Method for detecting multiple mutation types of genome

Abstract

The invention relates to a method for detecting multiple mutation types of a genome. A BAM file is used as an input file, a tumor tissue sample is detected, multiple mutation information of differentsample types is detected, and detection of the tumor tissue sample comprises the steps as follows: 1) detection of mononucleotide mutation; 2) detection of insertion and deletion; 3) detection of structural variants; 4) detection of complex variants. According to the method for detecting multiple mutation types of the genome, insertion and deletion are detected with a GeneReader algorithm obtainedthrough combination of a supervised method and an unsupervised method, the local comparison precision is optimized, linear growth with increase of the sequencing depth is realized in the aspect of speed, the detection effect is good, the detection result is accurate, and actual application demands can be well met.

Description

technical field [0001] The invention belongs to the technical field of gene detection, and in particular relates to a method for detecting multiple mutation types of a genome. Background technique [0002] With the maturity of next-generation sequencing technology and the decline in price, genome sequencing has been widely used in the medical field. Taking tumor drug use and clinical trials as an example, researchers can take cancer tissue or blood samples to study the effectiveness of drugs for cancer types with different biomarkers, the process mechanism of cancer metastasis and the markers produced, screening Check the markers of early tumor or recurrence. [0003] Genes have a variety of mutation types, generally can be divided into single nucleotide mutation (SNV), DNA fragment insertion (Insertion) and deletion (Deletion), and more complex copy number variation (Copy number variant) and structural variation (Structural variant). Different cancer types have different...

AI Technical Summary

Problems solved by technology

For the GATK algorithm, first of all, it is not sensitive enough to detect extremely low mutation rate sites in blood

Second, the model parameters used by GATK are trained using tissue data and are not suitable for blood samples

Third, for the efficiency of the algorithm, GATK will use random downsampling (downsampling) to reduce the amount of data, and this processing method will make us miss the real mutation information

Fourth, GATK does not have flexible and diverse filtering parameters, and cannot make good use of multi-parameter combinations to filter out false positive mutation sites