A Survey of Computational Intelligence Techniques in Protein Function Prediction

Table 11

The results analysis of different classifiers to predict protein functions.

Computational intelligence based techniques		DNA	RNA	Membrane	Enzyme	Nuclear receptor	G-protein coupled receptor	Overall
Random forest	ACC	78.6	64.7	89.2	81.6	97.4	94.6	86.7
	MCC	0.74	0.71	0.86	0.74	0.94	0.97	0.84
Support vector machine	ACC	81.3	68.4	95	92	96.7	99.5	91.5
	MCC	0.86	0.76	0.91	0.83	0.96	0.98	0.90
-nearest neighbor	ACC	66.9	60.3	96.8	66.3	76.8	94.8	78.8
	MCC	0.74	0.51	0.68	0.70	0.85	0.97	0.76
Naïve Bayes	ACC	64.8	86	84.4	60.7	98.8	97.2	80.7
	MCC	0.61	0.64	0.81	0.65	0.87	0.98	0.77
SVM with AAC	ACC	72.4	73.5	87.7	89.3	99.8	70.8	84.1
	MCC	0.83	0.81	0.90	0.82	0.74	0.82	0.82
SVM with AAC + DC	ACC	82	71.3	95	91.8	96.9	94.3	91
	MCC	0.86	0.78	0.91	0.84	0.94	0.95	0.89