Methods for Inferring Haplotype Blocks and Informative SNP Selection

Bjarni Halldorsson, Celera Genomics

The sequencing of the human genome has created tremendous opportunity for advances in pinpointing disease-related genes, predicting disease risk, and defining population structure and history. The most common form of variants among individuals are called single nucleotide polymorphisms (SNPs) and consist of a substitution of one nucleotide for another at a specific location. Cost effective technology has recently been developed for SNP genotyping. However, the human genome has millions of SNPs and we are left with the problem of choosing a subset of these SNPs for our studies. Due to strong correlations among SNPs a smaller fraction of them can determine the rest. Here, we present methods for choosing such an informative subset of SNPs. We further present results of an empirical study on the profile of SNP variation across human chromosomes 6, 21 and 22 in African-American and Caucasian populations.