An Efficient Tagging SNP Selection Method Using Normalized Mutual Information and Joint Entropy

Abstract

Single nucleotide polymorphisms (SNPs) are ideal markers for candidate gene association studies for their abundance and density. The selection of a minimal subset of SNPs which can represent the haplotype diversity optimally is both important and valuable and it has received extensive study in recent years. In this paper, we develop a new LD measure and an approach for tagSNP selection, using the concept of mutual information and joint entropy in the information theory. An algorithm based on the objective of capturing the haplotype diversity and SNP association is build. Experimental results on real datasets illustrated the efficiency of the proposed method and its ability to capture LD pattern elaborately. Informative SNPs with low redundancy can be selected using our method.