{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T16:03:04Z","timestamp":1783699384315,"version":"3.55.0"},"reference-count":39,"publisher":"Oxford University Press (OUP)","issue":"6","license":[{"start":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T00:00:00Z","timestamp":1778803200000},"content-version":"vor","delay-in-days":1,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Reliable predictions of protein\u2013protein binding affinities are essential for molecular biology and therapeutic discovery. However, most computational methods rely on three-dimensional structural models, which are often unavailable for many complexes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We introduce BindPred, a structure-agnostic input framework that predicts affinities directly from amino acid sequences by combining embeddings from large protein language models with gradient boosting trees. On the protein\u2013protein binding (PPB)-Affinity benchmark, which comprises 11 919 diverse complexes, BindPred achieves a Pearson correlation coefficient of 0.86 in random split five-fold cross-validation. Ablation analysis indicates that evolutionary embeddings alone capture most of the predictive signals, while augmenting with physics-based energy terms from PyRosetta and BindCraft increases the correlation only by 0.01. A more stringent protein-level split that places entire protein families (wild-type and all mutants) exclusively in either training or testing sets, resulting in only a modest decline in performance, demonstrating robust generalization to novel interaction pairs. Because BindPred operates exclusively on sequence input, it enables rapid inference [approximately 3 million complexes per GPU (T4) hour], making proteome-scale screening computationally feasible.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability<\/jats:title>\n                    <jats:p>The pretrained model and inference pipeline are available in a Google Colab notebook: BindPred Colab notebook. The training dataset, code, and model weights are available on the hugging face: https:\/\/huggingface.co\/hbp5181\/BindPred.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag309","type":"journal-article","created":{"date-parts":[[2026,5,12]],"date-time":"2026-05-12T11:43:52Z","timestamp":1778586232000},"source":"Crossref","is-referenced-by-count":1,"title":["BindPred: a framework for predicting protein\u2013protein binding affinity from language model embeddings"],"prefix":"10.1093","volume":"42","author":[{"given":"Haixing","family":"Piao","sequence":"first","affiliation":[{"name":"Department of Chemical Engineering, The Pennsylvania State University , University Park, PA 16802,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Veda Sheersh","family":"Boorla","sequence":"additional","affiliation":[{"name":"Department of Chemical Engineering, The Pennsylvania State University , University Park, PA 16802,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Somtirtha","family":"Santra","sequence":"additional","affiliation":[{"name":"Department of Chemical Engineering, The Pennsylvania State University , University Park, PA 16802,","place":["United States"]},{"name":"The Center for Bioenergy Innovation , Oak Ridge National Laboratory, Oak Ridge, TN 37831,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1508-1398","authenticated-orcid":false,"given":"Costas D","family":"Maranas","sequence":"additional","affiliation":[{"name":"Department of Chemical Engineering, The Pennsylvania State University , University Park, PA 16802,","place":["United States"]},{"name":"The Center for Bioenergy Innovation , Oak Ridge National Laboratory, Oak Ridge, TN 37831,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,5,14]]},"reference":[{"key":"2026062409121943800_btag309-B1","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate structure prediction of biomolecular interactions with AlphaFold 3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"key":"2026062409121943800_btag309-B2","doi-asserted-by":"crossref","first-page":"3031","DOI":"10.1021\/acs.jctc.7b00125","article-title":"The Rosetta All-Atom energy function for macromolecular modeling and design","volume":"13","author":"Alford","year":"2017","journal-title":"J Chem Theory Comput"},{"key":"2026062409121943800_btag309-B3","author":"Banerjee","year":"2025"},{"key":"2026062409121943800_btag309-B5","doi-asserted-by":"crossref","first-page":"908","DOI":"10.1002\/prot.25260","article-title":"ATLAS: a database linking binding affinities with structures for wild-type and mutant TCR-pMHC complexes","volume":"85","author":"Borrman","year":"2017","journal-title":"Proteins"},{"key":"2026062409121943800_btag309-B6","doi-asserted-by":"crossref","first-page":"1265","DOI":"10.1038\/s41467-022-28865-w","article-title":"Improved prediction of protein\u2013protein interactions using AlphaFold2","volume":"13","author":"Bryant","year":"2022","journal-title":"Nat Commun"},{"key":"2026062409121943800_btag309-B7","doi-asserted-by":"crossref","first-page":"689","DOI":"10.1093\/bioinformatics\/btq007","article-title":"PyRosetta: a script-based interface for implementing molecular modeling algorithms using rosetta","volume":"26","author":"Chaudhury","year":"2010","journal-title":"Bioinformatics"},{"key":"2026062409121943800_btag309-B8","doi-asserted-by":"crossref","first-page":"2602","DOI":"10.1021\/acs.jcim.4c01641","article-title":"Estimating absolute protein\u2013protein binding free energies by a super learner model","volume":"65","author":"Chaves","year":"2025","journal-title":"J Chem Inf Model"},{"key":"2026062409121943800_btag309-B9","doi-asserted-by":"crossref","first-page":"6600","DOI":"10.1038\/s41467-022-34232-6","article-title":"Absolute protein quantification using fluorescence measurements with FPCountR","volume":"13","author":"Csibra","year":"2022","journal-title":"Nat Commun"},{"key":"2026062409121943800_btag309-B10","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1126\/science.add2187","article-title":"Robust deep learning\u2013based protein sequence design using ProteinMPNN","volume":"378","author":"Dauparas","year":"2022","journal-title":"Science"},{"key":"2026062409121943800_btag309-B12","first-page":"1","article-title":"ProtTrans: towards cracking the language of lifes code through Self-Supervised deep learning and high performance computing","volume":"44","author":"Elnaggar","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2026062409121943800_btag309-B13","author":"Evans","year":"2021"},{"key":"2026062409121943800_btag309-B14","doi-asserted-by":"crossref","first-page":"606","DOI":"10.1016\/j.str.2021.01.005","article-title":"An expanded benchmark for antibody-antigen docking and affinity prediction reveals insights into antibody recognition determinants","volume":"29","author":"Guest","year":"2021","journal-title":"Structure"},{"key":"2026062409121943800_btag309-B15","doi-asserted-by":"crossref","DOI":"10.1093\/bioinformatics\/btaf270","article-title":"ProBASS\u2014a language model with sequence and structural features for predicting the effect of mutations on binding affinity","volume":"41","author":"Gurusinghe","year":"2025","journal-title":"Bioinformatics"},{"key":"2026062409121943800_btag309-B16","doi-asserted-by":"crossref","first-page":"729513","DOI":"10.3389\/fmolb.2021.729513","article-title":"Structural biology in the clouds: the WeNMR-EOSC ecosystem","volume":"8","author":"Honorato","year":"2021","journal-title":"Front Mol Biosci"},{"key":"2026062409121943800_btag309-B17","author":"Hsu","year":"2022"},{"key":"2026062409121943800_btag309-B18","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1093\/bioinformatics\/bty635","article-title":"SKEMPI 2.0: an updated benchmark of changes in protein\u2013protein binding energy, kinetics and thermodynamics upon mutation","volume":"35","author":"Jankauskaite","year":"2019","journal-title":"Bioinformatics"},{"key":"2026062409121943800_btag309-B19","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein structure prediction with alphafold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"key":"2026062409121943800_btag309-B20","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbae480","article-title":"Protein language models are performant in structure-free virtual screening","volume":"25","author":"Lam","year":"2024","journal-title":"Brief Bioinform"},{"key":"2026062409121943800_btag309-B21","doi-asserted-by":"crossref","first-page":"1498","DOI":"10.1002\/prot.26821","article-title":"PCANN program for structure-based prediction of protein\u2013protein binding affinity: comparison with other neural-network predictors","volume":"93","author":"Lebedenko","year":"2025","journal-title":"Proteins"},{"key":"2026062409121943800_btag309-B22","doi-asserted-by":"crossref","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"key":"2026062409121943800_btag309-B23","doi-asserted-by":"crossref","first-page":"1316","DOI":"10.1038\/s41597-024-03997-4","article-title":"PPB-Affinity: protein\u2013protein binding affinity dataset for AI-based protein drug discovery","volume":"11","author":"Liu","year":"2024","journal-title":"Sci Data"},{"key":"2026062409121943800_btag309-B24","author":"Lou","year":"2025"},{"key":"2026062409121943800_btag309-B25","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1021\/acs.jcim.5c02428","article-title":"Advances in machine learning models for predicting enzyme kinetic parameters","volume":"66","author":"Malli","year":"2026","journal-title":"J Chem Inf Model"},{"key":"2026062409121943800_btag309-B27","doi-asserted-by":"crossref","first-page":"1629","DOI":"10.1007\/s00439-021-02411-y","article-title":"Embeddings from protein language models predict conservation and variant effects","volume":"141","author":"Marquet","year":"2022","journal-title":"Hum Genet"},{"key":"2026062409121943800_btag309-B28","doi-asserted-by":"crossref","DOI":"10.1073\/pnas.1111471108","article-title":"Direct-coupling analysis of residue coevolution captures native contacts across many protein families","volume":"108","author":"Morcos","year":"2011","journal-title":"Proc Natl Acad Sci USA"},{"key":"2026062409121943800_btag309-B29","doi-asserted-by":"crossref","first-page":"483","DOI":"10.1038\/s41586-025-09429-6","article-title":"One-shot design of functional protein binders with BindCraft","volume":"646","author":"Pacesa","year":"2025","journal-title":"Nature"},{"key":"2026062409121943800_btag309-B30","doi-asserted-by":"crossref","first-page":"2102","DOI":"10.1021\/acs.jctc.2c01237","article-title":"Free energy differences from molecular simulations: exact confidence intervals from transition counts","volume":"19","author":"K\u0159\u00ed\u017e","year":"2023","journal-title":"J Chem Theory Comput"},{"key":"2026062409121943800_btag309-B31","doi-asserted-by":"crossref","first-page":"i590","DOI":"10.1093\/bioinformatics\/btaf192","article-title":"Deep learning models for unbiased sequence-based PPI prediction Plateau at an accuracy of 0.65","volume":"41","author":"Reim","year":"2025","journal-title":"Bioinformatics"},{"key":"2026062409121943800_btag309-B33","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1186\/s12859-024-05737-2","article-title":"Using protein language models for protein interaction hot spot prediction with limited data","volume":"25","author":"Sargsyan","year":"2024","journal-title":"BMC Bioinformatics"},{"key":"2026062409121943800_btag309-B34","doi-asserted-by":"crossref","first-page":"1724","DOI":"10.1002\/prot.22689","article-title":"A comprehensive examination of the contributions to the binding entropy of protein-ligand complexes","volume":"78","author":"Singh","year":"2010","journal-title":"Proteins"},{"key":"2026062409121943800_btag309-B35","doi-asserted-by":"crossref","first-page":"969","DOI":"10.1016\/j.cels.2021.08.010","article-title":"D-SCRIPT translates genome to phenome with sequence-based, structure-aware, genome-scale predictions of protein\u2013protein interactions","volume":"12","author":"Sledzieski","year":"2021","journal-title":"Cell Syst"},{"key":"2026062409121943800_btag309-B36","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1007\/978-1-4939-8760-3_12","article-title":"Use of surface plasmon resonance (SPR) to determine binding affinities and kinetic parameters between components important in fusion machinery","volume":"1860","author":"Sparks","year":"2019","journal-title":"Methods Mol Biol"},{"key":"2026062409121943800_btag309-B38","doi-asserted-by":"crossref","first-page":"1026","DOI":"10.1038\/nbt.3988","article-title":"MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets","volume":"35","author":"Steinegger","year":"2017","journal-title":"Nat Biotechnol"},{"key":"2026062409121943800_btag309-B40","article-title":"The STRING database in 2025: protein networks with directionality of regulation","volume":"53","author":"Szklarczyk","year":"2024","journal-title":"Nucleic Acids Res"},{"key":"2026062409121943800_btag309-B41","doi-asserted-by":"crossref","first-page":"4391","DOI":"10.1038\/s41598-025-88445-y","article-title":"PIPENN-EMB ensemble net and protein embeddings generalise protein interface prediction beyond homology","volume":"15","author":"Thomas","year":"2025","journal-title":"Sci Rep"},{"key":"2026062409121943800_btag309-B42","doi-asserted-by":"crossref","first-page":"1199","DOI":"10.1038\/s41467-025-67971-3","article-title":"Learning the language of protein\u2013protein interactions","volume":"17","author":"Ullanat","year":"2026","journal-title":"Nat Commun"},{"key":"2026062409121943800_btag309-B628861","doi-asserted-by":"publisher","first-page":"3031","DOI":"10.1016\/j.jmb.2015.07.016","article-title":"Updates to the integrated protein\\u2013protein interaction benchmarks: docking benchmark version 5 and affinity benchmark version 2","volume":"427","author":"Vreven","year":"2015","journal-title":"Journal of Molecular Biology"},{"key":"2026062409121943800_btag309-B43","author":"Wee","year":"2024"},{"key":"2026062409121943800_btag309-B44","author":"Wu","year":"2024"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btag309\/68289286\/btag309.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/6\/btag309\/68289286\/btag309.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/6\/btag309\/68289286\/btag309.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T13:12:26Z","timestamp":1782306746000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag309\/8678529"}},"subtitle":[],"editor":[{"given":"Daisuke","family":"Kihara","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,5,14]]},"references-count":39,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2026,6,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag309","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2025.09.27.678407","asserted-by":"object"}]},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,6]]},"published":{"date-parts":[[2026,5,14]]},"article-number":"btag309"}}