

[0116] 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.

[0117] In this embodiment, the somatic cell variation detection device based on whole exon sequencing data of normal blood is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit), a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0118] This invention also provides a computer device having the somatic cell variation detection device based on whole exon sequencing data of normal blood as described above.

[0119] Please see Figure 7 , Figure 7This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 7 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 7 Take a processor 10 as an example.

[0120] 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.

[0121] 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.

[0122] 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.

[0123] 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.

[0124] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0125] 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.

[0126] 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.

[0127] 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 detecting somatic cell variations based on whole-exome sequencing data of normal blood, characterized in that, The method includes: Variant sites obtained from whole-exome sequencing data of normal blood; The variant sites are sequentially preprocessed by filtering, haplotype typing, and haplotype classification. The preprocessed variant sites are input into the somatic cell variant prediction model to obtain the somatic cell variant detection results in normal blood.

2. The method according to claim 1, characterized in that, The filtering preprocessing of the variant sites includes: Based on the mutation sites, mutations marked with preset identifiers by biological software are retained; Remove mutations that are less than the preset sequencing depth, less than or equal to the preset variant read length, and those that have supporting variants only in the positive or reverse strand; Remove mutations whose allele frequencies are less than a first preset value, or greater than or equal to a second preset value; Mutations located in repetitive regions were removed, including simple repeats, RepeatMasker annotated regions, and repetitive region annotations obtained from the UCSC Genome Browser database; Remove mutations located in cluster regions enriched with haplotype numbers greater than or equal to 3; Remove mutations located within regions of reproductive copy number variation.

3. The method according to claim 2, characterized in that, Preprocessing for haplotype typing of the aforementioned variant sites includes: The number of haplotypes at a variant site is determined using haplotype typing based on read length. The number of haplotypes includes 2, 3, and greater than 3, where 2 represents a variant site that is a heterozygous germ cell mutation, 3 represents a somatic cell mutation site, and greater than 3 represents a variant site located in a low-mapping region. Copy number variation regions may be sequencing errors.

4. The method according to claim 3, characterized in that, Preprocessing for haplotype classification of the variant sites includes: Randomly select a preset number of variant sites with a haplotype number of 3; The variant sites were examined using a comprehensive genome browser; Sites in cluttered areas were marked as repeating regions, sites in clear areas were marked as somatic regions, and the remaining sites were classified as indeterminate regions.

5. The method according to claim 3, characterized in that, Further refinement of the classification of variant sites with a haplotype number of 3 includes: A multinomial logistic regression model with 31 read length features was used to redistribute the genotype categories of all variant sites, including germline, somatic, and repeat categories. The read length features include two features reflecting mapping quality: `mapq_p` and `mapq_difference`. Three features are used to record the number of mismatches in each read segment: `major_mismatches_mean`, `minor_mismatches_mean`, and `mismatches_p`. Six features reflect base quality: baseq_p, baseq_t, ref_baseq1b_p, ref_baseq1b_t, alt_baseq1b_p, and alt_baseq1b_t. The remaining features include read depth, variant allele frequency, genotyping likelihood, strand bias, read pair bias, sequencing cycle bias, read mapping position bias, allele query position bias, local mappability score, proportion of spliced ​​reads, multi-allele test, GC content, and three nucleotide bases of variants.

6. The method according to claim 1, characterized in that, Prior to the step of obtaining variant sites from whole-exome sequencing data in normal blood, the method further includes: Obtaining the training set corresponding to the mutation sites used to train the somatic mutation prediction model includes: The training set corresponding to the mutation sites is sequentially filtered, haplotyped, and preprocessed for haplotype classification.

7. A somatic cell variation detection device based on whole-exome sequencing data of normal blood, characterized in that, include: The acquisition module is used to obtain variant sites from whole-exome sequencing data in normal blood. The processing module sequentially performs filtering, haplotype typing, and haplotype classification preprocessing on the variant sites; The detection module is used to input the preprocessed variant sites into the somatic cell variant prediction model to obtain the somatic cell variant detection results in normal blood.

8. A computer device, characterized in that, include: A memory and a processor are interconnected, the memory storing computer instructions, and the processor executing the computer instructions to perform the somatic cell variation detection method based on whole exon sequencing data of normal blood as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the somatic cell variation detection method based on whole exon sequencing data of normal blood, as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The method includes computer instructions for causing a computer to perform the somatic variation detection method based on whole exon sequencing data of normal blood, as described in any one of claims 1 to 6.

Images