Table 1.

Performance Evaluation During 10‐Fold Cross Validation

S. No.	Feature	No. of Features	PCC
			Training/Testing, T683 (10×)				Validation, V76
			SVM	RF	IBk	K*	SVM	RF	IBk	K*
1	Amino acid composition (Mono)	20	0.59	0.61	0.44	0.41	0.64	0.64	0.42	0.41
2	Di‐peptide composition (Di)	400	0.61	0.60	0.47	0.43	0.66	0.62	0.47	0.45
3	C8 Binary profile (C8 Bin)	160	0.56	0.57	0.45	0.42	0.59	0.60	0.43	0.41
4	N8 Binary profile (N8 Bin)	160	0.51	0.54	0.45	0.43	0.48	0.60	0.45	0.43
5	Physicochemical properties (Physico)	315	0.59	0.54	0.46	0.44	0.63	0.68	0.46	0.45
6	Solvent accessibility (SA)	21	0.22	0.20	0.18	0.19	0.21	0.18	0.15	0.16
7	Secondary structure (SS)	3	0.18	0.18	0.16	0.17	0.19	0.16	0.17	0.18
8	1 + 2	420	0.60	0.61	0.47	0.45	0.67	0.62	0.48	0.48
9	3 + 4	320	0.59	0.62	0.51	0.48	0.62	0.65	0.52	0.50
10	1 + 2+5	735	0.63	0.61	0.52	0.51	0.70	0.64	0.54	0.51
11	3 + 4+5	635	0.63	0.60	0.51	0.50	0.72	0.67	0.52	0.50
12	1 + 2+3 + 4	740	0.61	0.62	0.51	0.49	0.67	0.63	0.51	0.50
13	1 + 2+3 + 4+5	1055	0.62	0.61	0.50	0.51	0.66	0.64	0.54	0.53
14	6 + 7	23	0.22	0.20	0.18	0.21	0.23	0.19	0.20	0.18
15	1 + 2+5 + 6+7	758	0.66	0.63	0.55	0.54	0.74	0.68	0.59	0.57
16	3 + 4+5 + 6+7	658	0.65	0.64	0.56	0.55	0.73	0.70	0.58	0.56

10‐Fold cross validation performance of predictive models on AVP dataset of 683 sequences (T⁶⁸³) and evaluation of performance of predictive models on validation dataset of 76 peptides (V⁷⁶) using SVM, RF, IBk, and K* MLTs.

Abbreviations: SVM: support vector machine; RF: random forest; IBk: instance‐based classifier (Weka); K*: KStar (Weka); T⁶⁸⁵: Training dataset of 683 AVPs; 10×: 10‐fold cross validation; V⁷⁶: independent dataset of 76 AVPs.