[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122925-en":3,"doc-seo-122925-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122925,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","Machine learning coarse-grained potentials of protein thermodynamics","General understanding of protein dynamics remains unsolved, limiting interpretation of structure–function relationships essential to biological processes. This work constructs coarse-grained molecular potentials using artificial neural networks and grounding them in statistical mechanics. Training uses an unbiased all-atom molecular dynamics dataset of ~9 ms across twelve proteins with multiple secondary-structure arrangements. The resulting coarse-grained models accelerate dynamics by over three orders of magnitude while preserving system thermodynamics, revealing ensemble states with comparable energies and integrating multiple proteins with experimental relevance to mutated forms.","Article [https://doi.org/10.1038/s41467-023-41343-1](https://doi.org/10.1038/s41467-023-41343-1)  \nMachine learning coarse-grained potentials of protein thermodynamics  \nReceived: 2 June 2023  \n\n| Accepted: 29 August 2023 |\n| --- |\n|  |\n| Check for updates |\n\nMaciej Majewski 1,2,14, Adrià Pérez 1,2,14, Philipp Thölke 1, Stefan Doerr2, Nicholas E. Charron3,4,5, Toni Giorgino 6, Brooke E. Husic7,8,9,10,  \nCecilia Clementi 3,4,5,11 , Frank Noé 5,7,11,12  & Gianni De Fabritiis 1,2,13   \nA generalized understanding of protein dynamics is an unsolved scientific problem, the solution of which is critical to the interpretation of the structurefunction relationships that govern essential biological processes. Here, we approach this problem by constructing coarse-grained molecular potentials based on artificial neural networks and grounded in statistical mechanics. For training, we build a unique dataset of unbiased all-atom molecular dynamics simulations of approximately 9 ms for twelve different proteins with multiple secondary structure arrangements. The coarse-grained models are capable of accelerating the dynamics by more than three orders of magnitude while preserving the thermodynamics of the systems. Coarse-grained simulations identify relevant structural states in the ensemble with comparable energeticsto the all-atom systems. Furthermore, we show that a single coarse-grained potential can integrate all twelve proteins and can capture experimental structural features of mutated proteins. These results indicate that machine learning coarse-grained potentials could provide a feasible approach to simulate and understand protein dynamics.  \nProteins are complex dynamical systems that exist in an equilibrium of distinct conformational states, and their multi-state behavior is critical for their biological functions1–5. A complete description ofthe dynamics of a protein requires the determination of (1) its stable and metastable conformational states,(2) the relative probabilities of these states, and (3) the rates of interconversion among them. Here, we focus on addressing the first two problems by demonstrating how to learn coarse-grained potentials that preserve protein thermodynamics.  \nDue to the structural heterogeneity of proteins and the ranges of time and length scales over which their dynamics occur, there is no  \nsingle technique that is able to successfully model protein behavior across the whole spatiotemporal scale. Computationally, the main method to study protein dynamics has traditionally been molecular dynamics (MD). The first MD simulation ever made was carried out in 1977 on the BPTI protein in vacuum, and only accounted for 9.2 picoseconds of simulation time6. As remarked by Karplus & McCammon7, these simulations were pivotal towards the realization that proteins are dynamic systems and that those dynamics play a fundamental role in their biological function2. When compared with experimental methods such as X-ray crystallography, MD simulations  \n1Computational Science Laboratory, Universitat Pompeu Fabra, Barcelona Biomedical Research Park (PRBB), Carrer Dr. Aiguader 88, 08003 Barcelona, Spain. 2Acellera Labs, Doctor Trueta 183, 08005 Barcelona, Spain. 3Department of Physics, Rice University, Houston, TX 77005, USA. 4Center for Theoretical Biological Physics, Rice University, Houston, TX 77005, USA. 5Department of Physics, FU Berlin, Arnimallee 12, 14195 Berlin, Germany. 6Biophysics Institute, National Research Council(CNR-IBF), 20133 Milan, Italy. 7Department of Mathematics and Computer Science, FU Berlin, Arnimallee 12,14195 Berlin, Germany. 8Lewis Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ 08540, USA. 9Princeton Center for Theoretical Science, Princeton University, Princeton, NJ 08540, USA. 10Center for the Physics of Biological Function, Princeton University, Princeton, NJ 08540, USA. 11Department of Chemistry, Rice University, Houston, TX77005, USA. 12Microsoft Research AI4Sc","cbCaicLdoN2FL0aL","https://ap.wps.com/l/cbCaicLdoN2FL0aL","pdf",2674144,1,13,"English","en",105,"# Introduction\n## Protein dynamics and thermodynamics\n## Limitations of existing modeling approaches\n## Sampling strategies for slow processes","[{\"question\":\"What problem does this study address in protein dynamics?\",\"answer\":\"It targets the challenge of building models that preserve protein thermodynamics while learning coarse-grained representations of conformational dynamics.\"},{\"question\":\"How are the coarse-grained potentials constructed and trained?\",\"answer\":\"They are constructed using artificial neural networks and trained on an unbiased all-atom molecular dynamics dataset of about 9 ms for twelve proteins with different secondary-structure arrangements.\"},{\"question\":\"What performance and validation results are reported?\",\"answer\":\"The coarse-grained models accelerate dynamics by more than three orders of magnitude while preserving thermodynamics, identifying relevant ensemble structural states and integrating all twelve proteins with the ability to capture features of mutated proteins.\"}]","Machine learning coarse-grained potentials of protein thermodynamics | PDF",1785813712,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-coarse-grained-potentials-of-protein-thermodynamics","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-coarse-grained-potentials-of-protein-thermodynamics/122925/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does this study address in protein dynamics?","Question",{"text":75,"@type":76},"It targets the challenge of building models that preserve protein thermodynamics while learning coarse-grained representations of conformational dynamics.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How are the coarse-grained potentials constructed and trained?",{"text":80,"@type":76},"They are constructed using artificial neural networks and trained on an unbiased all-atom molecular dynamics dataset of about 9 ms for twelve proteins with different secondary-structure arrangements.",{"name":82,"@type":73,"acceptedAnswer":83},"What performance and validation results are reported?",{"text":84,"@type":76},"The coarse-grained models accelerate dynamics by more than three orders of magnitude while preserving thermodynamics, identifying relevant ensemble structural states and integrating all twelve proteins with the ability to capture features of mutated proteins.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]