Polypharmacology Browser PPB3: A Web-based Deep Learning Tool for Target Prediction Using ChEMBL Data

Drug-like molecules often interact with multiple biological targets. Assessing this polypharmacology is essential for drug development. Here we trained deep neural networks to associate bioactive molecules up to 80 non-hydrogen atoms in size reported in ChEMBL 34, represented as binary substructure fingerprints, with lists of targets on which the molecules are > 50% active at ≤10 µM. We included 2,496,555 interactions between 1,187,089 molecules and 7,546 targets having at least five reported active molecules, including single proteins, protein complexes, protein families, cell lines, organisms, and further target types. This represents a much larger dataset than previously reported models, which were mostly limited to protein targets. Our DNNs achieve excellent performance in terms of recall and precision per molecule and per target among the different target types, as illustrated by overall statistics and by a case study in comparison with other online prediction tools. PPB3 predictions can be performed online at https://ppb3.gdb.tools/.