Premium
Hit Dexter: A Machine‐Learning Model for the Prediction of Frequent Hitters
Author(s) -
Stork Conrad,
Wagner Johannes,
Friedrich NilsOle,
de Bruyn Kops Christina,
Šícho Martin,
Kirchmair Johannes
Publication year - 2018
Publication title -
chemmedchem
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.817
H-Index - 100
eISSN - 1860-7187
pISSN - 1860-7179
DOI - 10.1002/cmdc.201700673
Subject(s) - pubchem , computer science , artificial intelligence , in silico , machine learning , chembl , computational biology , bioinformatics , drug discovery , biology , biochemistry , gene
False-positive assay readouts caused by badly behaving compounds-frequent hitters, pan-assay interference compounds (PAINS), aggregators, and others-continue to pose a major challenge to experimental screening. There are only a few in silico methods that allow the prediction of such problematic compounds. We report the development of Hit Dexter, two extremely randomized trees classifiers for the prediction of compounds likely to trigger positive assay readouts either by true promiscuity or by assay interference. The models were trained on a well-prepared dataset extracted from the PubChem Bioassay database, consisting of approximately 311 000 compounds tested for activity on at least 50 proteins. Hit Dexter reached MCC and AUC values of up to 0.67 and 0.96 on an independent test set, respectively. The models are expected to be of high value, in particular to medicinal chemists and biochemists who can use Hit Dexter to identify compounds for which extra caution should be exercised with positive assay readouts. Hit Dexter is available as a free web service at http://hitdexter.zbh. uni-hamburg.de.