Table 4.

The performance of different kinds of feature descriptors in non-redundant dataset by random forest method

Features	ACC	SN	SP	AUC	MCC	F1
OAAC	0.849	0.856	0.817	0.900	0.581	0.904
Dipeptide = 0	0.872	0.892	0.780	0.910	0.612	0.921
Dipeptide = 1	0.879	0.900	0.781	0.912	0.625	0.925
Dipeptide = 2	0.870	0.885	0.797	0.908	0.612	0.918
AAindex	0.819	0.844	0.698	0.846	0.475	0.886
PSSM	0.836	0.855	0.744	0.884	0.527	0.896
All features	0.887	0.908	0.788	0.919	0.647	0.930