Predicting antifreeze proteins with weighted generalized dipeptide composition and multi-regression feature selection ensemble

Table 1 Evaluation based on four singer filters and LRMR-Ri for antifreeze proteins dataset with three models

Model	Evaluating indicator	Lasso	Ridge	Mic	Relief	LRMR-Ri
DT	RE (%)	77.65	90.59	88.82	92.94	95.88
	ACC (%)	76.93	92.39	89.81	91.34	93.21
	MCC	0.5328	0.8545	0.7932	0.8283	0.8627
SVM	RE (%)	77.65	95.88	87.06	94.71	97.06
	ACC (%)	82.79	93.56	90.75	93.21	95.56
	MCC	0.6574	0.9001	0.8138	0.8639	0.9105
SGD	RE (%)	76.47	88.82	81.87	87.13	91.76
	ACC (%)	81.87	85.38	83.04	87.13	90.53
	MCC	0.6675	0.7205	0.6698	0.7450	0.8332

ISSN: 1471-2105