[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125855-en":3,"doc-seo-125855-105":31,"detail-sidebar-cat-0-en-105":93},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},125855,1099523882367,"Hazel","https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc",8,"Research & Report","MACE-OFF - Short Range Transferable Machine Learning Force Fields for Organic Molecules - Abstract Summary","MACE-OFF introduces a series of short-range transferable machine learning force fields for organic molecules, built from state-of-the-art ML methods and first-principles reference data computed with high-level quantum theory. The approach delivers accurate predictions for diverse gas and condensed-phase properties, including reliable dihedral torsion scans for unseen molecules and credible descriptions of molecular crystals and liquids with quantum nuclear effects. It further enables free-energy surfaces in explicit solvent and folding dynamics of peptides, supporting first-principles simulations at high accuracy with comparatively low cost.","arXiv :2312 . 15211v5 [physics .chem-ph] 22 Aug 2025  \nMACE-OFF: Short Range Transferable Machine Learning Force Fields for Organic Molecules  \nD􀀓avid P􀀓eter Kov􀀓acs†, 1 J. Harry Moore†* , 1, 2 Nicholas J. Browning,3 Ilyes Batatia, 1 Joshua T. Horton,4 Yixuan Pu,5 Venkat Kapil,6, 5, 7 William C. Witt,8 Ioan-Bogdan Magd􀀕au,4 Daniel J. Cole,4 and G􀀓abor Cs􀀓anyi 1, 2  \n1 Engineering Laboratory, University of Cambridge, Cambridge, CB2 1PZ, UK  \n2 ngstr¨om AI, 2325 3rd Street, San Francisco, CA 94107, USA  \n3 Swiss National Supercomputing Centre (CSCS), 6900, Lugano, Switzerland  \n4 School of Natural and Environmental Sciences,  \nNewcastle University, Newcastle upon Tyne NE1 7RU, UK  \n5 Department of Physics and Astronomy, University College, London WC1E 6BT, UK  \n6 Yusuf Hamied Department of Chemistry, University of Cambridge, Lens􀀌eld Road, Cambridge, CB2 1EW, UK  \n7 Thomas Young Centre and London Centre for Nanotechnology, London WC1E 6BT, UK  \n8 Department of Materials Science and Metallurgy, University of Cambridge,  \n27 Charles Babbage Road, CB3 0FS, Cambridge, United Kingdom (Dated: August 25, 2025)  \nAbstract  \nClassical empirical force 􀀌elds have dominated biomolecular simulation for over 50 years. Although widely used in drug discovery, crystal structure prediction, and biomolecular dynamics, they generally lack the accuracy and transferability required for 􀀌rst-principles predictive modeling. In this paper, we introduce MACE-OFF, a series of short range transferable force 􀀌elds for organic molecules created using state-of-the-art machine learning technology and 􀀌rst-principles reference data computed with a high level of quantum mechanical theory. MACE-OFF demonstrates the remarkable capabilities of short range models by accurately predicting a wide variety of gas and condensed phase properties of molecular systems. It produces accurate, easy-to-converge dihedral torsion scans of unseen molecules, as well as reliable descriptions of molecular crystals and liquids, including quantum nuclear e􀀋ects. We further demonstrate the capabilities of MACE-OFF by determining free energy surfaces in explicit solvent, as well as the folding dynamics of peptides and nanosecond simulation of a fully solvated protein. These developments enable 􀀌rst-principles simulations of molecular systems for the broader chemistry community at high accuracy and relatively low computational cost.  \nI. Introduction  \nMachine learning (ML) force 􀀌elds have recently undergone major improvements in accuracy, robustness, and computational speed [1–13] . They are now routinely used in materials chemistry contexts where density functional theory was previously the method of choice. In these applications, available empirical force 􀀌elds, such as the embedded-atom method [14], do not provide su􀀎cient accuracy and transferability to describe many scienti􀀌cally interesting and challenging phenomena. Successful applications of ML potentials include simulation of quenching of amorphous silicon [15], determination of the phase diagrams of inorganic perovskites [16] and alloys [17], and device-scale simulation of phase-change memory materials [18] .  \nIn contrast, simulating bio-organic systems entails adi􀀋erent set of trade-o􀀋s, with greater emphasis on simulating large systems over long timescales. This means  \n† These authors contributed equally.  \n􀀃 Corresponding [author: jhm72@cam.ac.uk](author: jhm72@cam.ac.uk)  \nthat empirical force 􀀌elds, which sacri􀀌ce accuracy for computational speed, continue to be used routinely to study molecular liquids, crystals, biological systems, and drug-like molecules [19–22] .  \nTwo alternatives to empirical force 􀀌elds are available. The 􀀌rst is semi-empirical quantum mechanics, such as the series of extended tight-binding models [23], which represents a low-cost solution for small molecules. The method is limited by its moderate accuracy compared to quantum chemistry methods, its restriction to modelling non-periodic systems (i","cbCaisztYXpSnqat","https://ap.wps.com/l/cbCaisztYXpSnqat","pdf",4345460,7,1,47,"English","en",105,"# Abstract\n## Introduction\n## Background on ML force fields\n## Trade-offs in bio-organic simulations\n## Alternatives to empirical force fields","[{\"question\":\"What problem does MACE-OFF address compared with classical empirical force fields?\",\"answer\":\"Classical empirical force fields are widely used but typically lack the accuracy and transferability needed for first-principles predictive modeling. MACE-OFF is designed to improve both accuracy and transferability for organic molecules.\"},{\"question\":\"How does MACE-OFF demonstrate its predictive capability?\",\"answer\":\"It accurately predicts a wide range of gas and condensed-phase properties. It also provides accurate dihedral torsion scans for molecules not seen during training and dependable descriptions of molecular crystals and liquids, including quantum nuclear effects.\"},{\"question\":\"What additional capabilities beyond property prediction does MACE-OFF enable?\",\"answer\":\"It supports first-principles style simulations by determining free-energy surfaces in explicit solvent and by modeling folding dynamics of peptides, including nanosecond simulation of fully solvated proteins.\"}]","MACE-OFF - Short Range Transferable Machine Learning Force Fields for Organic Molecules - Abstract Summary | PDF",1785901611,118,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"mace-off-short-range-transferable-machine-learning-force-fields-for-organic-molecules-abstract-summary","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/mace-off-short-range-transferable-machine-learning-force-fields-for-organic-molecules-abstract-summary/125855/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What problem does MACE-OFF address compared with classical empirical force fields?","Question",{"text":77,"@type":78},"Classical empirical force fields are widely used but typically lack the accuracy and transferability needed for first-principles predictive modeling. MACE-OFF is designed to improve both accuracy and transferability for organic molecules.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How does MACE-OFF demonstrate its predictive capability?",{"text":82,"@type":78},"It accurately predicts a wide range of gas and condensed-phase properties. It also provides accurate dihedral torsion scans for molecules not seen during training and dependable descriptions of molecular crystals and liquids, including quantum nuclear effects.",{"name":84,"@type":75,"acceptedAnswer":85},"What additional capabilities beyond property prediction does MACE-OFF enable?",{"text":86,"@type":78},"It supports first-principles style simulations by determining free-energy surfaces in explicit solvent and by modeling folding dynamics of peptides, including nanosecond simulation of fully solvated proteins.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,112,117,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":108,"doc_module":4,"doc_module_name":47,"category_name":109,"show_sort_weight":110,"slug":111},5,"Comic",60,"comic",{"id":113,"doc_module":4,"doc_module_name":47,"category_name":114,"show_sort_weight":115,"slug":116},6,"Technology",50,"technology",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":108,"slug":139},19,"General","general"]