THE USE OF THE ANT COLONY ALGORITHM FOR THE DETECTION OF MARKER ASSOCIATIONS IN THE PRESENCE OF GENE INTERACTIONS

In recent years there has been much focus on the use of single nucleotide polymorphism (SNP) fine genomemapping to identify causative mutations for traits of interest; however, many studies focus only on the marginal effects ofmarkers, ignoring potential gene interactions. Simulation studies have shown that this approach may not be powerfulenough to detect important loci when gene interactions are present. Although several attempts have been made to studypotential gene interaction, the number of SNP markers considered in these studies is often limited. Given the prohibitivecomputation cost of modeling interactions in studies involving a large number SNP, there is a need for methods that canaccount for potential gene interactions in a computationally efficient manner to be developed. In this study, the ant colonyoptimization algorithm (ACA) and logistic regression on large number of SNP genotypes were used. Our procedure wascompared to sliding window (SW/H), and single locus genotype association (RG) methods used in haplotype analyses. Abinary trait simulated using an epistatic model and HapMap ENCODE SNP genotypes was used to evaluate each algorithm.Results show that the ACA outperformed SW/H and RG under several simulation scenarios, yielding substantial increases inpower to detect genomic regions associated with the simulated trait.