{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T16:06:41Z","timestamp":1790179601223,"version":"4.1.0"},"reference-count":41,"publisher":"American Chemical Society (ACS)","issue":"5","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,23]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Predicting protein\u2013protein binding free energy (\u0394G) from structure remains a central challenge in computational biophysics. Here, we present GULP (Graph-based Unified Learning for Protein binding), a graph neural network (GNN) that jointly learns from a residue-level graph representation of the binding interface and global physicochemical descriptors. We systematically investigate how training data distribution affects model performance by comparing a full training set with a balanced subset enriched for extreme-affinity complexes. GULP is computationally efficient and provides interpretable insights into residue-level and physicochemical contributions to binding. On external validation, GULP achieves a mean absolute error (MAE) of 2.31 kcal\/mol and shows moderate agreement with experimental \u0394G values (Pearson r = 0.54, Spearman \u03c1 = 0.58).<\/jats:p>","DOI":"10.1021\/acsphyschemau.6c00060","type":"journal-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T14:34:09Z","timestamp":1785854049000},"page":"1012-1017","source":"Crossref","is-referenced-by-count":0,"title":["Learning Protein\u2013Protein\nBinding Free Energies\nfrom Interface Graphs and Physicochemical Descriptors"],"prefix":"10.1021","volume":"6","author":[{"given":"Qingshu","family":"Zhao","sequence":"first","affiliation":[{"name":"The University of Wisconsin-Milwaukee , , ,","place":["Milwaukee, Wisconsin, United States, 53211"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1198-1820","authenticated-orcid":true,"given":"Arjun","family":"Saha","sequence":"additional","affiliation":[{"name":"The University of Wisconsin-Milwaukee , , ,","place":["Milwaukee, Wisconsin, United States, 53211"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,1]]},"reference":[{"issue":"1","key":"2026092310063797800_cit1","doi-asserted-by":"publisher","first-page":"341","DOI":"10.1038\/s41392-024-02036-3","article-title":"New Insights into Protein\u2013Protein Interaction Modulators in\nDrug Discovery and Therapeutic Advance","volume":"9","author":"Nada","year":"2024","journal-title":"Sig\nTransduct Target Ther"},{"issue":"6","key":"2026092310063797800_cit2","doi-asserted-by":"publisher","first-page":"288","DOI":"10.5501\/wjv.v10.i6.288","article-title":"Protein-Protein\nInteractions: Methods, Databases, and Applications in Virus-Host Study","volume":"10","author":"Farooq","year":"2021","journal-title":"World J. Virol"},{"issue":"1","key":"2026092310063797800_cit3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41392-020-00315-3","article-title":"Recent Advances\nin the Development\nof Protein\u2013Protein Interactions Modulators: Mechanisms and\nClinical Trials","volume":"5","author":"Lu","year":"2020","journal-title":"Sig Transduct Target Ther"},{"issue":"4","key":"2026092310063797800_cit4","doi-asserted-by":"publisher","DOI":"10.1016\/j.jsb.2024.108118","article-title":"Studying Protein\u2013Protein\nInteractions: Latest and Most Popular Approaches","volume":"216","author":"Akbarzadeh","year":"2024","journal-title":"J. Struct. Biol."},{"issue":"4","key":"2026092310063797800_cit5","doi-asserted-by":"publisher","first-page":"423","DOI":"10.1007\/s10867-019-09537-1","article-title":"Gibbs Free\nEnergy of Protein-Protein Interactions Correlates with ATP Production\nin Cancer Cells","volume":"45","author":"Golas","year":"2019","journal-title":"J. Biol. Phys."},{"issue":"2","key":"2026092310063797800_cit6","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1016\/j.bpj.2022.12.019","article-title":"Thermodynamic\nAnalysis of an Entropically\nDriven, High-Affinity Nanobody-HIV P24 Interaction","volume":"122","author":"Brookes","year":"2023","journal-title":"Biophys. J."},{"key":"2026092310063797800_cit7","doi-asserted-by":"publisher","first-page":"10023","DOI":"10.1039\/D5CP01009H","article-title":"Exploring the Binding Free Energy\nLandscape of Intrinsically Disordered\nProtein-Protein Interactions: Insights into the AF9-BCOR Complex Implicated\nin Leukemia","volume":"27","author":"Sharma","year":"2025","journal-title":"Phys. Chem. Chem. Phys."},{"issue":"5","key":"2026092310063797800_cit8","doi-asserted-by":"publisher","first-page":"1509","DOI":"10.1021\/ct900587b","article-title":"Prediction of Absolute\nSolvation Free Energies Using\nMolecular Dynamics Free Energy Perturbation and the OPLS Force Field","volume":"6","author":"Shivakumar","year":"2010","journal-title":"J. Chem. Theory Comput."},{"issue":"16","key":"2026092310063797800_cit9","doi-asserted-by":"publisher","first-page":"9478","DOI":"10.1021\/acs.chemrev.9b00055","article-title":"End-Point\nBinding Free Energy Calculation with MM\/PBSA\nand MM\/GBSA: Strategies and Applications in Drug Design","volume":"119","author":"Wang","year":"2019","journal-title":"Chem. Rev."},{"issue":"6","key":"2026092310063797800_cit10","doi-asserted-by":"publisher","first-page":"1129","DOI":"10.1016\/j.neuron.2018.08.011","article-title":"Molecular Dynamics\nSimulation for All","volume":"99","author":"Hollingsworth","year":"2018","journal-title":"Neuron"},{"issue":"6","key":"2026092310063797800_cit11","doi-asserted-by":"publisher","first-page":"3031","DOI":"10.1021\/acs.jctc.7b00125","article-title":"The Rosetta\nAll-Atom Energy Function for Macromolecular Modeling and Design","volume":"13","author":"Alford","year":"2017","journal-title":"J. Chem. Theory Comput."},{"issue":"5","key":"2026092310063797800_cit12","doi-asserted-by":"publisher","first-page":"2602","DOI":"10.1021\/acs.jcim.4c01641","article-title":"Estimating Absolute\nProtein\u2013Protein Binding Free Energies by a Super Learner Model","volume":"65","author":"Chaves","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026092310063797800_cit13","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.07454","article-title":"Contacts-Based Prediction\nof Binding Affinity in Protein\u2013Protein\nComplexes","volume":"4","author":"Vangone","year":"2015","journal-title":"eLife"},{"key":"2026092310063797800_cit14","doi-asserted-by":"publisher","first-page":"1065703","DOI":"10.3389\/fbinf.2022.1065703","article-title":"Machine Learning Methods for Protein-Protein Binding\nAffinity Prediction in Protein Design","volume":"2","author":"Guo","year":"2022","journal-title":"Front.\nBioinform."},{"issue":"23","key":"2026092310063797800_cit15","doi-asserted-by":"publisher","first-page":"8796","DOI":"10.1021\/acs.jcim.4c01850","article-title":"ProAffinity-GNN: A\nNovel Approach to Structure-Based\nProtein\u2013Protein Binding Affinity Prediction via a Curated Data\nSet and Graph Neural Networks","volume":"64","author":"Zhou","year":"2024","journal-title":"J. Chem. Inf.\nModel."},{"issue":"1","key":"2026092310063797800_cit16","doi-asserted-by":"publisher","first-page":"1761","DOI":"10.1038\/s41598-020-80900-2","article-title":"Classification and Prediction of Protein\u2013Protein\nInteraction Interface Using Machine Learning Algorithm","volume":"11","author":"Das","year":"2021","journal-title":"Sci. Rep"},{"key":"2026092310063797800_cit17","doi-asserted-by":"publisher","first-page":"1214","DOI":"10.1016\/j.csbj.2024.03.009","article-title":"SpatialPPI: Three-Dimensional Space Protein-Protein\nInteraction Prediction\nwith AlphaFold Multimer","volume":"23","author":"Hu","year":"2024","journal-title":"Computational and Structural\nBiotechnology Journal"},{"issue":"12","key":"2026092310063797800_cit18","doi-asserted-by":"publisher","first-page":"2977","DOI":"10.1021\/jm030580l","article-title":"The PDBbind\nDatabase:\nCollection of Binding Affinities for Protein\u2013Ligand Complexes\nwith Known Three-Dimensional Structures","volume":"47","author":"Wang","year":"2004","journal-title":"J.\nMed. Chem."},{"issue":"D1","key":"2026092310063797800_cit19","doi-asserted-by":"publisher","first-page":"D1140","DOI":"10.1093\/nar\/gkt1043","article-title":"SAbDab: The Structural\nAntibody Database","volume":"42","author":"Dunbar","year":"2014","journal-title":"Nucleic Acids Res."},{"issue":"3","key":"2026092310063797800_cit20","doi-asserted-by":"publisher","first-page":"482","DOI":"10.1002\/pro.580","article-title":"A Structure-Based Benchmark for Protein-Protein\nBinding Affinity","volume":"20","author":"Kastritis","year":"2011","journal-title":"Protein Sci."},{"issue":"5","key":"2026092310063797800_cit21","doi-asserted-by":"publisher","first-page":"908","DOI":"10.1002\/prot.25260","article-title":"ATLAS: A Database\nLinking Binding\nAffinities with Structures for Wild-Type and Mutant TCR-pMHC Complexes","volume":"85","author":"Borrman","year":"2017","journal-title":"Proteins: Struct., Funct., Bioinf."},{"issue":"1","key":"2026092310063797800_cit22","doi-asserted-by":"publisher","first-page":"1316","DOI":"10.1038\/s41597-024-03997-4","article-title":"PPB-Affinity:\nProtein-Protein Binding Affinity Dataset\nfor AI-Based Protein Drug Discovery","volume":"11","author":"Liu","year":"2024","journal-title":"Sci. Data"},{"issue":"1","key":"2026092310063797800_cit23","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1016\/0022-2836(81)90087-5","article-title":"Identification of\nCommon Molecular Subsequences","volume":"147","author":"Smith","year":"1981","journal-title":"J. Mol. Biol."},{"issue":"1","key":"2026092310063797800_cit24","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1186\/s13040-020-00231-w","article-title":"ISLAND:\nIn-Silico Proteins Binding Affinity Prediction Using Sequence Information","volume":"13","author":"Abbasi","year":"2020","journal-title":"BioData Mining"},{"issue":"8","key":"2026092310063797800_cit25","doi-asserted-by":"publisher","first-page":"1829","DOI":"10.1021\/acs.jproteome.2c00020","article-title":"PPI-Affinity:\nA Web Tool for the\nPrediction and Optimization of Protein\u2013Peptide and Protein\u2013Protein\nBinding Affinity","volume":"21","author":"Romero-Molina","year":"2022","journal-title":"J. Proteome Res."},{"issue":"10","key":"2026092310063797800_cit26","doi-asserted-by":"publisher","first-page":"7257","DOI":"10.1039\/D2CP05644E","article-title":"An Artificial Neural\nNetwork Model to Predict Structure-Based Protein\u2013Protein Free\nEnergy of Binding from Rosetta-Calculated Properties","volume":"25","author":"Ferraz","year":"2023","journal-title":"Phys. Chem. Chem. Phys."},{"key":"2026092310063797800_cit27","unstructured":"Sundararajan, M.; Taly, A.; Yan, Q.\n          Axiomatic\nAttribution for Deep\nNetworks.  arXiv June 13, 2017. 10.48550\/arXiv.1703.01365."},{"key":"2026092310063797800_cit28","unstructured":"Kokhlikyan, N.; Miglani, V.; Martin, M.; Wang, E.; Alsallakh, B.; Reynolds, J.; Melnikov, A.; Kliushkina, N.; Araya, C.; Yan, S.; Reblitz-Richardson, O.\n          Captum: A Unified and Generic\nModel Interpretability Library for PyTorch.  arXiv September 16, 2020. 10.48550\/arXiv.2009.07896."},{"issue":"1","key":"2026092310063797800_cit29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1006\/jmbi.1998.1843","article-title":"Anatomy of Hot Spots\nin Protein Interfaces","volume":"280","author":"Bogan","year":"1998","journal-title":"J. Mol. Biol."},{"issue":"9","key":"2026092310063797800_cit30","doi-asserted-by":"publisher","first-page":"716","DOI":"10.1038\/nchembio.1580","article-title":"Comprehensive\nAnalysis of Loops at Protein-Protein Interfaces for Macrocycle Design","volume":"10","author":"Gavenonis","year":"2014","journal-title":"Nat. Chem. Biol."},{"issue":"Supplement_1","key":"2026092310063797800_cit31","doi-asserted-by":"publisher","first-page":"i544","DOI":"10.1093\/bioinformatics\/btad231","article-title":"Deep Local Analysis\nDeconstructs Protein\u2013Protein Interfaces and Accurately Estimates\nBinding Affinity Changes upon Mutation","volume":"39","author":"Mohseni Behbahani","year":"2023","journal-title":"Bioinformatics"},{"issue":"48","key":"2026092310063797800_cit32","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2200018119","article-title":"Learning\nthe Relationship between\nNanoscale Chemical Patterning and Hydrophobicity","volume":"119","author":"Rego","year":"2022","journal-title":"Proc. Natl. Acad. Sci. U.S.A."},{"issue":"1","key":"2026092310063797800_cit33","doi-asserted-by":"publisher","first-page":"12087","DOI":"10.1038\/s41598-022-16338-5","article-title":"Investigating the Side-Chain Structural\nOrganization\nbehind the Stability of Protein Folding and Binding","volume":"12","author":"Desantis","year":"2022","journal-title":"Sci. Rep"},{"issue":"7","key":"2026092310063797800_cit34","doi-asserted-by":"publisher","first-page":"2576","DOI":"10.1021\/pr070018+","article-title":"Protein\u2013Protein Interfaces:\nProperties, Preferences, and Projections","volume":"6","author":"Headd","year":"2007","journal-title":"J.\nProteome Res."},{"issue":"2","key":"2026092310063797800_cit35","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1016\/j.jmb.2009.03.058","article-title":"Insights into Positive and Negative\nRequirements for Protein\u2013Protein Interactions by Crystallographic\nAnalysis of the \u03b2-Lactamase Inhibitory Proteins BLIP, BLIP-I,\nand BLP","volume":"389","author":"Gretes","year":"2009","journal-title":"J. Mol. Biol."},{"issue":"27","key":"2026092310063797800_cit36","doi-asserted-by":"publisher","first-page":"11034","DOI":"10.1073\/pnas.0902463106","article-title":"Structure\nof a Serine Protease Poised\nto Resynthesize a Peptide Bond","volume":"106","author":"Zakharova","year":"2009","journal-title":"Proc. Natl.\nAcad. Sci. U. S. A."},{"issue":"2","key":"2026092310063797800_cit37","doi-asserted-by":"publisher","first-page":"247","DOI":"10.1107\/S0907444902020887","article-title":"Structure of a Hybrid Squash Inhibitor\nin Complex with\nPorcine Pancreatic Elastase at 1.8 \u00c5 Resolution","volume":"59","author":"Ay\u0308","year":"2003","journal-title":"Acta Cryst. D"},{"issue":"4","key":"2026092310063797800_cit38","doi-asserted-by":"publisher","first-page":"480","DOI":"10.1107\/S010876818300275X","article-title":"The Geometry\nof the Reactive Site and of the Peptide Groups in Trypsin, Trypsinogen\nand Its Complexes with Inhibitors","volume":"39","author":"Marquart","year":"1983","journal-title":"Acta Crystallographica\nSection B"},{"issue":"11","key":"2026092310063797800_cit39","doi-asserted-by":"publisher","first-page":"5988","DOI":"10.1093\/nar\/gkz355","article-title":"Alternative\n\u03c3I\/Anti-\u03c3I Factors Represent a Unique Form of Bacterial\n\u03c3\/Anti-\u03c3 Complex","volume":"47","author":"Wei","year":"2019","journal-title":"Nucleic Acids\nRes."},{"issue":"49","key":"2026092310063797800_cit40","doi-asserted-by":"publisher","first-page":"38524","DOI":"10.1074\/jbc.M110.172072","article-title":"Two Independent\nHistidines, One in Human Prolactin and One in Its Receptor, Are Critical\nfor pH-Dependent Receptor Recognition and Activation *","volume":"285","author":"Kulkarni","year":"2010","journal-title":"J. Biol. Chem."},{"issue":"2","key":"2026092310063797800_cit41","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1038\/nsmb.2231","article-title":"Structural Basis\nfor Autoinhibition and Phosphorylation-Dependent\nActivation of c-Cbl","volume":"19","author":"Dou","year":"2012","journal-title":"Nat. Struct Mol. Biol."}],"container-title":["ACS Physical Chemistry Au"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acsphyschemau.6c00060","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/apcach\/article-pdf\/6\/5\/1012\/66353733\/acsphyschemau.6c00060.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/apcach\/article-pdf\/6\/5\/1012\/66353733\/acsphyschemau.6c00060.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T15:46:23Z","timestamp":1790178383000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/apcach\/article\/6\/5\/1012\/5238271\/Learning-Protein-Protein-Binding-Free-Energies"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,1]]},"references-count":41,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,8,1]]},"published-print":{"date-parts":[[2026,9,23]]}},"URL":"https:\/\/doi.org\/10.1021\/acsphyschemau.6c00060","relation":{},"ISSN":["2694-2445"],"issn-type":[{"value":"2694-2445","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,1]]}}}