COE: a general approach for efficient genome-wide two-locus epistasis test in disease association study.

摘要: The availability of high-density single nucleotide polymorphisms (SNPs) data has made genome-wide association study computationally challenging. Two-locus epistasis (gene-gene interaction) detection attracted great research interest as a promising method for genetic analysis complex diseases. In this article, we propose general approach, COE, efficient large scale gene-gene interaction analysis, which supports wide range tests. particular, show that many commonly used statistics are convex functions. From the observed values events in two-locus test, can develop an upper bound test value. Such only depends on single-locus and genotype SNP-pair. We thus group index SNP-pairs by their genotypes. This indexing structure benefit computation all statistics. Utilizing structure, prune most without compromising optimality result. Our approach is especially permutation test. Extensive experiments demonstrate our provides orders magnitude performance improvement over brute force approach.