A storage and index processing method for genome mutation data

Abstract

The invention belongs to the technical field of gene data analysis, and in particular relates to a storage and index processing method for genome mutation data. The present invention provides a variety of fast retrieval methods such as retrieving a certain mutation according to mutation site information, retrieving its related mutation list according to single or multiple gene names, retrieving its related mutation list according to single or multiple genome regions, etc. In the field of disease-assisted diagnosis, it is very convenient to quickly find pathogenic sites, and in the field of tumor targeted drug guidance, it is very convenient to quickly find targeted drugs related to tumor mutations, which significantly reduces the processing time of genomic data analysis and interpretation, and greatly reduces The analysis of genomic mutation data is difficult.

Description

technical field [0001] The invention belongs to the technical field of gene data analysis, and in particular relates to a storage and index processing method for genome mutation data. Background technique [0002] In the prior art, genome mutation data is generally stored locally in vcf / bcf file format, and the index of vcf / bcf file is based on the binning indexing algorithm and file pointer to index genome mutation data by region, and only supports vcf / bcf output format . [0003] There are many data storage and indexing technologies, but at present, the storage of genomic mutations is not used in mainstream SQL or NoSQL databases. Usually mutation screening is realized by commonly used vcftools, vcflib, bcftools or self-developed scripts, the functions are relatively single, and personalized screening and customization are difficult. [0004] Although GA4GH defines mutation data to obtain mutation data through RESTful API, it does not define the specific storage format a...

AI Technical Summary

Problems solved by technology

[0006] 1) vcf / bcf files can only store a small number of samples in the form of files, and the indexing method based on file pointers is not efficient. In addition to indexing by genomic region, it is difficult to index by mutation site ID, gene name, etc. to index

[0007] 2) When the sample size reaches a certain level, it is difficult to calculate the mutation frequency of a specified site in a large number of samples in real time, and it is difficult to screen out a set of mutations associated with a specified phenotype according to a certain phenotype, or Search for previously recorded pathogenic loci based on a disease name

[0008] 3) When the sample size increases, a large number of vcf / bcf files need to be used for storage, and the occupied storage resources also gradually increase

If you want to mine more data, parsing the vcf file will also become quite complicated

Images