An incomplete dominance genetic algorithm approach to microarray data analysis

Abstract

We address the problem of analyzing the vast amount of data involved in microarray studies. The finality is to discover, from a large pool of candidates, a limited number of genes that could be causally related with a specific pathology. In this context, we propose a new genetic algorithm (GA) approach for feature selection, with diploid number of chromosomes and an incomplete dominance model for genotype to phenotype mapping. We test our algorithm on a familiar data set for performance evaluation purposes.