Table 2.

Five-Fold cross-validation performance of the three methods based on different descriptors.

Method	Descriptor(s)	AUC	Accuracy	Sensitivity	Specificity
RF	RDKit	0.87 ± 0.02	0.81 ± 0.02	0.55 ± 0.04	0.93 ± 0.02
RF	RDKit + RLM	0.89 ± 0.01	0.83 ± 0.01	0.62 ± 0.04	0.93 ± 0.01
XGBoost	RDKit	0.87 ± 0.02	0.81 ± 0.02	0.65 ± 0.02	0.89 ± 0.02
XGBoost	RDKit + RLM	0.89 ± 0.02	0.83 ± 0.01	0.70 ± 0.03	0.89 ± 0.02
GCNN	Graph	0.84 ± 0.02	0.80 ± 0.02	0.57 ± 0.07	0.91 ± 0.03
GCNN	Graph + RDKit	0.86 ± 0.02	0.80 ± 0.02	0.62 ± 0.07	0.89 ± 0.03
GCNN	Graph + RDKit + RLM	0.87 ± 0.03	0.82 ± 0.02	0.67 ± 0.05	0.89 ± 0.02