Copy number variation detection method based on neural network and improved particle swarm algorithm

• Main Authors: LI HUANRONG, WANG XUAN, LI JIAKE, XU YING, YAN MEI, CHEN PAN, GENG XIAOXUE, ZHANG YONG, SHU JINLONG, MEI LI, LI JUNQING, LIANG XIANJUAN, WANG SHOUFENG
• Format: Patent
• Language: Chinese; English
• Published: 16.08.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The invention relates to the technical field of deep learning, in particular to a copy number variation detection method based on a neural network and an improved particle swarm algorithm. The method comprises the steps that a to-be-detected sample is input into a neural network model optimized based on an improved particle swarm algorithm, and the neural network model determines the difference value of the occurrence rate between a sequencing sample and a reference genome, the GC content, the base quality, the correlation between adjacent sites of the genome and the number of repeated read segments in the bin range; calculating a copy number variation state probability value of the to-be-tested sample, wherein the improved particle swarm algorithm comprises three different updating strategies; obtaining a plurality of copy number variation state probability values output by the neural network model; and selecting the target copy number variation state corresponding to the maximum probability value as the cur