Using docking-derived protein-ligand interaction descriptors to increase performance of QSAR models for human CYP450 inhibition
Sergii Novotarskyi, Iurii Sushko, Robert Koerner and Igor Tetko, eADMET GmbH, Germany
Cytochromes P450 (CYP450) are a superfamily of enzymes, involved in metabolism of a large number of xenobiotic compounds. CYP450 are involved in metabolism of a large amount of drugs, currently present on the market. Individual CYP enzymes in families 1, 2 and 3 metabolize xenobiotics, including the majority of small molecule drugs currently in use. The distinctive feature of CYP450 enzymes is broad and overlapping substrate specificity. Therefore, prediction of CYP450 inhibition activity of small molecules poses an important task due to high risk of drug-drug interactions.
In this work the quality of novel docking-derived protein-ligand interaction descriptors is benchmarked in QSAR modeling of HTS data for human CYP450 inhibition.
The calculation of descriptors involves a flexible docking of the molecule to the rigid binding cite of the cytochrome (in this study the AutoDock Vina tool was used). The obtained top-ranked conformation is then processed to obtain the descriptors. The descriptors include binding affinity (as predicted by the docking tool) and summary atom counts and atom charges for every type of atom in the protein and the ligand, binned by distance between these atoms.
The training sets for the benchmarked models were obtained from PubChem BioAssay database. The identifiers of PubChem BioAssays with training data for this study are AID410 for CYP1A2, AID883 for CYP2C9, AID884 for CYP3A4, and AID891 for CYP2D6. The test sets are obtained from the AID1851 assay. The additional test sets were formed from literature data.
The study shows, that the use of docking-derived descriptors allows a statistically significant increase in QSAR model performance.
The datasets and the benchmarked models are available on the Online Chemical Modeling Environment (http://ochem.eu)
(presenting author: Sergii Novotarskyi)
