Biclustering of Microarray Data with Multi-Objective Immune Optimization
Feifei Liu · China Journal of Bioinformatics · 2009
The development of DNA microarray technologies provides an efficient tool for the experimental study of gene expression.Analysis of those large scale genomics data has initially focused on clustering methods.Recently,biclustering techniques were proposed for revealing submatrices showing unique patterns.Multi-objective optimization approach,which optimizes simultaneously several objectives in conflict with each other,is very good for solving biclustering problem.This paper proposes a novel multi-objective immune optimization biclustering algorithm based on the clonal selection principle to mining biclusters from microarray data.Experimental are conducted on two real datasets,which shows that multi-objective immune optimization biclustering algorithm exhibits better and more stable performance than other multi-objective evolutionary biclustering algorithms.