Extended Data Fig. 4: The effect of different metrics of GD on the correlation between GD and accuracy.
From: Polygenic scoring accuracy varies across the genetic ancestry continuum
The y-axis \(-{cor}\left({r}_{i}^{2}\,,{d}_{i}\right)\) is the correlation between the GD and PGS accuracy; a larger correlation means GD has a better prediction of accuracy. The x-axis are different GD metrics: (1) GD based on PCA with varying number of PCs (from J = 1 to J = 20) and (2) GD based on GRM using pruned PCA SNPs only or all SNPs in PGS models. The GRM GD is computed as \({d}_{i}({\rm{GRM}})=\sqrt{\frac{1}{K}{\sum }_{k=1}^{K}{\left({x}_{i}-{x}_{k}\right)}^{2}}\), where \({x}_{i}\) is the standardized genotype of \({i}_{{th}}\) testing individual and \({x}_{k}\) is the standardized genotype of \({k}_{{th}}\) training individual.
