Tumor gene point mutation feature extraction method combining contour coefficient and RJMCMC algorithm

Abstract

The invention provides a tumor gene point mutation feature extraction method combining a contour coefficient and an RJMCMC algorithm, and relates to the technical field of tumor gene feature extraction. The tumor gene point mutation feature extraction method combining the contour coefficient and the RJMCMC algorithm comprises the following steps: S1, data set acquisition: mutation types of a mutation data set comprise Somic SNV and Somic INDEL, and the Somic SNV and the Somic INDEL are subjected to overall statistics by using MuTect software; According to the tumor gene point mutation feature extraction method combining the contour coefficient and the RJMCMC algorithm, an input mode of annotation files is realized, use is convenient, early-stage data processing time is saved, the efficiency is improved, mutation spectrum 3D visual display is realized, researchers can visually see mutation conditions of each type from spatial vision, the comparison effect display of the type is is enhanced, and the contour coefficient is innovatively combined to construct the RJMCMCNMF model and algorithm implementation;code software device design is completed, and a software device for feature spectrum and gene association acquisition is achieved.

Description

technical field [0001] The invention relates to the technical field of tumor gene feature extraction, in particular to a feature extraction method for tumor gene point mutations combined with a contour coefficient and an RJMCMC algorithm. Background technique [0002] Cancer is a genetic disease caused by mutations in the cells of an organism. With the development of genetic testing technologies such as next-generation sequencing (NGS), it is found that these mutations are caused by the combination of specific mutational signatures, which often have known underlying processes, which can better provide information on cancer mechanisms, It also helps in the prevention and treatment of cancer. The human genome is composed of chromosomes, and each chromosome is composed of four different nucleotides - A / C / G / T. The four nucleotides actually form two pairs of A-T, C-G. When A is on one strand, T is on the other strand. When G is on one strand, C must be formed at the same positi...

AI Technical Summary

Problems solved by technology

However, in most human cancer types, the mutation signatures imprinted by DNA damage and repair processes are very limitedly characterized, and these methods have certain limitations, the functions are relatively single, and the analysis results for some data sets are not satisfactory, especially The results of small sample data or low-depth data have relatively large errors

Images