[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126404-en":3,"doc-seo-126404-105":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},126404,962085571259,"Theodora","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Integrating Machine Learning into Free Energy Perturbation Workflows","Free energy perturbation (FEP) methods provide high-accuracy predictions of protein–ligand binding affinities in structure-based drug design, yet their uptake is constrained by heavy computation and elaborate setup. This review examines how integrating machine learning—especially active learning and deep learning—can improve efficiency, accessibility, accuracy, and precision of FEP workflows. It covers ML-driven sampling, protocol optimization, and force-field development; active learning reduces needed FEP calculations, while cofolding models automate complex structure generation. Neural network potentials trained on quantum data further improve force-field accuracy, supporting a hybrid human–ML strategy for faster, more democratized drug discovery.","University of Groningen  \nIntegrating Machine Learning into Free Energy Perturbation Workflows  \nvan Pinxteren, Donald J. M. ; Jespers, Willem  \nPublished in:  \nJournal of chemical information and modeling  \nDOI:  \n10.1021/acs.jcim.5c01449  \nIMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.  \nDocument Version  \nPublisher's PDF, also known as Version of record  \nPublication date: 2025  \nLink to publication in University of Groningen/UMCG research database  \nCitation for published version (APA):  \nvan Pinxteren, D. J. M. , & Jespers, W. (2025) . Integrating Machine Learning into Free Energy Perturbation Workflows. Journal of chemical information and modeling, 65(19), 9856-9864.  \n[https://doi.org/10.1021/acs.jcim.5c01449](https://doi.org/10.1021/acs.jcim.5c01449)  \nCopyright  \nOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons) .  \nThe publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license. More information can be found on the University of Groningen website: [https://www.rug.nl/library/open-access/self-archiving-pure/taverne](https://www.rug.nl/library/open-access/self-archiving-pure/taverne)amendment.  \nTake-down policy  \nIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.  \nDownloaded from the University of Groningen/U MCG research database (Pure): [http://www.rug. nl/research/portal. For technical reasons the](http://www.rug. nl/research/portal. For technical reasons the)[ ](http://www.rug. nl/research/portal. For technical reasons the)[number of authors shown on this cover page is limited to 10 maximum.](number of authors shown on this cover page is limited to 10 maximum.)  \nDownload date: 30-12-2025  \nThis article is licensed under CC-BY 4.0   \n[pubs.acs.org/jcim](pubs.acs.org/jcim)  Perspective   \nIntegrating Machine Learning into Free Energy Perturbation Workflows  \nDonald J. M. van Pinxteren and Willem Jespers*  \n Cite This: J. Chem. Inf. Model. 2025, 65, 9856−9864  \nRead Online  \nDownloaded via UNIV GRONINGEN on October 30, 2025 at 10:57:45 (UTC) . See [https://pubs.acs.org/sharingguidelines](https://pubs.acs.org/sharingguidelines) for options on how to legitimately share published articles.  \nACCESS  \n Metrics & More  \n Article Recommendations  \nABSTRACT: Free energy perturbation (FEP) methods are among the most accurate tools in structure-based drug design for predicting protein−ligand binding affinities. However, their adoption remains limited due to high computational demands and complex setup procedures. This review explores how integrating machine learning (ML), especially active learning (AL) and deep learning (DL), can enhance the efficiency, accessibility, accuracy, and precision of FEP workflows. It examines three key areas where ML has been successfully applied: sampling strategies, protocol optimization, and force field development. AL algorithms can significantly reduce the number of FEP calculations needed during virtual screening by guiding the molecule selection. DL-based protein−ligand cofolding methods such as AlphaFold, NeuralPLexer, and DragonFold enable the automated generation of accurate complex structures for FEP, bypassing traditional docking and preparation steps. Additionally, MLderived neural network potentials (NNPs), trained on quantum mechanical data, offer improved force field accuracy, although at the cost of higher computational expenses. This review emphasizes a hybrid approach combining human expertise with ML tools as the most promising strategy for accelerating and democratizing FEP-ba","cbCaim2Uwa2LYmSX","https://ap.wps.com/l/cbCaim2Uwa2LYmSX","pdf",3228564,9,1,10,"English","en",105,"# Abstract\n# Introduction\n## Computational drug discovery and CADD\n## FEP methods: ABFE and RBFE\n# ML integration into FEP workflows\n## Sampling strategies via active learning\n## Protocol optimization\n## Force field development\n## Deep learning for protein–ligand complex generation\n# Hybrid human–ML perspective and future directions","[{\"question\":\"Why are FEP methods not widely adopted in drug discovery?\",\"answer\":\"FEP is limited by high computational costs and complex setup procedures despite its high accuracy for predicting binding affinities.\"},{\"question\":\"How does active learning improve FEP workflows?\",\"answer\":\"Active learning guides molecule selection in virtual screening, reducing the number of FEP calculations required to achieve effective results.\"},{\"question\":\"What roles do deep learning models and neural network potentials play in FEP?\",\"answer\":\"Deep learning cofolding models can generate accurate protein–ligand complex structures automatically, reducing reliance on traditional docking and preparation steps. ML-derived neural network potentials trained on quantum data can improve force-field accuracy, though they may increase computational expense.\"}]","Integrating Machine Learning into Free Energy Perturbation Workflows | PDF",1785904886,25,{"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},"integrating-machine-learning-into-free-energy-perturbation-workflows","",{"@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/integrating-machine-learning-into-free-energy-perturbation-workflows/126404/",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},"Why are FEP methods not widely adopted in drug discovery?","Question",{"text":77,"@type":78},"FEP is limited by high computational costs and complex setup procedures despite its high accuracy for predicting binding affinities.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How does active learning improve FEP workflows?",{"text":82,"@type":78},"Active learning guides molecule selection in virtual screening, reducing the number of FEP calculations required to achieve effective results.",{"name":84,"@type":75,"acceptedAnswer":85},"What roles do deep learning models and neural network potentials play in FEP?",{"text":86,"@type":78},"Deep learning cofolding models can generate accurate protein–ligand complex structures automatically, reducing reliance on traditional docking and preparation steps. ML-derived neural network potentials trained on quantum data can improve force-field accuracy, though they may increase computational expense.","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},[95,99,103,107,112,117,122,125,129,132,135],{"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":118,"doc_module":4,"doc_module_name":47,"category_name":119,"show_sort_weight":120,"slug":121},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":123,"slug":124},30,"research-report",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},"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":22,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":22,"slug":134},"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":108,"slug":138},19,"General","general"]