The resulting genotype data was inspected for genotyping success rate and compliance with Hardy Weinberg equilibrium. Loci with less than 95% success rate or deviating from Hardy Weinberg equilibrium at less than p<0.01 were discarded. The surviving genotypic data was then analyzed for linkage disequilibrium (D′) using Haploview [18] and position data provided by The National Center for Biotechnology (NCBI) dbSNP database, Build 127 (http://www.ncbi.nlm.nih.gov/projects/SNP/). Guided by this analysis, haplotypes were inferred using PHASE version 2.1.1 [19].
