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

Table 7

Performance for independent datase2 (trained on expanded benchmark dataset1100).

Method	ACC (%)	MCC	SE (%)	SP (%)	F1-M (%)
DNAbinder(P21)	75.11	0.51	64.41	85.27	71.59
DNAbinder(P400)	81.65	0.65	67.14	95.42	78.09
DNA-Prot	79.07	0.60	65.32	92.03	75.19
iDNA-Prot	75.60	0.54	57.01	93.14	69.41
enDNA-Prot	83.48	0.67	84.29	82.72	83.21

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