Table 1.

Comparison of models for the public dataset

Cross validation using the public dataset (8201 structures)
Fragments	BAC	SEN	SPEC	PPV	NPV
SOHN	0.78	0.78	0.78	0.79	0.78
Random Forest	0.79	0.83	0.76	0.78	0.76
SVM	0.81	0.82	0.81	0.82	0.81
kNN	0.76	0.73	0.80	0.79	0.80
Decision Tree	0.77	0.76	0.79	0.79	0.79

BAC = balanced accuracy, SEN = sensitivity, SPEC = specificity, PPV = positive predictivity, NPV = negative predictivity.