Improving chemical similarity ensemble approach in target prediction

Table 1 Statistics of the training and test sets

Data set	Target	Molecule	Ligand-target pair	Active
Training set (5)	2,809	393,090	666,313	All
Training set (6)	2,297	294,877	407,296	All
Training set (7)	1,711	179,710	246,651	All
Kinase training set (5)	429	42,164	101,502	All
Test set	1190	26,498	80,066	37,138
Kinase test	259	2,225	3010	2,192

The size of 4 training data sets and 2 test sets. Numbers in brackets denote activity thresholds

ISSN: 1758-2946