Second generation PLINK49 was used to construct polygenic scores for each phenotype for the 1000Genomes participants (N = 1940, see below for exclusions), using 13 thresholds, pT, as follows: p < 5 × 10−8, p < 1 × 10−6, p < 1 × 10−4, p < 1 × 10−3, p < 1 × 10−2, p < .05, p < .1, p < .2, p < .3, p < .4, p < .5, p < .75, p < = 1. The statistical package R51 was used for plotting and statistical analyses (t-tests, correlations). For correlations between polygenic scores and principal components, see Supplementary Data 3. Height phenotype data were downloaded from a compiled table of average heights, for males and females, by countries. Heights for males and females were averaged. Certain populations were excluded from the analysis of correlations between polygenic risk scores for height and height phenotypes for three reasons: Four populations were excluded due to lack of height phenotype data: Puerto Rican in Puerto Rico (PUR; N = 105), Bengali in Bangladesh (BEB; N = 86), Punjabi in Lahore,
