Having analyzed the use of polygenic scores in different ancestry groups (above), we next assessed the performance of polygenic scores in multiple ancestry groups. Since most large-scale GWAS have been conducted in primarily (or exclusively) European ancestry individuals42, our a priori hypothesis was that polygenic scores would perform best among European ancestry individuals, and less well for other populations. Figure 2 provides an overview of polygenic score performance across ancestry groups. Results from all complex genetic phenotypes are analyzed together in order to increase the amount of data available for analysis. In Fig. 2, each point represents one within-study comparison between a non-European ancestry sample and the matched (within-study) European ancestry sample. The vertical black line represents equal performance in the non-European ancestry sample, as compared to the matched European ancestry sample from the same study.
