Table 2.

Statistical results of the seven classification models based on 144 molecular features for the training (random fivefold cross-validation) and test sets

	Training set (random fivefold cross-validation)				Test set
	GA	BA	MCC	AUC	GA	BA	MCC	AUC
SVM	0.902	0.893	0.793	0.947	0.911	0.905	0.812	0.958
DNN	0.894	0.892	0.780	0.950	0.907	0.904	0.806	0.960
XGBoost	0.902	0.894	0.793	0.956	0.891	0.883	0.770	0.957
SGB	0.901	0.894	0.792	0.952	0.886	0.879	0.759	0.958
RLR	0.875	0.872	0.740	0.932	0.873	0.867	0.734	0.936
k-NN	0.863	0.862	0.717	0.913	0.857	0.856	0.705	0.917
NB	0.826	0.834	0.654	0.898	0.780	0.793	0.572	0.888
Consensus1	0.902	0.893	0.793	NA	0.903	0.897	0.797	NA
Consensus2	0.901	0.895	0.793	0.956	0.909	0.903	0.808	0.963

NA not available