Table 4.

Performance of five major types of features for the training and independent datasets.

Methods		Training		Independent
H. sapiens		RF	SVM	RF	SVM
	pCKSAAP	0.856	0.838	0.695	0.691
	CKSAAP	0.816	0.831	0.677	0.663
	AAindex	0.739	0.728	0.759	0.755
	Binary	0.767	0.754	0.822	0.809
	PseAAC	0.819	0.822	0.658	0.649
H. capsulatum	pCKSAAP	0.789	0.792	0.638	0.634
	CKSAAP	0.788	0.783	0.619	0.607
	AAindex	0.712	0.722	0.658	0.666
	Binary	0.713	0.698	0.665	0.647
	PseAAC	0.759	0.743	0.612	0.614
M. musculus	pCKSAAP	0.801	0.788	0.637	0.634
	CKSAAP	0.777	0.767	0.646	0.651
	AAindex	0.648	0.655	0.679	0.672
	Binary	0.639	0.641	0.677	0.659
	PseAAC	0.711	0.722	0.609	0.611
E. coli	pCKSAAP	0.769	0.761	0.679	0.684
	CKSAAP	0.773	0.782	0.646	0.631
	AAindex	0.719	0.721	0.633	0.619
	Binary	0.689	0.674	0.619	0.607
	PseAAC	0.733	0.734	0.608	0.603
M. tuberculosis	pCKSAAP	0.708	0.712	0.688	0.679
	CKSAAP	0.689	0.675	0.664	0.671
	AAindex	0.667	0.658	0.656	0.655
	Binary	0.629	0.617	0.639	0.634
	PseAAC	0.643	0.634	0.629	0.617
S. cerevisiae	pCKSAAP	0.882	0.869	0.776	0.772
	CKSAAP	0.879	0.863	0.752	0.744
	AAindex	0.742	0.733	0.759	0.749
	Binary	0.741	0.745	0.798	0.787
	PseAAC	0.790	0.768	0.699	0.675
T. gondii	pCKSAAP	0.834	0.836	0.657	0.666
	CKSAAP	0.826	0.822	0.655	0.638
	AAindex	0.726	718	0.663	0.647
	Binary	0.744	0.745	0.679	0.671
	PseAAC	0.801	0.788	0.678	0.664
S. lycopersicum	pCKSAAP	0.842	0.836	0.649	0.642
	CKSAAP	0.833	0.824	0.648	0.637
	AAindex	0.753	0.765	0.644	0.629
	Binary	0.729	0.722	0.637	0.631
	PseAAC	0.801	0.783	0.678	0.658
T. aestivum	pCKSAAP	0.822	0.826	0.649	0.654
	CKSAAP	0.821	0.811	0.638	0.634
	AAindex	0.736	0.734	0.604	0.611
	Binary	0.726	0.719	0.612	0.596
	PseAAC	0.778	0.769	0.632	0.628

AUC values are used to assess the prediction performance.