Leveraging molecular structure and bioactivity with chemical language models for de novo drug design.

Autor: Moret M; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Pachon Angona I; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Cotos L; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Yan S; University of Zurich, University Children's Hospital, Children's Research Center, Pediatric Molecular Neuro-Oncology Research, Lengghalde 5, 8008, Zurich, Switzerland., Atz K; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Brunner C; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland., Baumgartner M; University of Zurich, University Children's Hospital, Children's Research Center, Pediatric Molecular Neuro-Oncology Research, Lengghalde 5, 8008, Zurich, Switzerland., Grisoni F; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland. f.grisoni@tue.nl.; Eindhoven University of Technology, Institute for Complex Molecular Systems and Eindhoven Artificial Intelligence Systems Institute, Department of Biomedical Engineering, Groene Loper 7, 5612AZ, Eindhoven, The Netherlands. f.grisoni@tue.nl.; Center for 393 Living Technologies, Alliance TU/e, WUR, UU, UMC 394 Utrecht, Utrecht, 3584 CB, The Netherlands. f.grisoni@tue.nl., Schneider G; ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland. gisbert@ethz.ch.; ETH Singapore SEC Ltd, 1 CREATE Way, #06-01 CREATE Tower, Singapore, 138602, Singapore. gisbert@ethz.ch.
Jazyk: angličtina
Zdroj: Nature communications [Nat Commun] 2023 Jan 07; Vol. 14 (1), pp. 114. Date of Electronic Publication: 2023 Jan 07.
DOI: 10.1038/s41467-022-35692-6
Abstrakt: Generative chemical language models (CLMs) can be used for de novo molecular structure generation by learning from a textual representation of molecules. Here, we show that hybrid CLMs can additionally leverage the bioactivity information available for the training compounds. To computationally design ligands of phosphoinositide 3-kinase gamma (PI3Kγ), a collection of virtual molecules was created with a generative CLM. This virtual compound library was refined using a CLM-based classifier for bioactivity prediction. This second hybrid CLM was pretrained with patented molecular structures and fine-tuned with known PI3Kγ ligands. Several of the computer-generated molecular designs were commercially available, enabling fast prescreening and preliminary experimental validation. A new PI3Kγ ligand with sub-micromolar activity was identified, highlighting the method's scaffold-hopping potential. Chemical synthesis and biochemical testing of two of the top-ranked de novo designed molecules and their derivatives corroborated the model's ability to generate PI3Kγ ligands with medium to low nanomolar activity for hit-to-lead expansion. The most potent compounds led to pronounced inhibition of PI3K-dependent Akt phosphorylation in a medulloblastoma cell model, demonstrating efficacy of PI3Kγ ligands in PI3K/Akt pathway repression in human tumor cells. The results positively advocate hybrid CLMs for virtual compound screening and activity-focused molecular design.
(© 2023. The Author(s).)
Databáze: MEDLINE
Externí odkaz: