Ant Colony Optimisation for Exploring Logical Gene-Gene Associations in Genome Wide Association Studies

In this paper a search for the logical variants of gene-gene interactions in genome-wide association study (GWAS) data using ant colony optimisation is proposed. The method based on stochastic algorithms is tested on a large established database from the Wellcome Trust Case Control Consortium and is shown to discover logical operations between combinations of single nucleotide polymorphisms that can discriminate Type II diabetes. A variety of logical combinations are explored and the best discovered associations are found within reasonable computational time and are shown to be statistically significant