Journal of Cheminformatics

Table 2 Overall performance of LigTMap using MM or MMD as ligand similarity search, for all-target class prediction or class-specific (CS) prediction

From: LigTMap: ligand and structure-based target identification and activity prediction for small molecular compounds

	Average success rate			Average (Top 10)
	Top 1	Top 5	Top 10	Precision	Recall	F1 score
Validation set
MM	0.56	0.68	0.72	0.35	0.27	0.23
MMD	0.53	0.70	0.73	0.35	0.25	0.20
MM-CS	0.63	0.75	0.77	0.55	0.52	0.45
MMD-CS	0.61	0.74	0.76	0.54	0.38	0.35
Benchmark set
MM	0.40	0.59	0.63	0.29	0.27	0.26
MMD	0.44	0.64	0.65	0.33	0.27	0.24
MM-CS	0.60	0.74	0.84	0.53	0.52	0.65
MMD-CS	0.57	0.73	0.82	0.55	0.45	0.44

Back to article page

ISSN: 1758-2946

Contact us

Submission enquiries: journalsubmissions@springernature.com