Verifying the fully “Laplacianised” posterior Naïve Bayesian approach and more

Table 1 Enzyme data set: comprising 4,658 ligands to classify according to the enzyme they inhibit

Activity class	Target ID	No. of active compounds
Vascular endothelial growth factor 2	10980	268
Carbonic anhydrase II	15	262
11-beta-hydroxysteroid dehydrogenase 1	11489	233
Carbonic anhydrase I	10193	228
Beta-secretase 1	12252	212
Dipeptidyl peptidase IV	11140	208
Epidermal growth factor erbB1	9	202
MAP kinase p38 alpha	10188	197
Carbonic anhydrase IX	12952	184
Cyclooxygenase-2	194	180
Acetylcholinesterase	93	158
Coagulation factor X	126	156
Histone deacetylase 1	12697	156
Monoamine oxidase B	104	148
Thrombin	11	147
Renin	11225	139
Epoxide hydratase	11727	134
Matrix metalloproteinase 13	11024	125
Cathepsin K	10495	116
Cathepsin S	11534	112
Matrix metalloproteinase-2	13001	112
Protein-tyrosine phosphatase 1B	13061	110
Serine–threonine-protein kinase AKT	12666	109
Butyrylcholinesterase	10532	104
Cytochrome P450 19A1	65	102
Receptor protein-tyrosine kinase erbB2	188	99
Tyrosine-protein kinase SRC	10434	98
Hepatocyte growth factor receptor	11451	94
Matrix metalloproteinase-1	13000	91
Glycogen synthase kinase-3 beta	10197	89
Carbonic anhydrase XII	12209	85

Columns 1, 2 and 3 denote the protein, the protein identifier (ID) in our dataset and the number of ligands reported for each protein, respectively.

ISSN: 1758-2946