We have also shown the improvement in mapping power that can be gained by the choice of genotype imputation strategy and characterized the source of gains in power. These findings are generally applicable to GWAS as, despite more than 900 studies that have detected disease susceptibility loci at genome-wide significant level (Hindorff et al. 2009), these discoveries have only explained a small portion of familial clustering for many traits (Manolio et al. 2009). Genotype imputation has been widely used in the past 2 yr to aid in fine mapping of candidate regions, to help improve power of GWASs, and to facilitate meta-analysis of results from studies using different genotyping platforms (Marchini et al. 2007; Servin and Stephens 2007; Li et al. 2009, 2010; Gaffney et al. 2012). We have shown here that 1000G imputation (based on 112 CEU [Utah residents {CEPH} with Northern and Western European ancestry] haplotypes available from the 1000 Genomes pilot phase) (1000 Genomes Project Consortium 2010) produces substantial increases in the number of loci that can be mapped, that it improves the fine resolution of genetic
