Table 2. Internal Cross-Validation Results of the Final CP Models and XGBoost Models^a.

model	ROC value	sensitivity	specificity
CP PCM model	0.762	0.773	0.750
CP OATP1B1	0.773	0.777	0.769
CP OATP1B3	0.800	0.833	0.767
CP OATP2B1	0.664	0.726	0.602
XGBoost PCM model	0.903	0.916	0.702
XGBoost models OATP1B1	0.847	0.867	0.651
XGBoost models OATP1B3	0.872	0.909	0.565
XGBoost models OATP2B1	0.764	0.799	0.627

^aShown are the 5-fold ROC values and the calculated sensitivity and specificity from the confusion matrix. In each column, the highest value is highlighted in bold.