Algorithm for Disease Association Studies Using Functionally Informative Haplotype Motif

Selecting a subset of genetic polymorphism data is considered to be an essential step for locating disease related genes. Haplotypes, contiguous sets of correlated single nucleotide polymorphisms (SNPs), may provide a promising way out for analyzing linkage data involved. At present, some flexible block- free "haplotype motif models are demonstrated to deduce conserved structure within haplotype. However, while those haplotype motifs may characterize true haplotype conservation patterns of a target genomic region, they do not necessarily contain functional significant SNPs directly associated with disease. In this paper, we address this challenge by redefining the mathematical model of haplotype motif. We present an integrative haplotype motif selection system which can simultaneously deduce motif not only statistically significant but also highly correlated to deleterious variation. To evaluate our system, we compared our new model with some existing models on finding informative haplotype motifs on real data set to show its advantages in disease association studies.