Table 1.

Prediction performance of the new strategy under five-fold cross-validation, in comparison with four component predictors, another four recently-designed predictors.

	DisEMBL	IUPred	VSL2	Espritz	PONDR-FIT	MFDp2	IUPred2	AUCpreD	This Work
Sens	0.440 ± 0.008	0.650 ± 0.003	0.817 ± 0.004	0.514 ± 0.009	0.713 ± 0.004	0.777 ± 0.004	0.640 ± 0.003	0.592 ± 0.006	0.807 ± 0.012
Spec	0.914± 0.002	0.874 ± 0.004	0.736 ± 0.003	0.939 ± 0.002	0.859 ± 0.004	0.859 ± 0.004	0.877 ± 0.004	0.909 ± 0.002	0.856 ± 0.007
Acc	0.779 ± 0.003	0.810 ± 0.003	0.759 ± 0.002	0.818 ± 0.004	0.817 ± 0.004	0.836 ± 0.003	0.810 ± 0.003	0.819 ± 0.003	0.842 ± 0.003
Acc-b	0.677 ± 0.006	0.762 ± 0.002	0.776 ± 0.002	0.726 ± 0.004	0.786 ± 0.003	0.818 ± 0.003	0.759 ± 0.002	0.751 ± 0.003	0.831 ± 0.004
MCC	0.410 ± 0.007	0.529 ± 0.006	0.504 ± 0.004	0.521 ± 0.007	0.561 ± 0.007	0.614 ± 0.006	0.526 ± 0.006	0.535 ± 0.006	0.635 ± 0.006
F1	0.531 ± 0.006	0.660 ± 0.003	0.658 ± 0.003	0.616 ± 0.006	0.689 ± 0.004	0.729 ± 0.003	0.657 ± 0.004	0.651 ± 0.005	0.744 ± 0.004
AUC_ROC	0.776 ± 0.004	0.823 ± 0.001	0.841 ± 0.003	0.886 ± 0.003	0.857 ± 0.003	0.879 ± 0.002	0.822 ± 0.001	0.869 ± 0.003	0.899 ± 0.004
AUC_PR	0.607 ± 0.007	0.675 ± 0.007	0.656 ± 0.020	0.752 ± 0.006	0.696 ± 0.004	0.629 ± 0.006	0.657 ± 0.004	0.716 ± 0.007	0.788 ± 0.010

Note Bene. The measures of predictor performance include: sensitivity (Sens), specificity (Spec), accuracy (Acc), balanced accuracy (Acc-b), Matthews Correlation Coefficient (MCC), F1 score, Area Under ROC Curve (AUC_ROC), and Area Under Precision-Recall Curve (AUC_PR). The highest value in each of these measures is in bold and highlighted (red).