Polygenic risk prediction: why and when out-of-sample prediction R can exceed SNP-based heritability.

In polygenic score (PGS) analysis, the coefficient of determination (R) is a key statistic to evaluate efficacy. R is the proportion of phenotypic variance explained by the PGS, calculated in a cohort that is independent of the genome-wide association study (GWAS) that provided estimates of allelic effect sizes. The SNP-based heritability (h, the proportion of total phenotypic variances attributable to all common SNPs) is the theoretical upper limit of the out-of-sample prediction R. However, in real data analyses R has been reported to exceed h, which occurs in parallel with the observation that h estimates tend to decline as the number of cohorts being meta-analyzed increases. Here, we quantify why and when these observations are expected. Using theory and simulation, we show that if heterogeneities in cohort-specific h exist, or if genetic correlations between cohorts are less than one, h estimates can decrease as the number of cohorts being meta-analyzed increases. We derive conditions when the out-of-sample prediction R will be greater than h and show the validity of our derivations with real data from a binary trait (major depression) and a continuous trait (educational attainment). Our research calls for a better approach to integrating information from multiple cohorts to address issues of between-cohort heterogeneity.