Information Content-Based Gene Ontology Semantic Similarity Approaches: Toward a Unified Framework Theory

Table 2

Comparison of performance of different approaches for GO BP ontology. This comparison is done using Pearson’s correlation with enzyme commission (EC), Pfam and sequence similarity, and resolution. Results are obtained from the CESSM online tool. The best scores among each group are in bold, and Nmax, Nunif, and Nunivers are suffixes indicating different IC normalization strategies, namely, the highest IC value, uniform, and GO-universal strategies, respectively.

Family	Approach	Similarity measure correlation			Resolution
		EC	PFAM	Seq Sim
Annotation	Resnik-Nmax	0.41166	0.29151	0.54563	0.55874
	Resnik-Nunif	0.41166	0.29151	0.54563	0.49265
	Nunivers	0.48967	0.41280	0.62349	0.48490
	Lin	0.48032	0.38900	0.57956	0.43343
	Li et al.	0.49531	0.42010	0.62173	0.49017
	Relevance	0.48188	0.38682	0.57550	0.43823
	GraSM-Lin	0.48673	0.45470	0.61739	0.51701
	GraSM-Nmax	0.44826	0.35941	0.63497	0.54996
	GraSM-Nunif	0.44826	0.35941	0.63497	0.48491
	GraSM-Nunivers	0.49301	0.44158	0.65671	0.92975
	XGraSM-Lin	0.39811	0.49859	0.68669	0.92067
	XGraSM-Nmax	0.45493	0.37152	0.69892	0.53910
	XGraSM-Nunif	0.45493	0.37152	0.69892	0.47533
	XGraSM-Nunivers	0.49782	0.45220	0.70732	0.91425
Topology	Wang et al.	0.45451	0.47867	0.65214	0.91475
	Zhang et al.	0.47888	0.45527	0.61862	0.44350
	GO-universal	0.45958	0.48175	0.68953	0.43772