We estimated the admixture proportion for each individual based on SNPs that were informative for ancestry using the STRUCTURE program (v.2.3.4)17. We selected 1,634 such SNPs based on genetic information downloaded from the 1000 Genomes CEU (Northern Europeans from Utah) and AFR (African ancestry superpopulation, including Esan, Gambian, Luhyu, Mende and Yoruba populations) samples. Markers were chosen based on the following criteria: (1) absolute difference (δ) in allele frequency between the two ancestry populations > 0.5; (2) r2 between each pair of SNPs < 0.1 within each population; (3) P > 0.01 to test the Hardy–Weinberg equilibrium within each population; and (4) successfully imputed in our brain samples (info > 0.8). The structure was run within a two-ancestry population model with 5,000 burn-in and 10,000 iterations.
