PRS models have great clinical potential: previous studies have shown that individuals with higher PRS have increased risk for disease8–12. In the future, polygenic risk assessment may become as common as screening for known mutations of monogenic disease, especially as it has been shown that individuals with severely high PRS may be at similar risk to disease as are carriers of rare monogenic mutations12. However, since PRS rely heavily on GWAS with large sample sizes to estimate effect sizes accurately, there is specific demand for the transferability of PRS from populations with larger GWAS to populations underrepresented by GWAS2,6–8,17,18,22. Here, we chose pruning and thresholding (P+T) as our PRS model6,8. P+T models select an independent subset of all SNPs genome wide by pruning away SNPs correlated by LD and then further thresholding on GWAS P value. We elected to use P+T rather than LDpred2,22 or AnnoPred21, which compute a posterior effect size estimate for all SNPs genome-wide based on membership to functional categories. With P+T, we can partition the genome by IMPACT-prioritized and deprioritized SNPs, whereas the assumptions of the
