A method that predicts the probability that a molecule or set of molecules will be active against a particular target by considering its pairwise similarities
to all known ligands for that target.
Source:
PAC, 2016, 88, 239. 'Glossary of terms used in computational drug design, part II (IUPAC Recommendations 2015)' on page 256 (https://doi.org/10.1515/pac-2012-1204)