enDNA-Prot: Identification of DNA-Binding Proteins by Applying Ensemble Learning

Table 6

Performance for independent dataset1 (trained on expanded benchmark dataset1100).

Method	ACC (%)	MCC	SE (%)	SP (%)	F1-M (%)
DNAbinder(P21)	72.93	0.52	42.24	100	57.39
DNAbinder(P400)	78.45	0.61	52.44	100	68.80
DNA-Prot	76.37	0.58	47.56	100	64.46
iDNA-Prot	76.92	0.58	50.00	99.00	66.13
enDNA-Prot	89.56	0.79	80.48	97.00	87.42

P400 and P21 denote the two vectorization methods PSSM-400 based DNAbinder and PSSM-21 based DNAbinder, respectively.