Joint Modeling of Effect Sizes for Two Correlated Traits: Characterizing Trait Properties to Enhance Polygenic Risk Prediction

Abstract

Recent years have witnessed a surge in the development of innovative polygenic score (PGS) methods, driving their extensive application in disease prevention, monitoring, and treatment. However, the accuracy of genetic risk prediction remains moderate for most traits. Currently, most PGSs were built based on the summary statistics from the target trait, while many traits exhibit varied degrees of shared genetic architecture or pleiotropy. Appropriate leveraging of pleiotropy from correlated traits can potentially improve the performance of PGS of the target trait. In this study, we present PleioSDPR, a novel method that jointly models the genetic effects of complex traits to characterize conditions under which considering pleiotropy enhances polygenic risk prediction. PleioSDPR models the joint distribution of effect sizes across traits, allowing SNPs to be null for both traits, causal for only one trait, or causal for both traits, while accommodating region-specific genetic correlations and unequal heritability between traits. Through extensive simulations and real trait applications, we demonstrate that PleioSDPR improves prediction performance compared with several univariant and multivariate PGS methods, especially when there is no validation dataset. For example, by incorporating information from schizophrenia or leg fat-free mass, PleioSDPR effectively improves the prediction accuracy of bipolar disease (14.2% accuracy gain) and hip circumstance (20.65% accuracy gain), respectively. Moreover, our findings demonstrate that traits exhibiting high genetic correlations and heritability, and low overlapping sample sizes contribute more to the improvement of prediction accuracy of the target trait. Overall, our study highlights the potential of PleioSDPR to enhance the accuracy of genetic risk prediction by leveraging pleiotropy and considering a broader spectrum of traits and diseases. These findings contribute to the understanding of polygenic risk prediction and underscore the importance of incorporating pleiotropic information for improved utilization in disease prevention and treatment strategies.