CN109616155B - Data processing system and method for genetic variation pathogenicity classification of coding region - Google Patents

Abstract

The invention discloses a data processing system and a data processing method for genetic variation pathogenicity classification of a coding region, wherein the system comprises a variation locus discovery module, a variation locus annotation module, a data and resource loading module, a pathogenicity distinguishing and classifying module and a result explaining and verifying module which are sequentially connected, wherein the variation locus discovery module is used for searching the specific position of a genetic variation pathogenicity variation locus in the coding region; the variant locus annotation module is used for carrying out information annotation on variant loci and generating a data file corresponding to each variant locus; the data and resource loading module is used for reading an external resource file and the data file which are judged by pathogenicity; the pathogenicity judging and classifying module is used for calculating values of all discriminants in each variation site in the data file and classifying the pathogenicity of the genetic variation; the result interpretation and verification module is used for interpreting and manually verifying the classification results; the invention improves the working efficiency of the personnel reading the genetic disease data.

Description

A data processing system and method for pathogenicity classification of genetic variation in coding regions

technical field

The invention belongs to the field of bioinformatics, and in particular relates to a data processing system and method for pathogenicity classification of genetic variation in coding regions.

Background technique

With the development of next-generation high-throughput sequencing technology and modern information processing technology, it is possible to process and interpret massive genetic data based on target gene sets, whole exomes, and whole genome sequencing. Among them, the functional analysis of genetic variation, especially the classification and interpretation of the pathogenicity of genetic variation in coding regions, is currently the focus of attention in the field of gene sequencing data processing and analysis. It provides an important basis for clinical auxiliary decision-making, accurate interpretation of data, genetic counseling and other applications.

At present, the pathogenicity classification of genetic variants is mainly guided by the classification and interpretation standards of disease variant sites released by the American College of Medical Genetics and Genomics (ACMG) in 2015. This guideline gives 28 evaluation criteria for judging the clinical significance of variant sites. Since the guideline itself cannot clearly specify the details and parameters of each criteria, different data interpreters may have certain differences in specific operations. , leading to a high degree of inconsistency in interpretation conclusions. In addition, for each test sample, data interpretation personnel are required to manually query various database resources and compare relevant evidence information one by one, making the entire data processing process very cumbersome and inefficient, and prone to errors. It can be seen that the degree of automation and systematization of existing technical methods is very low, and there is a lack of efficient sequencing data interpretation tools, which cannot meet the needs of rapid processing of large-scale sample data and consistent results.

Contents of the invention

The main purpose of the present invention is to provide a data processing system and method for pathogenicity classification of genetic variation in coding regions, which can realize semi-automatic and systematic processing of large-scale sample data and massive variation site information, and speed up genetic The speed of disease gene analysis greatly improves the work efficiency of genetic disease data interpreters and avoids errors caused by cumbersome processing; it solves the problems of low efficiency and high cost in existing technologies; the specific technical solutions are as follows:

On the one hand, a data processing system for pathogenicity classification of genetic variation in coding regions is provided. The system is constructed based on ACMG guidelines, and the system includes a variant site discovery module, a variant site annotation module, and a data And resource loading module, pathogenicity discrimination and classification module and result interpretation and verification module, in which:

The variant site discovery module is used to find the specific position of the genetic variation pathogenic variant site in the coding region; the variant site includes SNPs and small fragment INDELs;

The variable site annotation module is used to annotate the variable site information and generate a data file corresponding to each of the variable sites; the information annotation includes the chromosome where the variable site is located, the reference allele, the replacement Alleles, exon positions, rarity, genes, amino acid changes, various calculable tools to calculate the harmfulness of mutations and prediction results, and annotations of mutation frequency information in different populations;

Data and resource loading module, for reading the external resource file and the data file of pathogenicity discrimination; The external resource file includes the gene list, Clinvar, OMIM, dbscSNV and dbNSFP database of the pathogenicity discrimination;

The pathogenicity judgment and classification module is used to calculate the value of all discriminant items in each of the mutation sites in the data file, and perform classification operations on the pathogenicity of genetic variation; the discriminant items include PVS1, PS1, PS4, PM1, PM2, PM4, PM5, PP2, PP3, PP5, BS1, BS2, BP1, BP3, BP4, BP6, BP7 and BA1;

The result interpretation and verification module is used for performing result interpretation and manual verification on the classification.

Further, the variation site discovery module includes a sequence comparison and mapping unit, a sequence data preprocessing unit, and a SNPs and small fragment INDELs variation discovery unit; the sequence comparison and mapping unit is used to receive the original Sequencing data, and mapping the sequence data to the reference genome; the sequence data preprocessing unit is used to perform preprocessing operations on the sequence data mapped to the reference genome; the SNPs and small fragment INDELs variation discovery unit is used to identify the preprocessing The processed sequence data is compared with the variation sites of the reference genome, and the genotype of each variation site is calculated.

Further, the input of the variable site discovery module is the original sequencing data file in fastq format, and the output of the variable site discovery module is a vcf format file containing all variable sites.

Further, the variable site discovery module uses the BWA-MEM algorithm to complete the mapping operation of the original sequencing data; the variable site discovery module uses the GATK tool to realize the search operation for the variable site.

Further, the variable site annotation module includes a site annotation unit, which annotates the SNPs and the small fragment INDELs, and can select and specify the variable site for information annotation.

Further, the data and resource loading module includes an annotation data loading unit and an external resource loading unit, the annotation data loading unit is used to read and store the data file; the external resource loading unit is used to read the External resource files.

Further, the pathogenicity discrimination and classification module includes a pathogenicity discrimination unit and a pathogenicity classification unit, and the pathogenicity discrimination unit is used to calculate the value of all the discrimination items in each of the mutation sites. value; the pathogenicity classification unit classifies all the variant sites according to the calculation results in response to the pathogenicity discrimination unit.

Further, the result interpretation and verification module includes a result interpretation unit and a verification unit, and the result interpretation unit is used to give the classification basis of the discrimination results and classification results of the pathogenicity discrimination and classification module; the verification unit It is used to compare the categories and submit them to manual for further review and confirmation based on the comparison results.

In another aspect, a data processing method for classification of pathogenicity of genetic variation in coding regions is provided, which is applied to the above data processing system for classification of pathogenicity of genetic variation in coding regions, characterized in that the method includes the steps of:

S1. Input the original sequencing data composed of sequence data to the variation site discovery module, use the BWA-MEM algorithm to map the sequence data to the reference genome, use the Picard tool to preprocess the mapped sequence data, and use the GATK tool to find The variation site of the sequence data;

S2. Use the variable site annotation module to annotate information on all the variable sites, and generate a data file corresponding to each of the variable sites;

S3. Use the data and resource loading module to read the data file, and at the same time read the external resource file used for pathogenicity discrimination;

S4. Calculating the values of all discriminant items for each of the variant sites based on the ACMG guidelines, and scoring the discriminant items, performing a summary operation based on the scoring, and performing a classification operation on pathogenicity based on the summary;

S5. Explain the classification of the pathogenicity, and use the explanation as the basis for the classification, compare the classification results with the interpretation results of the Clinvar and InterVar genetic variation database interpretation tools, and submit them to manual processing based on the comparison results Further review and confirmation.

Further, in the step S5, if some of the classification results are inconsistent with the interpretation results of the Clinvar and InterVar genetic variation database interpretation tools, the classification results are submitted to manual review and confirmation; otherwise, the corresponding classification results are completed. disease classification.

The data processing system and method for pathogenicity classification of genetic variation in coding regions of the present invention, the processing system consists of a variation site discovery module, a variation site annotation module, a data and resource loading module, a pathogenicity discrimination and classification module and a result The interpretation and verification modules are connected in sequence, and the variable site discovery module completes the search for all genetic variation sites in the coding region, and annotates all the information in the variable site through the variable site annotation module to generate a corresponding to each variable site The data file, and then the data and resource loading module reads the data file and the external resource file used for pathogenicity discrimination; then, the specific value of the discriminant item of each mutation site is calculated through the pathogenicity discrimination and classification module , to score each discriminant item, perform a summary operation of all discriminant items according to the scoring, and classify the pathogenicity of all genetic variations according to the summary; finally, the result interpretation and verification module gives the pathogenicity of all genetic variants. The basis for the classification of the disease, and compare it with the results of the genetic variation data interpretation tool. If the comparison is inconsistent, further review and confirmation will be performed manually again; compared with the prior art, the present invention can target gene sets, Whole exome sequencing data realizes semi-automatic and systematic processing of large-scale samples and massive variation site information; the invention integrates a variation site discovery module, a variation site annotation module, a data and resource loading module, and a The entire data processing process is standardized and systematic; the invention can speed up the analysis speed of genetic disease genetic data, and greatly improve the work efficiency of genetic disease data interpreters. Avoid errors caused by cumbersome processing.

Description of drawings

Fig. 1 is a schematic block diagram of the structural composition of the data processing system for the classification of the pathogenicity of the genetic variation in the coding region described in the embodiment of the present invention;

Fig. 2 is a block diagram schematic diagram of the discovery module of the variable site described in the embodiment of the present invention to find the variable site;

Fig. 3 is a schematic flowchart of the information annotation of the variable site by the self-defined annotation unit described in the embodiment of the present invention;

Fig. 4 is a block diagram of a data processing method for pathogenicity classification of genetic variation in the coding region described in the embodiment of the present invention.

1-variant site discovery module, 2-variant site annotation module, 3-data and resource loading module, 4-pathogenicity discrimination and classification module, 5-result interpretation and verification module.

Detailed ways

In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention.

The data processing system and method for the pathogenicity classification of genetic variation in the coding region of the present invention is constructed based on the ACMG guidelines, wherein the coding region is all proteins on the genome Coding region, that is, all exon regions on the genome; in the process of data processing, the present invention relates to BWA-MEM algorithm, Picard tool and GATK tool, and Picard tool includes AddOrReplaceReadGroups algorithm and MarkDuplicate algorithm, GATK tool includes HaplotypeCaller algorithm, GenotypeGVCFs algorithm, SelectVariants algorithm, VariantRecalibrator algorithm, ApplyRecalibration algorithm and CombineVariants algorithm.

With reference to Figures 1 to 4, in the data processing system and method for the classification of pathogenicity of genetic variation in coding regions of the present invention, the data processing system for classification of pathogenicity of genetic variation in coding regions includes a sequentially connected variation site discovery module 1 , variant site annotation module 2, data and resource loading module 3, pathogenicity discrimination and classification module 4 and result interpretation and verification module 5, variant site discovery module 1 is used to find the specific position of the variant site in the coding region, Variation sites include SNPs and INDELs of small fragments; the variation site annotation module 2 is used to annotate the information of the variation sites and generate data files corresponding to each variation site; the information annotation includes the chromosome where the variation site is located, the reference Alleles, replacement alleles, exon positions, rarity, genes, amino acid changes, various calculation tools for the calculation of mutation harmfulness and prediction results, annotations of mutation frequency information in different populations; data and The resource loading module 3 is used to read the external resource files and data files of pathogenicity discrimination; the external resource files include the gene list, Clinvar, OMIM, dbscSNV and dbNSFP databases of pathogenicity discrimination; the pathogenicity judgment and classification module 4 is used Calculate the value of all discriminant items in each mutation site in the data file, and classify the pathogenicity of genetic variation; discriminant items include PVS1, PS1, PS4, PM1, PM2, PM4, PM5, PP2, PP3 , PP5, BS1, BS2, BP1, BP3, BP4, BP6, BP7, and BA1; the result interpretation and verification module 5 is used to perform result interpretation and manual verification of the classification; the data processing method of the pathogenicity classification of the genetic variation in the coding region Including steps: S1. Input the original sequencing data composed of sequence data to the variation site discovery module, use the BWA-MEM algorithm to map the sequence data to the reference genome, use the Picard tool to preprocess the mapped sequence data, and use The GATK tool finds the variation site of the sequence data; S2, uses the variation site annotation module to perform information annotation on all the variation sites, and generates a data file corresponding to each of the variation sites; S3, uses the variation site annotation module The above data and resource loading module reads the data file, and at the same time reads the external resource file used for pathogenicity discrimination; S4, calculates the value of all discriminant items for each of the mutation sites based on the ACMG guidelines, and Score the discriminant items, perform a summary operation based on the scores, and classify the pathogenicity based on the summary; S5. Explain the classification of the pathogenicity, and use the explanation as a basis for classification, and use the explanation as a basis for classification. The classification results are compared with the interpretation results of the Clinvar and InterVar genetic variation database interpretation tools, and based on the comparison results, they are submitted to manual for further review and confirmation.

In the embodiment of the present invention, the variant site discovery module 1 includes a sequence comparison and mapping unit, a sequence data preprocessing unit, and a SNPs and small fragment INDELs variation discovery unit; the sequence comparison and mapping unit is used to receive the sequence data composed of Raw sequencing data, and map the original sequencing data to the reference genome; the sequence data preprocessing unit is used to preprocess the mapped sequence data; the SNPs and small fragment INDELs variation discovery unit is used to identify the variant position relative to the reference genome point, and calculate the genotype of each variation site; wherein, the present invention utilizes the BWA-MEM algorithm to realize the mapping operation of the original sequencing data to the reference genome; the preprocessing operation is realized by the Picard tool, and the Picard tool includes AddOrReplaceReadGroups algorithm and MarkDuplicate Algorithm, the sequence data information is added to the mapped bam file by the AddOrReplaceReadGroups algorithm; the repeated information in the sequence data is marked by the MarkDuplicate algorithm to alleviate the bias caused by the data generation steps such as PCR amplification; and the Picard tool is used to The sequence data is sorted; finally, the GATK tool is used to recalibrate the base quality score; the specific process of using the variable site discovery module to find the variable site is as follows:

First use GATK to run the HaplotypeCaller method on each sample separately in GVCF mode to generate the GVCF intermediate file format; then use GATK's GenotypeGVCFs method to combine single-sample GVCF files to generate multi-sample VCF files; and use GATK's SelectVariants method to distinguish SNPS and INDELs; Next, use GATK's VariantRecalibrator algorithm and GATK's ApplyReclibration algorithm to correct the mass score of genetic variation to achieve filtering of variant sites; finally use GATK's CombineVariants method to combine SNPs and INDELs into a VCF file middle.

In the embodiment of the present invention, the variable site annotation module 2 realizes the information annotation of the variable site through the site annotation unit. Specifically, the dbNSFP database can be used to annotate all the variable sites or select a specific variable site. Specifically The process is as described below:

First parse the VCF file generated by the variant site discovery module, convert the VCF file of a single sample into a format separated by Tab keys, and adjust the position of INDELs to eliminate redundancy, and the output file contains the variant chromosome number, variant Coordinate start and end positions, reference alleles, replacement alleles; then use this file as input, through transcriptome code, transcriptome code to gene and protein code and other data resources, get the gene name, gene region, transcriptome code , protein coding and other information output files; then, through amino acid and sequence data resources, obtain the exon position where the variation is located, and classify the type of variation, and obtain the change of the mutated base and the change information of the amino acid; and analyze Whether the mutation occurs at the splice site, and obtain the code of the splice site; finally, with the help of the dbNSFP database resource, get the scores and predictions of the variant's SIFT, Polyphen2, MutationTaster, LRT, FATHMM, CADD, MetaSVM, Clinvar, InterVar function prediction tools result.

In the embodiment of the present invention, the data and resource loading module 3 includes an annotation data loading unit and an external resource loading unit, the annotation data loading unit is used to read and store data files; the external resource loading unit is used to read all external resources to be analyzed Resource files, for example, read the LOF (Lost of Function) gene list; read the missense gene list, read all pathogenic variation information in the Clinvar database, and the variation information includes the pathogenic chromosome and start coordinates of the genetic variation , base changes, amino acid changes and corresponding gene names; read the relevant information of the OMIM database, which includes the gene names corresponding to OMIM numbers, the list of OMIM numbers for recessive genetic diseases, the list of OMIM numbers for dominant genetic diseases, and the OMIM numbers Corresponding to the list of Orpha numbers; obtain the variation information related to PS4 from gwasdb, including chromosome number, position in hg19, SNP number, reference allele and replacement allele; read the benign protein domain used for PM1 judgment; Read the BP1 gene list to obtain the gene name; read the rmsk range used to judge PM4 and BP3, including chromosome number, start position and end position; read heterozygote and homozygote information for BS2 judgment; load dbscSNV Data information, data information includes chromosome number, position, reference allele, replacement allele, ada score, rf score.

In the embodiment of the present invention, the pathogenicity discrimination and classification module includes a pathogenicity discrimination unit and a pathogenicity classification unit, and the pathogenicity discrimination unit is used to calculate the value of all discrimination items in each mutation site; The taxonomic unit classifies all the variant sites according to the calculation results of the pathogenicity discriminant unit; for the discriminant item PVS1, first check whether the gene of the variant site is in the LOF (Lost of Function) gene list, and further check the variant site Point annotation information, if it is nonsense variation, frameshift variation, +/-1 or 2 splice sites, start codon, single or multiple exon deletion, then PVS1 is established, and the value is 1; for Discriminant item PS1, first check the type of gene where the mutation site is located, if it is a missense mutation, further check whether the mutation exists in the loaded Clinvar pathogenic variant data, and then check whether the mutation causes different nucleic acid changes, and judge whether it is For the same amino acid change, if it is true, PS1 is established, and the value is 1, otherwise, it is 0; for the discriminant item PS4, check whether the mutation rate of the affected gene is significantly increased compared with the prevalence rate of the control group, first check the variation Type, if it is a missense variation, then check whether the variation appears in the PS4 resource file list, if so, PS4 is established, and the value is 1, otherwise the value is 0; Therefore, missense variants in these regions tend to cause disease, first check the variant type, if it is a missense variant, further check whether the variant appears in the PM1 benign protein domain resource file list, if so, then PM1 is not established, the value is 0, otherwise the value is 1; for the discriminant item PM2, first check whether the variation is missing in the control group such as ESP (Exome Sequencing Project), G1000 (1000 Genomes Project), EXAC (Exome Aggregation Consortium), if missing Then PM2 is established, and the value is 1; check the OMIM recessive genetic disease data resource list, if it is a recessive genetic disease, then check whether the variation is an extremely low frequency in the above-mentioned control group, and the present invention uses 0.005 as the threshold of frequency; For the discriminant item PM4, check the RMSK database from the UCSC genome browser, if the mutation belongs to the non-frameshift INDEL of the non-repeated region, or the mutation type is Stop Loss, then PM4 is established, and the value is 1, otherwise the value is 0; for Discriminant item PM5, first check the variant type, if it is a missense variant, further check whether the variant exists in the loaded Clinvar pathogenic variant data, and then check whether the variant leads to different nucleic acid changes, if it leads to different amino acid changes, if If yes, PM5 is established, and the value is 1; otherwise, the value is 0; for the discriminant item PP2, first check the variation type, if it is a missense variation, further check whether the missense variation is a common cause of disease, and whether the gene where the variation is located is There are very few benign mutations, and the pathogenic data of Clinvar is used here as the basis for common disease mechanisms. If yes, PP2 is established, and the value is 1, otherwise, the value is 0; for the discriminant item PP3, multiple calculation evidences (sift, polyp2_hvar, lrt, mu_taster, mu_assessor, fathmm, cadd, metasvm) support detrimental effects of mutations, if only 1 computational evidence predicts benign, and less than 3 are unknown classifications; if ≤ 3 computational evidences have unknown outcomes, all others If the calculated evidence can predict harmfulness, then PP3 is established, and the value is 1; otherwise, the value is 0; for the discriminant item PP5, check the clnsig item in the variation label, if the clnsig value is equal to the possible pathogenicity or pathogenicity, then PP5 is established, and the value is 0 The value is 1, otherwise the value is 0; for the discriminant item BP1, first check the mutation type, if it is a missense mutation, further check whether the truncated mutation of the gene where the mutation is located can cause disease, here by checking whether the gene where the mutation is located is in the BP1 gene list If it is, BP1 is established, and the value is 1; otherwise, the value is 0; for the discriminant item BP3, check the RMSK database from the UCSC genome browser, if the mutation belongs to the non-frameshift INDEL of the repeated region, then BP3 is established, The value is 1, otherwise the value is 0; for the discriminant item BP4, multiple pieces of computational evidence (sift, polyp2_hvar, lrt, mu_taster, mu_assessor, fathmm, cadd, metasvm) support the harmful effects of mutations, if only 1 computational evidence predicts Harmful, and less than 3 are unknown classifications; if less than or equal to 3 calculation evidences are unknown results, all other calculation evidences will predict benign, then BP4 is established, and the value is 1, otherwise the value is 0; for the discriminant item BP6 , check the clnsig item in the variation annotation, if the clnsig value is equal to the possible benign or benign, then BP6 is established, and the value is 1, otherwise the value is 0; for the discriminant item BP7, if the synonymous variation has no effect on splicing, and the nucleoside If the acid position is not highly conserved, then the variant is classified as benign, and BP7 is established with a value of 1. When the predicted variation has no effect on splicing, both dbscSNV_RF_SCORE and dbscSNV_ADA_SCORE should be less than 0.6. The prediction of nucleotide conservation is retrieved from the dbnsfp30a database, and the GERP++ score is required to be greater than 2 to indicate that the nucleotide is highly conserved; for the discriminant item BA1, check whether the allele frequency of ESP, G1000, ExAC or SEC in the variant annotation is high For the specified threshold, the threshold here is set to 0.05, if it is greater than the threshold, BA1 is established, and the value is 1, otherwise it is 0; for the discriminant item BS1, in the ExAC browser, if the adjusted allele frequency is greater than the specified threshold, Here the threshold is set to 0.01, then BS1 is established, the value is 1, otherwise the value is 0; for the discriminant item BS2, first check the variation type, if it is a missense variation, further check whether the variation satisfies a homozygous, dominant, or X-linked diseases, if satisfied, then BS2 is established, and the value is 1, otherwise, the value is 0; the pathogenicity classification unit classifies according to the following rules according to the calculation results of the above-mentioned discrimination units, if PVS1 is equal to 1, and The number of PS is greater than or equal to 1, or the number of PM is greater than or equal to 2, or the number of PM is equal to 1 and the number of PP is equal to 1, or the number of PP is greater than or equal to 2, then the classification result is pathogenic; if If the number of PS is greater than or equal to 2, the classification result is still pathogenic; if the number of PS is equal to 1, if the number of PM is greater than or equal to 3, the classification result is still pathogenic; if the number of PS is equal to 1, if If the number of PM is equal to 2 and the number of PP is greater than or equal to 2, the classification result is still pathogenic; if the number of PVS1 is equal to 1 and the number of PM is equal to 1, or the number of PS is equal to 1 and the number of PM is equal to 1 or 2, Or the number of PS is equal to 1 and the number of PP is greater than or equal to 2, or the number of PM is greater than or equal to 3 or the number of PM is equal to 2 and the number of PP is greater than or equal to 2, then the classification result is possibly pathogenic; if the number of BA is equal to 1 Or the number of BS is greater than or equal to 2, then the classification result is benign; if the number of BS is equal to 1 and the number of BP is equal to 1, or the number of BP is greater than or equal to 2, then the classification result is possibly benign; other In this case, the classification result is indeterminate in significance.

In the embodiment of the present invention, the result interpretation and verification module 5 includes a result interpretation unit and a verification unit, and the result interpretation unit is used to provide the classification basis of the discrimination results and classification results of the pathogenicity discrimination and classification module, and provide a visual interface , for manual reference, and make the pathogenicity judgment of genetic variation according to the actual situation; the verification unit is used to compare the pathogenicity classification of genetic variation with the results of Clinvar and InterVar genetic variation data interpretation tools. If the results of the variation data interpretation tool are compared and the comparison results are inconsistent, it needs to be highlighted, and the manual will further review and determine the classification.

The data processing system and method for pathogenicity classification of genetic variation in coding regions of the present invention, the processing system consists of a variation site discovery module, a variation site annotation module, a data and resource loading module, a pathogenicity discrimination and classification module and a result The interpretation and verification modules are connected in sequence, and the variable site discovery module completes the search for all variable sites in the coding region, and annotates all the information in the variable site through the variable site annotation module to generate data corresponding to each variable site file, and then the data and resource loading module reads the data file and the external resource file used for pathogenicity discrimination; then, through the pathogenicity discrimination and classification module, the specific value of the discriminant item of each mutation site is calculated, Score each discriminant item, aggregate all discriminant items according to the score, and classify the pathogenicity of all genetic variations according to the summary; finally, the result interpretation and verification module gives the pathogenicity of all genetic variants. According to the basis of classification, it is compared with the results of genetic variation data interpretation tools. If the comparison is inconsistent, further review and confirmation will be performed manually again; compared with the prior art, the present invention can target gene sets, global Exome sequencing data realizes semi-automatic and systematic processing of large-scale samples and massive variation site information; the invention integrates a variation site discovery module, a variation site annotation module, a data and resource loading module, and a disease-causing The entire data processing process is standardized and systematic; the invention can accelerate the analysis speed of genetic disease genetic data, greatly improve the work efficiency of genetic disease data interpreters, and avoid Errors caused by cumbersome processing.

The above are only preferred embodiments of the present invention, but do not limit the scope of patents of the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, it can still understand the foregoing aspects The technical solutions described in the specific embodiments are modified, or some of the technical features are equivalently replaced. All equivalent structures made by utilizing the contents of the specification and drawings of the present invention and directly or indirectly used in other related technical fields are also within the protection scope of the patent of the present invention.

Claims (9)

1. A data processing system for genetic variation pathogenicity classification of a coding region is characterized by being constructed based on ACMG (access control message) guidelines as theoretical bases, comprising a variation site discovery module, a variation site annotation module, a data and resource loading module, a pathogenicity distinguishing and classifying module and a result interpreting and verifying module which are sequentially connected, aiming at sequencing data of a target gene set and a whole exome, semi-automatic and systematic processing of large-scale samples and massive variation site information is realized, wherein:

a mutation site discovery module for searching the specific position of the genetic mutation pathogenicity mutation site in the coding region; the variant sites comprise SNPs and INDELs of small fragments;

the variant locus annotation module is used for carrying out information annotation on variant loci and generating a data file corresponding to each variant locus; the information annotation comprises the chromosome where the variation locus is located, a reference allele, a replacement allele, the location of an exon, the rarity, the gene where the variation locus is located, amino acid change, computation scores and prediction results of variation harmfulness by various computational tools, and annotation of variation frequency information in different crowds;

the data and resource loading module is used for reading an external resource file and the data file which are judged by pathogenicity; the external resource file comprises a gene list for judging pathogenicity, clinvar, OMIM, dbscSNV and dbNSFP databases; the annotation data loading unit is used for reading and storing the data file;

the pathogenicity judging and classifying module is used for calculating values of all discriminants in each variation site in the data file, scoring each discriminant, summarizing all discriminants according to the scores, and classifying pathogenicity of all genetic variations according to the summarizing condition; the discriminant items comprise PVS1, PS4, PM1, PM2, PM4, PM5, PP2, PP3, PP5, BS1, BS2, BP1, BP3, BP4, BP6, BP7 and BA1;

the result interpretation and verification module comprises a result interpretation unit and a verification unit, wherein the result interpretation unit is used for providing the discrimination result of the pathogenicity discrimination and classification module and the classification basis of the classification result, providing a visual interface for manual reference, and performing pathogenicity discrimination of genetic variation according to the actual situation; the verification unit is used for comparing the pathogenicity classification of genetic variation with Clinvar and InterVar genetic variation data interpretation tool results, and if the comparison result of the comparison between the pathogenicity classification of genetic variation and Clinvar and InterVar genetic variation data interpretation tool results is inconsistent, the pathogenicity classification of genetic variation and the InterVar genetic variation data interpretation tool results need to be marked emphatically, and the verification unit is used for further auditing manually and determining the classification.

2. The data processing system for classification of pathogenicity of genetic variation in coding region according to claim 1, wherein the variation site discovery module comprises a sequence alignment and mapping unit, a sequence data preprocessing unit, and a SNPs and small fragment INDELs variation discovery unit; the sequence alignment and mapping unit is used for receiving original sequencing data consisting of sequence data and mapping the sequence data to a reference genome; the sequence data preprocessing unit is used for preprocessing the sequence data mapped on the reference genome; the SNPs and small-fragment INDELs mutation discovery unit is used for identifying mutation sites of the preprocessed sequence data relative to a reference genome and calculating the genotype of each mutation site.

3. The data processing system for classification of pathogenicity of genetic variation in coding region according to claim 2, wherein the input of the variant site discovery module is a raw sequencing data file in fastq format and the output of the variant site discovery module is a vcf format file containing all variant sites.

4. The data processing system for classification of pathogenicity of genetic variation in coding region according to claim 3, wherein said variation site discovery module performs mapping of said raw sequencing data using BWA-MEM algorithm; the mutation site discovery module realizes search operation of the mutation site by using a GATK tool.

5. The data processing system of claim 1, wherein said mutation site annotation module comprises a site annotation unit for annotating said SNPs and said small fragments INDELs and for enabling selection of said mutation sites for annotation of information; the specific process is as follows: firstly, analyzing a VCF file generated by a mutation locus discovery module, and outputting the file to contain a mutation chromosome number, a mutation coordinate starting point end point position, a reference allele and a replacement allele; then, the file is used as input to obtain an output file containing the information of the gene name, the gene region, the transcriptome code and the protein code; then, obtaining the position of the exon where the variation exists through amino acid and sequence data resources, and classifying the type of the variation to obtain the variation of the variation base and the variation information of the amino acid; analyzing whether variation occurs at the splice sites and obtaining the codes of the splice sites; and finally, obtaining the scores and the prediction results of the variant SIFT, polyphen2, mutationTaster, LRT, FATHMM, CADD, metaSVM, clinvar and InterVar function prediction tools by means of dbNSFP database resources.

6. The data processing system for classification of pathogenicity of genetic variation in coding region according to claim 1, wherein said data and resource loading module comprises an annotation data loading unit and an external resource loading unit, said annotation data loading unit is configured to read and store said data file; the external resource loading unit is used for reading all external resource files to be analyzed.

7. The data processing system of claim 1, wherein the pathogenicity judging and classifying module comprises a pathogenicity judging unit and a pathogenicity classifying unit, and the pathogenicity judging unit is configured to calculate values of all the judging items in each of the mutation sites; and the pathogenicity classification unit classifies all the mutation sites according to the calculation result of the corresponding pathogenicity judgment unit.

8. A data processing method for classification of pathogenicity of genetic variation of coding region, which is applied to the data processing system for classification of pathogenicity of genetic variation of coding region as claimed in any one of claims 1 to 7, the method comprising the steps of:

s1, inputting original sequencing data consisting of sequence data to a mutation site discovery module, mapping the sequence data to a reference genome by using a BWA-MEM algorithm, preprocessing the mapped sequence data by using a Picard tool, and finding out a mutation site of the sequence data by using a GATK tool;

s2, performing information annotation on all the variant loci by using the variant locus annotation module to generate a data file corresponding to each variant locus;

s3, reading the data file by using the data and resource loading module, and simultaneously reading an external resource file for pathogenicity judgment;

s4, calculating values of all discrimination items of each mutation site based on ACMG guidelines, scoring the discrimination items, performing summary operation according to the scoring, and performing classification operation on pathogenicity based on the summary;

and S5, explaining the pathogenicity classification, taking the explanation as a classification basis, comparing a classification result with an interpretation result of a Clinvar and InterVar genetic variation database interpretation tool, and submitting the comparison result to manual work for further auditing and confirmation.

9. The method of claim 8, wherein in step S5, if the classification results are inconsistent with the interpretation results of Clinvar and intersar genetic variation database interpretation tools, the classification results are submitted to human review and validation, otherwise, classification corresponding to the pathogenicity is completed.

Images