HMMPred: Accurate Prediction of DNA-Binding Proteins Based on HMM Profiles and XGBoost Feature Selection

Table 3

Performance comparison before and after feature selection.

Feature selection	CV methods	ACC	SN	SP	MCC	AUC
Before	10-fold	0.8034	0.8147	0.7927	0.6071	0.8720
	Jackknife	0.8015	0.8127	0.7909	0.6034	0.8805
RF	10-fold	0.8221	0.8243	0.8200	0.6441	0.8819
	Jackknife	0.8267	0.8262	0.8272	0.6533	0.8946
XGBoost	10-fold	0.8371	0.8301	0.8436	0.6738	0.8896
	Jackknife	0.8390	0.8398	0.8382	0.6778	0.9018