Application of Optimization Techniques for Gene Expression Data Analysis

The feature selection from gene expression data is the NP hard problem, few of evolutionary techniques give optimal solutions to find feature subsets. In this chapter, authors introduce some evolutionary optimization techniques and proposed a Binary Particle Swarm Optimization (BPSO) based algorithm for feature subset selection. The Feature selection is one of the important and challenging tasks for gene expression data where many traditional methods failed and evolutionary based methods were succeeded. In this study, the initial datasets are preprocessed using a quartile based fast heuristic technique to reduce the crude domain features which are less relevant in categorizing the samples of either group. The experimental results on three bench-mark datasets vis-a-vis colon cancer, defused B-cell lymphoma and leukemia data are evaluated by means of classification accuracies. Detailed comparative studies with some of popular existing algorithms like Genetic Algorithm (GA), Multi Objective GA are also made to show the superiority and effectiveness of the proposed method. Application of Optimization Techniques for Gene Expression Data Analysis