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

Table 14

Comparison of various algorithms in terms of significance of the predicted complexes for Collins data.

Method	#Predicted complexes	#Significant complexes	Proportion (%)
PPI-GA	71	67	94.36
DMCL-EHO	85	77	90.58
K-means clustering	120	88	73.33
MCODE	54	48	88.88
MCL	75	37	49.33
ClusterOne	114	85	74.56
Ramadan GA spectral population^1
(a) DC	162	150	92.59
(b) MC	180	148	82.22
(c) NC	193	135	69.94
(d) RC	161	110	68.32
Ramadan GA random population^1
(a) DC	164	143	87.19
(b) MC	162	113	69.75
(c) NC	138	124	89.85
(d) RC	154	120	77.92

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