Figure 1.
Overview of the PRiMeR framework for disease risk prediction. (A) PRiMeR utilizes matched health metrics and genetic data from a cohort of healthy individuals. (B) It integrates these with disease-specific GWAS summary statistics from an external cohort. (C) The framework trains risk predictors to align genetic effects with those observed in disease outcomes, maintaining adherence to two-sample MR principles. (D) Posttraining, the model's accuracy in predicting disease risk is evaluated, for example, through the receiver operating characteristic curve against actual follow-up disease onset data.
