Issue No. 03 - May-June (2012 vol. 9)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TCBB.2011.145
Michael A. Mooney , Oregon Health & Science University, Portland
Beth Wilmot , Oregon Clinical and Translational Research Institute, Portland
Shannon K. McWeeney , Oregon Health & Science University, Portland
Enormous data collection efforts and improvements in technology have made large genome-wide association studies a promising approach for better understanding the genetics of common diseases. Still, the knowledge gained from these studies may be extended even further by testing the hypothesis that genetic susceptibility is due to the combined effect of multiple variants or interactions between variants. Here, we explore and evaluate the use of a genetic algorithm to discover groups of SNPs (of size 2, 3, or 4) that are jointly associated with bipolar disorder. The algorithm is guided by the structure of a gene interaction network, and is able to find groups of SNPs that are strongly associated with the disease, while performing far fewer statistical tests than other methods.
Biology and genetics, evolutionary computing and genetic algorithms, graphs and networks.
S. K. McWeeney, M. A. Mooney, The Bipolar Genome Study and B. Wilmot, "The GA and the GWAS: Using Genetic Algorithms to Search for Multilocus Associations," in IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 9, no. , pp. 899-910, 2011.