Making the Most of Clumping and Thresholding for Polygenic Scores

Polygenic prediction has the potential to contribute to precision medicine. Clumping and Thresh-olding (C+T) is a widely used method to derive polygenic scores. When using C+T, it is common to test several p-value thresholds to maximize predictive ability of the derived polygenic scores. Along with this p-value threshold, we propose to tune three other hyper-parameters for C+T. We implement an efficient way to derive thousands of different C+T polygenic scores corresponding to a grid over four hyper-parameters. For example, it takes a few hours to derive 123,200 different C+T scores for 300K individuals and 1M variants on a single node with 16 cores.We find that optimizing over these four hyper-parameters improves the predictive performance of C+T in both simulations and real data applications as compared to tuning only the p-value threshold. A particularly large increase can be noted when predicting depression status, from an AUC of 0.557 (95% CI: [0.544-0.569]) when tuning only the p-value threshold in C+T to an AUC of 0.592 (95% CI: [0.580-0.604]) when tuning all four hyper-parameters we propose for C+T.We further propose Stacked Clumping and Thresholding (SCT), a polygenic score that results from stacking all derived C+T scores. Instead of choosing one set of hyper-parameters that maximizes prediction in some training set, SCT learns an optimal linear combination of all C+T scores by using an efficient penalized regression. We apply SCT to 8 different case-control diseases in the UK biobank data and find that SCT substantially improves prediction accuracy with an average AUC increase of 0.035 over standard C+T.