Predicting target profiles with confidence as a service using docking scores

Table 3 The table represents the model efficiency of predictions on the complete modelling set (from which training set was taken) and the external validation set

PDB entry	Eff on modelling set (%)	Eff on ext. val. set (%)	Inhibitors in top 10 (%) predicted ligands	Inhibitors in top 10 (%) docked molecules	Inhibitors in top 20 (%) predicted ligands	Inhibitors in top 20 (%) docked molecules
1RT2	93	97	32	31	68	68
1E66	93	94	60	52	67	70
1QCF	86	93	65	65	73	79
3ERD	93	92	65	58	78	69
3LN1	98	98	50	82	68	86
1BNU	87	87	47	55	75	78
1B8O	94	94	0	43	11	73
Average	92	94	46	55	63	74

The last four columns represents the predicted and the docking enrichment factor for inhibitors, i.e., the percent inhibitors found in the top 10% and 20% of the database search

ISSN: 1758-2946