Computational prediction of regioselectivity in the metabolism of xenobiotics

• Applicant: Professor Dr. Johannes Kirchmair
• Subject Area: Theoretical Chemistry: Electronic Structure, Dynamics, Simulation ; Bioinformatics and Theoretical Biology
• Term: from 2016 to 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 326167477

• Final Report Year: 2022

Final Report Abstract

The ability to predict the metabolic fate of small molecules is of essence to the discovery and design of safe and efficacious drugs, agrochemicals and cosmetics. Of particular value to compound optimization is site-of-metabolism (SoM) prediction. SoMs denote the atom positions in a molecule where biotransformations are initiated. Knowing these atom positions can give valuable pointers towards (i) how the metabolic properties (in particular, metabolic stability) can be improved and (ii) what the likely metabolites are. In this project, we have explored simple to complex atom representations and machine learning methods to develop accurate, informative and widely applicable models for SoM prediction. The models are built on a large set of compounds annotated with expert-curated SoMs. The newly developed models reach competitive performance while covering the full spectrum of phase 1 and phase 2 metabolism. They also cover natural products even though they are trained first and foremost on synthetic compounds. All predictions are accompanied by informative measures of their reliability. The models now form the basis also of metabolite structure predictors, which utilise the predicted probabilities of atoms being a SoM for the ranking of biotransformations and metabolites. The best models resulting from this work are accessible via a public web service.

Publications

• FAME 2: Simple and Effective Machine Learning Model of Cytochrome P450 Regioselectivity. J. Chem. Inf. Model. 2017, 57 (8), 1832–1846
  Šícho, M.; de Bruyn Kops, C.; Stork, C.; Svozil, D.; Kirchmair, J.
  (See online at https://doi.org/10.1021/acs.jcim.7b00250)
• ALADDIN: Docking Approach Augmented by Machine Learning for Protein Structure Selection Yields Superior Virtual Screening Performance. Mol. Inform. 2019
  Fan, N.; Bauer, C. A.; Stork, C.; Bruyn Kops, C.; Kirchmair, J.
  (See online at https://doi.org/10.1002/minf.201900103)
• FAME 3: Predicting the Sites of Metabolism in Synthetic Compounds and Natural Products for Phase 1 and Phase 2 Metabolic Enzymes. J. Chem. Inf. Model. 2019, 59 (8), 3400–3412
  Šícho, M.; Stork, C.; Mazzolari, A.; de Bruyn Kops, C.; Pedretti, A.; Testa, B.; Vistoli, G.; Svozil, D.; Kirchmair, J.
  (See online at https://doi.org/10.1021/acs.jcim.9b00376)
• GLORY: Generator of the Structures of Likely Cytochrome P450 Metabolites Based on Predicted Sites of Metabolism. Front. Chem. 2019, 7
  de Bruyn Kops, C.; Stork, C.; Šícho, M.; Kochev, N.; Svozil, D.; Jeliazkova, N.; Kirchmair, J.
  (See online at https://doi.org/10.3389/fchem.2019.00402)
• Hit Dexter 2.0: Machine-Learning Models for the Prediction of Frequent Hitters. J. Chem. Inf. Model. 2019, 59 (3), 1030–1043
  Stork, C.; Chen, Y.; Šícho, M.; Kirchmair, J.
  (See online at https://doi.org/10.1021/acs.jcim.8b00677)
• NERDD: A Web Portal Providing Access to in Silico Tools for Drug Discovery. Bioinformatics 2019
  Stork, C.; Embruch, G.; Šícho, M.; de Bruyn Kops, C.; Chen, Y.; Svozil, D.; Kirchmair, J.
  (See online at https://doi.org/10.1093/bioinformatics/btz695)
• Skin Doctor CP: Conformal Prediction of the Skin Sensitization Potential of Small Organic Molecules. Chem. Res. Toxicol. 2020
  Wilm, A.; Norinder, U.; Agea, M. I.; de Bruyn Kops, C.; Stork, C.; Kühnl, J.; Kirchmair, J.
  (See online at https://doi.org/10.1021/acs.chemrestox.0c00253)
• BonMOLière: Small-Sized Libraries of Readily Purchasable Compounds, Optimized to Produce Genuine Hits in Biological Screens across the Protein Space. Int. J. Mol. Sci. 2021, 22 (15), 7773
  Mathai, N.; Stork, C.; Kirchmair, J.
  (See online at https://doi.org/10.3390/ijms22157773)
• Computational Prediction of Frequent Hitters in Target-Based and Cell-Based Assays. Artif. Intell. Life Sci. 2021, 1, 100007
  Stork, C.; Mathai, N.; Kirchmair, J.
  (See online at https://doi.org/10.1016/j.ailsci.2021.100007)
• CYPlebrity: Machine Learning Models for the Prediction of Inhibitors of Cytochrome P450 Enzymes. Bioorg. Med. Chem. 2021, 46, 116388
  Plonka, W.; Stork, C.; Šícho, M.; Kirchmair, J.
  (See online at https://doi.org/10.1016/j.bmc.2021.116388)
• CYPstrate: A Set of Machine Learning Models for the Accurate Classification of Cytochrome P450 Enzyme Substrates and Non-Substrates. Molecules 2021, 26 (15), 4678
  Holmer, M.; de Bruyn Kops, C.; Stork, C.; Kirchmair, J.
  (See online at https://doi.org/10.3390/molecules26154678)
• Predicting the Skin Sensitization Potential of Small Molecules with Machine Learning Models Trained on Biologically Meaningful Descriptors. Pharmaceuticals 2021, 14 (8), 790
  Wilm, A.; Garcia de Lomana, M.; Stork, C.; Mathai, N.; Hirte, S.; Norinder, U.; Kühnl, J.; Kirchmair, J.
  (See online at https://doi.org/10.3390/ph14080790)