Detecting conserved protein complexes using a dividing-and-matching algorithm and unequally lenient criteria for network comparison

Table 2 Comparison of different methods in terms of how well matching with known proteins

Methods	PC	MPC	MKC	Recall	Precision	F-measure	CR	PM
Yeast-fly
UEDAMAlignCFinder (k = 4)	129	59	66	0.1471	0.4574	0.2226	0.1891	2
UEDAMAlignCMC	128	58	73	0.1476	0.4531	0.2226	0.2068	0
UEDAMAlignCoach	725	207	129	0.4259	0.2855	0.3419	0.3057	4
UEDAMAlignknowncomplex	148	145	172	0.3806	0.9797	0.5482	0.3432	45
UEDAMAlignMCL	862	159	137	0.3698	0.1845	0.2461	0.2401	9
AlignMCL	915	151	162	0.3804	0.165	0.2302	0.2479	9
Match-and-Split	27	12	20	0.03	0.4444	0.0562	0.0641	2
Mawish	41	16	26	0.0402	0.3902	0.0729	0.0318	1
NetworkBlast	179	9	10	0.0221	0.0503	0.0307	0.0391	0
Produles	46	29	26	0.0706	0.6304	0.1269	0.0573	3
Human-fly
UEDAMAlignCFinder (k = 4)	238	80	187	0.0531	0.3361	0.0917	0.106	3
UEDAMAlignCMC	404	67	182	0.0447	0.1658	0.0705	0.1585	2
UEDAMAlignCoach	1,538	428	493	0.2765	0.2783	0.2774	0.2983	10
UEDAMAlignknowncomplex	515	508	821	0.3908	0.9864	0.5598	0.4242	158
UEDAMAlignMCL	1,453	171	322	0.117	0.1177	0.1173	0.2008	9
AlignMCL	1,094	144	305	0.0992	0.1316	0.1131	0.1697	7
Match-and-Split	53	23	73	0.0147	0.434	0.0285	0.0558	3
Mawish	61	28	70	0.0178	0.459	0.0343	0.0333	1
NetworkBlast	164	45	107	0.029	0.2744	0.0525	0.0897	0
Produles	99	35	77	0.0223	0.3535	0.0419	0.0461	5

ISSN: 1748-7188