Fig. 3: Calibration in empirical analyses of nine disorders.
From: Estimating disorder probability based on polygenic prediction using the BPC approach
Calibration of the BPC and the Pain et al.14 approach was evaluated using the Integrated Calibration Index (ICI) for nine disorders, while varying the prior disorder probability. The BPC approach was applied using two Bayesian PGS methods, PRScs (BPC-PRScs) and SBayesR (BPC-SBayesR). The BPC-PRScs approach achieves the lowest mean ICI values across all prior disorder probabilities. BPC-SBayesR shows one fewer data points, as it did not converge for prostate cancer. Numerical values are presented in Supplementary Data 1. Error bars represent standard errors and their center represent means.
