Because the risk of confounding by population stratification may increase with sample size (i.e., confounded results become more significant with larger samples) [34], and because large sample GWAS are becoming increasingly common, another method has been developed that utilizes large samples and thousands of markers throughout the genome to adjust for population structure. Eigenstrat analysis [35,36] uses principal components analysis to explicitly detect and adjust for population stratification on a genome-wide scale in large sample sizes in a computationally efficient manner. This method may be preferred over a stratified analysis because the combined sample often yields more powerful statistical tests, even after adjusting for significant eigenvectors [37]. Eigensoft is freely available open-source software for conducting Eigenstrat analyses, available online (Table 1). Running Eigensoft requires dense genotyping coverage. We recommend using all the default options, including 100,000 randomly chosen high-quality markers. There are several SNPs in the HLA region on chromosome 6, in the lactase locus on chromosome 2, and in the inversion regions on 8p23 and 17q21.31 common in populations of European ancestry [38] that are sources of stratification that
