Biclustering of gene expression data using Particle Swarm Optimization integrated with pattern-driven local search

Abstract

Biclustering is of great significance in the analysis of gene expression data and is proven to be a NP-hard problem. Among the existing intelligent optimization algorithms used in the gene expression data analysis, most concentrate on the global search ability but ignore the inherent trajectory information of gene expression data, so the search efficiency is low. In this paper, a pattern-driven local search operator is incorporated in the binary Particle Swarm Optimization (PSO) algorithm in order to improve the search efficiency. Experiments show that our approach is valid.

HGAB3C: A new hybrid global optimization algorithm

“…success-history based adaptive differential evolution (SHADE), was modified to incorporate linear population size reduction in L-SHADE [11] and proved to be a very good optimization algorithm on the CEC-2014 test bench. Li et al also incorporated a pattern-driven local search operator in binary particle swarm optimization to improve its search efficiency [12]. Buzok et al [13] and Yavuz et al [14] also evaluated their hybrid algorithms, SHADE4 and SSEABC, respectively, on the CEC-2014 test suite and found them to be better than their parent algorithms.…”