[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120444-en":3,"doc-seo-120444-105":30,"detail-sidebar-cat-0-en-105":92},{"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":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},120444,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",6,"Technology","How to make machine learning scoring functions competitive with FEP","Machine learning supports fast, accurate binding affinity prediction, yet many models generalise poorly beyond training data and receive limited, non-robust evaluation across diverse benchmarks, constraining adoption in drug discovery. AEV-PLIG (Atomic Environment Vector-Protein Ligand Interaction Graph) addresses these gaps by using a graph neural network that encodes protein–ligand interactions through atomic environment vectors. The approach is validated on improved and out-of-distribution benchmarks, including an OOD Test set and systems tied to FEP calculations, demonstrating competitive performance with data augmentation strategies that further boost accuracy.","How to make machine learning scoring functions  \ncompetitive with FEP  \nMatthew T. Warren,†, § sak Valsson,‡, § Charlotte M. Deane,‡ Aniket  \nMagarkar,∗ ,¶ Garrett M. Morris, ∗ ,‡ and Philip C. Biggin ∗ ,†  \n†Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU,  \nUK  \n‡Department of Statistics, University of Oxford, 24-29 St Giles’, Oxford, OX1 3LB, UK ¶Boehringer Ingelheim Pharma GmbH & Co. KG, Birkendorfer Str. 65, 88397 Biberach  \nan de Riß, Germany  \n§These authors contributed equally to this work  \nE-mail: [aniket.magarkar@boehringer-ingelheim.com](aniket.magarkar@boehringer-ingelheim.com) ; [garrett.morris@dtc.ox.ac.uk](garrett.morris@dtc.ox.ac.uk) ;  \n[philip.biggin@bioch.ox.ac.uk](philip.biggin@bioch.ox.ac.uk)  \n1  \n[https://doi.org/10.26434/chemrxiv-2024-bth5z](https://doi.org/10.26434/chemrxiv-2024-bth5z ORCID:)[ ORCID:](https://doi.org/10.26434/chemrxiv-2024-bth5z ORCID:) [https://orcid.org/0000-0001-5100-8836 Content](https://orcid.org/0000-0001-5100-8836 Content) not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0  \nAbstract  \nMachine learning offers a promising approach for fast and accurate binding affinity predictions. However, current models often fail to generalise beyond their training data and are not robustly evaluated on a diverse range of benchmarks, limiting their application in drug discovery projects. In this work, we address these issues by introducing a novel graph neural network model called AEV-PLIG (Atomic Environment Vector-Protein Ligand Interaction Graph), which encodes protein-ligand interactions via atomic environment vectors to improve generalisation. We evaluate our model on improved benchmarks, including our new out-of-distribution test set we call OOD Test, and two alternative benchmark systems used for free energy perturbation (FEP) calculations, and highlight competitive performance of AEV-PLIG across the board. Moreover, we demonstrate how augmented data can be leveraged to enhance prediction accuracy, and how enriching the training data with three complexes from a congeneric series of ligands binding to a target of interest improves performance further. Altogether, we show that these strategies improve the applicability of machine learning scoring functions and enable state-of-the-art performance nearing the accuracy of physics-based simulation methods—but at a fraction of their computational cost. This practical approach extends the predictive capabilities of machine learning for molecular discovery, paving the way for its broader use in computer-aided drug design.  \n2  \n[https://doi.org/10.26434/chemrxiv-2024-bth5z](https://doi.org/10.26434/chemrxiv-2024-bth5z ORCID:)[ ORCID:](https://doi.org/10.26434/chemrxiv-2024-bth5z ORCID:) [https://orcid.org/0000-0001-5100-8836 Content](https://orcid.org/0000-0001-5100-8836 Content) not peer-reviewed by ChemRxiv. License: CC BY-NC 4.0  \nIntroduction  \nPredicting the change in free energy upon binding of protein and ligand represents a cornerstone of small molecule drug discovery. It is essential during hit identification, where one aims to identify binders that demonstrate high affinity for a target, as well as in hit-to-lead and lead optimisation, where binding affinity must be optimised alongside a number of other properties pertinent to safety and biological efficacy. Given the vastness of chemical space  \n– and hence possible drug candidates 1 – computer-aided drug design (CADD) plays an important role in accelerating the discovery process by enabling large numbers of compounds to be screened in silico, avoiding the significant cost and resources required for experimental measurements. 2  \nTo predict binding affinity, computational methods typically employ knowledge-or physicsbased approaches that depend on statistical potentials or those based on molecular mechanics force fields, respectively. 3 These approaches offer a trade-off between cost and accuracy: fast estimations of binding affinity, such as sco","cbCaiuCtBRkvQfs1","https://ap.wps.com/l/cbCaiuCtBRkvQfs1","pdf",2869418,1,38,"English","en",105,"# Abstract\n# Introduction\n## Binding free energy in small-molecule drug discovery\n## Computational trade-offs: scoring functions vs physics-based simulations\n## Limitations of FEP\n## Role of machine learning as an alternative","[{\"question\":\"Why are existing machine learning scoring functions limited in drug discovery?\",\"answer\":\"They often fail to generalise beyond their training data and are not evaluated robustly across diverse benchmarks, which reduces reliability for real project use.\"},{\"question\":\"What is AEV-PLIG and how does it improve prediction?\",\"answer\":\"AEV-PLIG is a graph neural network that encodes protein–ligand interactions using atomic environment vectors to strengthen generalisation.\"},{\"question\":\"How does the study improve results beyond the core model architecture?\",\"answer\":\"It uses augmented data and enrichment of training data with additional complexes from a congeneric ligand series to raise prediction accuracy further.\"}]","How to make machine learning scoring functions competitive with FEP | PDF",1785730153,96,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"how-to-make-machine-learning-scoring-functions-competitive-with-fep","",{"@graph":36,"@context":86},[37,54,69],{"@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/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/how-to-make-machine-learning-scoring-functions-competitive-with-fep/120444/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why are existing machine learning scoring functions limited in drug discovery?","Question",{"text":76,"@type":77},"They often fail to generalise beyond their training data and are not evaluated robustly across diverse benchmarks, which reduces reliability for real project use.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What is AEV-PLIG and how does it improve prediction?",{"text":81,"@type":77},"AEV-PLIG is a graph neural network that encodes protein–ligand interactions using atomic environment vectors to strengthen generalisation.",{"name":83,"@type":74,"acceptedAnswer":84},"How does the study improve results beyond the core model architecture?",{"text":85,"@type":77},"It uses augmented data and enrichment of training data with additional complexes from a congeneric ligand series to raise prediction accuracy further.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,114,119,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":112,"slug":113},50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":120,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":122,"slug":123},8,"Research & Report",30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]