Table 3. The Performance of SVM Models on PSSM based training dataset D3 & D4.

Features	C	g	SN	SP	ACC	MCC	AUC/ROC
AC&CC	32768	0	90.97	87.1	90.30	0.78	0.94
AC&HC	8	0.03	95.83	91.0	94.78	0.87	0.97
AC&HC&HYC	2	0.13	94.44	91.0	94.78	0.86	0.97
AC&HYC&CC	2048	0	91.67	90.32	91.42	0.82	0.96
AC&HYC	2048	0	91.67	88.7	91.04	0.8	0.95

The Performance of SVM Models on PSSM based training dataset D3 & D4 with different learning parameters on various hybrid models [γ (g) (in RBF kernel), c: parameter for trade-off between training error & margin] where SN–sensitivity, SP–specificity, ACC-accuracy, MCC–Matthews Correlation Coefficient, AUC/ROC-Area under curve/ Receiver Operating Curve.