Tools for efficient epistasis detection in genome-wide association study

We have developed a set of three efficient programs, FastANOVA, COE and TEAM, that support epistasis test in a variety of problem settings in GWAS. These programs utilize permutation test to properly control error rate such as family-wise error rate (FWER) and false discovery rate (FDR). They guarantee to find the optimal solutions, and significantly speed up the process of epistasis detection in GWAS.A web server with user interface and source codes are available at the website http://www.csbio.unc.edu/epistasis/ webcite. The source codes are also available at SourceForge http://sourceforge.net/projects/epistasis/ webcite.Genome-wide association study (GWAS) examines the genetic variants across the entire genome to identify genetic factors associated with observed phenotypes. It has been shown to be a promising design to locate genetic factors causing phenotypic differences [1,2]. Since most traits of interest are complex, finding gene-gene interaction has received increasing attention in recent years [3,4]. Unlike single-locus approaches, which test and estimate the association between the phenotype and one marker (or SNP) at a time, two-locus epistasis detection approaches search for associations between the phenotype and each SNP-pair.In general, there are two challenges in epistasis detection. The first is to develop statistical test that can effectively capture the interaction between SNPs. The second challenge is to reduce the computational burden since there are an extremely large number of SNP-pairs need to be tested in the whole genome. The computational challenge is further compounded by the multiple testing problem. Controlling family-wise error rate (FWER) and false discovery rate (FDR) are two standard approaches for controlling error rates [5]. With large number of SNPs correlated, permutation test is preferred over simple Bonferroni correction [6], which is often to conservative. The idea of permutation procedure is to randomly shuffle the phenotype