A method for detecting somatic mutations

Abstract

The invention discloses a method for detecting somatic cell mutation. The detection method includes the following steps: first, obtain a somatic cell sequence set; extract the characteristics of each somatic cell sequence in the somatic cell sequence set to describe genome candidate mutation sites, obtain a sample data set, and select from the sample data set The first training set and the second training set and test set; then, set up a fully connected neural network model; and use the first training set, the second training set and the test set to pair the fully connected neural network model Carry out training and verification to obtain the fully connected neural network model after training; finally, obtain the whole genome sequence to be detected, and extract the characteristics of the whole genome sequence to be detected to describe the candidate mutation sites of the genome to obtain the data to be detected; The data to be detected is input into the trained fully connected neural network model for detection, which realizes the detection of mutated genes, thereby improving the accuracy of tumor disease diagnosis.

Description

technical field [0001] The invention relates to the field of gene detection, in particular to a method for detecting somatic cell mutations. Background technique [0002] Current studies of cancer genomes typically use Next Generation Sequencing (NGS) to analyze tumors with single nucleotide variant (SNV) somatic mutations. Detection of somatic mutations in cancer using NGS typically involves sequencing tumor DNA and DNA from non-malignant (or normal) tissue (usually blood) from the same patient. Thus, cancer-focused NGS experiments differ significantly in experimental design from studies of Mendelian disorders or normal human variation. In cancer research, where reads from two matched samples need to be aligned with a reference human genome, unpredictable errors are likely to occur during the sequencing and alignment process, and mutations in some genes do not affect one determinant of cancer. Samtools, SOAPsnp, VarScan, SNVMix, GATK, VipR and other tools compare tumor a...

AI Technical Summary

Problems solved by technology

However, these simple alignment methods are not very accurate

Images