We performed a principal component analysis (PCA) based quantification of race using the African and European populations from the Human Genome Diversity panel as reference populations. The SNP set was trimmed using linkage disequilibrium (LD)-based SNP pruning, removing all SNPs for with high pairwise LD (R2>0.8), as in [59]. PCA was performed using smartpca, as implemented in EIGENSOFT [60]. Four samples were flagged as outliers and removed from all further analyses. As expected, the first principal component separated African from non-African individuals. We therefore used this loading vector as an estimated quantification of African ancestry for further analyses.
