An Improved Particle Swarm Optimization with Dynamic Scale-Free Network for Detecting Multi-omics Features

摘要

Along with the rapid development of high-throughput sequencing technology, a large amount of multi-omics data sets are generated, which provide more opportunities to understand the mechanism of complex diseases. In this study, an improved particle swarm optimization with dynamic scale-free network, named DSFPSO, is proposed for detecting multi-omics features. The highlights of DSFPSO are the introduced scale-free network and velocity updating strategies. The scale-free network is employed to DSFPSO as its population structure, which can dynamically adjust the iteration processes. Three types of velocity updating strategies are used in DSFPSO for fully considering the heterogeneity of particles and their neighbors. Both gene function analysis and pathway analysis on colorectal cancer (CRC) data show that DSFPSO can detect CRC-associated features effectively.