Fig. 3From: Predicting the presence of coronary plaques featuring high-risk characteristics using polygenic risk scores and targeted proteomics in patients with suspected coronary artery diseasePredictive performance of individual and combined models. Models included features from up to three feature groups, as shown on the x-axis, revealing the CRF + GPSMult model to be the most predictive in the full cohort. Stratifying patients by age group resulted in improved GPSMult performance in the group ≤ 55 years of age, while the CRF + GPSMult models were best in both groups. AUC, area under the curve; CRF, clinical risk factor; GPSMult, multi-trait multi-ancestry genome-wide polygenic score; Full, CRF + protein + GPSMultBack to article page