To assess the role of genetic variation in global ancestry-associated DEGs, we initially mapped main effect cis-eQTLs in BAs (n = 120, 45, 121 and 131 for the caudate nucleus, dentate gyrus, DLPFC and hippocampus, respectively) examining genetic variants within ±500 kb of each feature (gene, transcript, exon and junction). To improve detection, we applied mash and identified at least one cis-eQTL for 13,857 genes (‘eGenes’) across brain regions (LFSR < 0.05; n = 10,867 for the caudate nucleus; n = 11,664 for the dentate gyrus; n = 11,173 for the DLPFC; and n = 10,408 for the hippocampus; Supplementary Table 3 and Supplementary Data 6). Most of these eGenes (64.1%; Fig. 4a) were shared across all brain regions with only about 0.25–14.5% showing brain region specificity. However, when considering the direction of effect, more than 96% showed sign matching across brain regions (Fig. 4b).
