Our simulation finding that prediction accuracy of a European-derived PRS linearly decreases with increasing proportion of African ancestry in admixed African and European populations is consistent with a recent study of height, where there was a 1.3% decrease for each 10% increase in African ancestry.8 This decrease in prediction accuracy has been attributed to LD and allele frequency differences, as well as differences in effect sizes across populations contributing to height.8 Our work adds further insights into this reduction in PRS accuracy, showing that (1) it exists in the absence of trans-ancestry effect size differences consistent with previous theoretical models that did look at admixture,2,5 and (2) variants selected from an African population may not have these same biases. Recent work found that known GWAS loci discovered in Europeans have allele frequencies that are upwardly biased by 1.15% in African ancestry populations, which results in a misestimated PRS, a phenomenon that likely arises due to population bottlenecks and ascertainment bias from GWAS arrays.25 In our simulation study, which was not impacted by ascertainment bias, we show that GWASs in African
