PPI-GA: A Novel Clustering Algorithm to Identify Protein Complexes within Protein-Protein Interaction Networks Using Genetic Algorithm

Table 10

Comparison of PPI-GA algorithm results with existing algorithm results.

Method	#Clusters	CYC2008			MIPS
		Recall	Precision	F-measure	Recall	Precision	F-measure
PPI-GA	71	0.89	0.63	0.74	0.36	0.51	0.42
DMCL-EHO	85	0.74	0.61	0.66	0.42	0.41	0.41
K-means clustering	120	0.55	0.67	0.60	0.28	0.40	0.32
MCODE	54	0.66	0.59	0.63	0.27	0.48	0.35
MCL	75	0.65	0.45	0.54	0.27	0.34	0.30
ClusterOne	114	0.55	0.43	0.49	0.20	0.34	0.25
Ramadan GA spectral population^1
(a) DC	162	0.74	0.6	0.66	0.32	0.45	0.37
(b) MC	180	0.71	0.47	0.6	0.38	0.40	0.39
(c) NC	193	0.67	0.5	0.57	0.39	0.37	0.37
(d) RC	161	0.73	0.38	0.5	0.39	0.33	0.36
Ramadan GA random population^1
(a) DC	164	0.72	0.54	0.62	0.3	0.41	0.35
(b) MC	162	0.71	0.45	0.56	0.4	0.35	0.38
(c) NC	138	0.66	0.57	0.61	0.36	0.44	0.41
(d) RC	154	0.61	0.55	0.58	0.34	0.43	0.38

¹DC : Density cut; MC : Maximum cut; NC : Normalized cut; RC : Ratio cut.