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

Table 4

Comparison of performance of different approaches for GO CC 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	0.34355	0.43796	0.55437	0.40935
	Resnik-Nunif	0.34355	0.43796	0.55437	0.33651
	Nunivers	0.32079	0.40931	0.53991	0.95526
	Lin	0.29912	0.38851	0.4998	0.96253
	Li et al.	0.32110	0.40980	0.54062	0.95511
	Relevance	0.30183	0.39132	0.50435	0.96131
	GraSM-Lin	0.30463	0.38749	0.51142	0.95896
	GraSM-Nmax	0.36341	0.45946	0.60546	0.40677
	GraSM-Nunif	0.36341	0.45946	0.60546	0.33439
	GraSM-Nunivers	0.32473	0.41170	0.55626	0.95247
	XGraSM-Lin	0.30812	0.39642	0.57390	0.95087
	XGraSM-Nmax	0.37079	0.47364	0.68564	0.39428
	XGraSM-Nunif	0.37079	0.47364	0.68564	0.32412
	XGraSM-Nunivers	0.32451	0.41225	0.59762	0.94673
Topology	Wang et al.	0.34404	0.39297	0.61451	0.94019
	Zhang et al.	0.00477	0.00246	0.00188	0.36749
	GO-universal	0.15787	0.19982	0.13119	1.00000