MoCHI: neural networks to fit interpretable models and quantify energies, energetic couplings, epistasis, and allostery from deep mutational scanning data.

Autor: Faure AJ; Centre for Genomic Regulation (CRG), The Barcelona Institute of Science and Technology, Barcelona, Spain. andre.faure@crg.eu.; Current Address: ALLOX, PRBB Building, C/Dr. Aiguader, 88, 08003, Barcelona, Spain. andre.faure@crg.eu., Lehner B; Centre for Genomic Regulation (CRG), The Barcelona Institute of Science and Technology, Barcelona, Spain. bl11@sanger.ac.uk.; University Pompeu Fabra (UPF), Barcelona, Spain. bl11@sanger.ac.uk.; Institució Catalana de Recerca I Estudis Avançats (ICREA), Barcelona, Spain. bl11@sanger.ac.uk.; Wellcome Sanger Institute, Wellcome Genome Campus, Hinxton, UK. bl11@sanger.ac.uk.
Jazyk: angličtina
Zdroj: Genome biology [Genome Biol] 2024 Dec 02; Vol. 25 (1), pp. 303. Date of Electronic Publication: 2024 Dec 02.
DOI: 10.1186/s13059-024-03444-y
Abstrakt: We present MoCHI, a tool to fit interpretable models using deep mutational scanning data. MoCHI infers free energy changes, as well as interaction terms (energetic couplings) for specified biophysical models, including from multimodal phenotypic data. When a user-specified model is unavailable, global nonlinearities (epistasis) can be estimated from the data. MoCHI also leverages ensemble, background-averaged epistasis to learn sparse models that can incorporate higher-order epistatic terms. MoCHI is freely available as a Python package ( https://github.com/lehner-lab/MoCHI ) relying on the PyTorch machine learning framework and allows biophysical measurements at scale, including the construction of allosteric maps of proteins.
Competing Interests: Declarations: Ethics approval and consent to participate Ethical approval was not needed for the study. Competing interests: A.J.F. and B.L. are founders, employees, and shareholders of ALLOX.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz: