{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Martin J"],"funding":["HHS | National Institutes of Health","Welch Foundation (The Welch Foundation)","Ministerio de Ciencia e Innovación (MCIN)","HHS | National Institutes of Health (NIH)","Cancer Prevention and Research Institute of Texas (CPRIT)","Ministerio de Ciencia e Innovación","National Science Foundation (NSF)","Welch Foundation","NIGMS NIH HHS","Cancer Prevention and Research Institute of Texas","National Science Foundation","HPC Europe Program"],"pagination":["e2311807121"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC11228481"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["121(27)"],"pubmed_abstract":["Machine learning has been proposed as an alternative to theoretical modeling when dealing with complex problems in biological physics. However, in this perspective, we argue that a more successful approach is a proper combination of these two methodologies. We discuss how ideas coming from physical modeling neuronal processing led to early formulations of computational neural networks, e.g., Hopfield networks. We then show how modern learning approaches like Potts models, Boltzmann machines, and the transformer architecture are related to each other, specifically, through a shared energy representation. We summarize recent efforts to establish these connections and provide examples on how each of these formulations integrating physical modeling and machine learning have been successful in "],"journal":["Proceedings of the National Academy of Sciences of the United States of America"],"pubmed_title":["Machine learning in biological physics: From biomolecular prediction to design."],"pmcid":["PMC11228481"],"funding_grant_id":["PID2022-139467OB-I00","N/A","R35 GM133631","PRTR-C17.I1","EHPC-DEV-2023D06-018","R35GM133631","PHY-2019745","EHPC-BEN-2023B07-015","C-1792","MCB-1943442","PHY-2210291"],"pubmed_authors":["Morcos F","Martin J","Coluzza I","Lequerica Mateos M","Onuchic JN"],"additional_accession":[]},"is_claimable":false,"name":"Machine learning in biological physics: From biomolecular prediction to design.","description":"Machine learning has been proposed as an alternative to theoretical modeling when dealing with complex problems in biological physics. However, in this perspective, we argue that a more successful approach is a proper combination of these two methodologies. We discuss how ideas coming from physical modeling neuronal processing led to early formulations of computational neural networks, e.g., Hopfield networks. We then show how modern learning approaches like Potts models, Boltzmann machines, and the transformer architecture are related to each other, specifically, through a shared energy representation. We summarize recent efforts to establish these connections and provide examples on how each of these formulations integrating physical modeling and machine learning have been successful in ","dates":{"release":"2024-01-01T00:00:00Z","publication":"2024 Jul","modification":"2026-06-01T07:37:24.632Z","creation":"2025-04-04T11:25:02.675Z"},"accession":"S-EPMC11228481","cross_references":{"pubmed":["38913893"],"doi":["10.1073/pnas.2311807121"]}}