{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T20:46:53Z","timestamp":1792788413370,"version":"4.5.0"},"reference-count":52,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001711","name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","doi-asserted-by":"publisher","award":["208179"],"award-info":[{"award-number":["208179"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001711","name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","doi-asserted-by":"publisher","award":["219859"],"award-info":[{"award-number":["219859"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"publisher"}]},{"name":"EPFL and the Ludwig Institute for Cancer Research","award":["core funding"],"award-info":[{"award-number":["core funding"]}],"id":[{"id":"https:\/\/ror.org\/05qdwtz81","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["www.plosbiology.org"],"crossmark-restriction":false},"short-container-title":["PLoS Biol"],"abstract":"<jats:p>Recent advances in artificial intelligence have transformed protein structure prediction and design. However, protein function is governed not only by static structures but also by the conformational dynamics that allow proteins to access distinct functional states. Predicting these dynamic transitions, central to many biological processes, remains challenging due to the scarcity of high-resolution experimental data, which limits the training of machine-learning models for dynamic and energetic property prediction. Here, we present AlloPool, a graph neural network (GNN)-based framework that interprets molecular dynamics simulations by iteratively pruning residue\u2013residue interactions to uncover minimal, time-resolved interaction networks that govern protein conformational dynamics and structural responses to chemical or mechanical perturbations, collectively known as allostery. By integrating temporal attention with graph aggregation, AlloPool learns evolving interaction graphs from equilibrium and non-equilibrium molecular dynamics simulations, enabling accurate reconstruction of dynamic trajectories and the interaction networks that drive conformational transitions. Validated across diverse dynamic protein systems, including binding domains, mechanosensors, signaling receptors, and enzymes, AlloPool maps allosteric communication pathways, predicts the effects of ligand binding, mechanical forces, and mutations, discovers transient dynamic states and outperforms existing machine-learning approaches in dynamic trajectory reconstruction. This advance provides a general framework for interpreting protein dynamics with broad implications for drug discovery, synthetic biology, and protein engineering.<\/jats:p>","DOI":"10.1371\/journal.pbio.3004002","type":"journal-article","created":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T17:49:42Z","timestamp":1790012982000},"page":"e3004002","update-policy":"https:\/\/doi.org\/10.1371\/journal.pbio.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["AlloPool is a deep learning framework that infers protein allostery from molecular dynamics simulations"],"prefix":"10.1371","volume":"24","author":[{"given":"Matthieu","family":"Marfoglia","sequence":"first","affiliation":[{"name":"Interfaculty Institute of Bioengineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","place":["Lausanne, Switzerland"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miguel A.","family":"Pedraza-Joya","sequence":"additional","affiliation":[{"name":"Interfaculty Institute of Bioengineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","place":["Lausanne, Switzerland"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lucas","family":"Guirardel","sequence":"additional","affiliation":[{"name":"Interfaculty Institute of Bioengineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","place":["Lausanne, Switzerland"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aisima","family":"Chatzi Souleiman","sequence":"additional","affiliation":[{"name":"Interfaculty Institute of Bioengineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","place":["Lausanne, Switzerland"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0744-6844","authenticated-orcid":true,"given":"Patrick","family":"Barth","sequence":"additional","affiliation":[{"name":"Interfaculty Institute of Bioengineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","place":["Lausanne, Switzerland"]},{"name":"Ludwig Institute for Cancer Research Lausanne","place":["Lausanne, Switzerland"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"issue":"6615","key":"pbio.3004002.ref001","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1126\/science.add2187","article-title":"Robust deep learning-based protein sequence design using ProteinMPNN","volume":"378","author":"J Dauparas","year":"2022","journal-title":"Science"},{"issue":"7873","key":"pbio.3004002.ref002","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":"J Jumper","year":"2021","journal-title":"Nature"},{"issue":"7976","key":"pbio.3004002.ref003","doi-asserted-by":"crossref","first-page":"1089","DOI":"10.1038\/s41586-023-06415-8","article-title":"De novo design of protein structure and function with RFdiffusion","volume":"620","author":"JL Watson","year":"2023","journal-title":"Nature"},{"key":"pbio.3004002.ref004","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","journal-title":"Science"},{"key":"pbio.3004002.ref005","article-title":"Molecular dynamics and protein function","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pbio.3004002.ref006","article-title":"Allostery and cooperativity revisited","year":"2008","journal-title":"Protein Sci"},{"issue":"4","key":"pbio.3004002.ref007","doi-asserted-by":"crossref","first-page":"566","DOI":"10.1016\/j.str.2019.01.003","article-title":"Allostery in its many disguises: from theory to applications","volume":"27","author":"SJ Wodak","year":"2019","journal-title":"Structure"},{"issue":"7496","key":"pbio.3004002.ref008","doi-asserted-by":"crossref","first-page":"331","DOI":"10.1038\/nature13001","article-title":"The ensemble nature of allostery","volume":"508","author":"HN Motlagh","year":"2014","journal-title":"Nature"},{"issue":"6","key":"pbio.3004002.ref009","article-title":"Allostery: an overview of its history, concepts, methods, and applications","volume":"12","author":"J Liu","year":"2016","journal-title":"PLoS Comput Biol"},{"key":"pbio.3004002.ref010","article-title":"Controlling allosteric networks in proteins","journal-title":"Chemical Reviews"},{"key":"pbio.3004002.ref011","doi-asserted-by":"crossref","first-page":"6503","DOI":"10.1021\/acs.chemrev.5b00590","article-title":"Protein allostery and conformational dynamics","volume":"116","author":"J Guo","year":"2016","journal-title":"Chem Rev"},{"issue":"4","key":"pbio.3004002.ref012","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1038\/s41573-024-01083-3","article-title":"Functional dynamics of G protein-coupled receptors reveal new routes for drug discovery","volume":"24","author":"P Conflitti","year":"2025","journal-title":"Nat Rev Drug Discov"},{"issue":"3","key":"pbio.3004002.ref013","doi-asserted-by":"crossref","first-page":"1423","DOI":"10.1021\/acs.jctc.1c01055","article-title":"GLOW: a workflow integrating gaussian-accelerated molecular dynamics and deep learning for free energy profiling","volume":"18","author":"HN Do","year":"2022","journal-title":"J Chem Theory Comput"},{"key":"pbio.3004002.ref014","doi-asserted-by":"crossref","first-page":"558","DOI":"10.1038\/s42256-024-00837-3","article-title":"Predicting equilibrium distributions for molecular systems with deep learning","volume":"6","author":"S Zheng","year":"2024","journal-title":"Nat Mach Intell"},{"issue":"6457","key":"pbio.3004002.ref015","article-title":"Boltzmann generators: sampling equilibrium states of many-body systems with deep learning","volume":"365","author":"F No\u00e9","year":"2019","journal-title":"Science"},{"issue":"1","key":"pbio.3004002.ref016","doi-asserted-by":"crossref","first-page":"5739","DOI":"10.1038\/s41467-023-41343-1","article-title":"Machine learning coarse-grained potentials of protein thermodynamics","volume":"14","author":"M Majewski","year":"2023","journal-title":"Nat Commun"},{"key":"pbio.3004002.ref017","doi-asserted-by":"crossref","first-page":"681364","DOI":"10.3389\/fmolb.2021.681364","article-title":"Editorial: understanding protein dynamics, binding and allostery for drug design","volume":"8","author":"G Hu","year":"2021","journal-title":"Front Mol Biosci"},{"key":"pbio.3004002.ref018","doi-asserted-by":"crossref","first-page":"102768","DOI":"10.1016\/j.sbi.2023.102768","article-title":"Protein dynamics underlying allosteric regulation","volume":"84","author":"MA Astore","year":"2024","journal-title":"Curr Opin Struct Biol"},{"issue":"1","key":"pbio.3004002.ref019","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1021\/acs.chemrev.6b00177","article-title":"GPCR dynamics: structures in motion","volume":"117","author":"NR Latorraca","year":"2017","journal-title":"Chem Rev"},{"issue":"5","key":"pbio.3004002.ref020","doi-asserted-by":"crossref","first-page":"1101","DOI":"10.1016\/j.cell.2015.04.043","article-title":"Structural insights into the dynamic process of \u03b22-adrenergic receptor signaling","volume":"161","author":"A Manglik","year":"2015","journal-title":"Cell"},{"key":"pbio.3004002.ref021","doi-asserted-by":"crossref","first-page":"102820","DOI":"10.1016\/j.sbi.2024.102820","article-title":"Probing allosteric communication with combined molecular dynamics simulations and network analysis","volume":"86","author":"M Bernetti","year":"2024","journal-title":"Curr Opin Struct Biol"},{"issue":"14","key":"pbio.3004002.ref022","doi-asserted-by":"crossref","first-page":"3590","DOI":"10.1093\/bioinformatics\/btac380","article-title":"Prediction of allosteric communication pathways in proteins","volume":"38","author":"T Haliloglu","year":"2022","journal-title":"Bioinformatics"},{"issue":"8","key":"pbio.3004002.ref023","doi-asserted-by":"crossref","first-page":"899","DOI":"10.1016\/j.bbapap.2018.04.008","article-title":"Dynamics-based community analysis and perturbation response scanning of allosteric interaction networks in the TRAP1 chaperone structures dissect molecular linkage between conformational asymmetry and sequential ATP hydrolysis","volume":"1866","author":"GM Verkhivker","year":"2018","journal-title":"Biochim Biophys Acta Proteins Proteom"},{"issue":"1","key":"pbio.3004002.ref024","doi-asserted-by":"crossref","first-page":"3862","DOI":"10.1038\/s41467-020-17618-2","article-title":"Mapping allosteric communications within individual proteins","volume":"11","author":"J Wang","year":"2020","journal-title":"Nat Commun"},{"issue":"1","key":"pbio.3004002.ref025","doi-asserted-by":"crossref","first-page":"2875","DOI":"10.1038\/s41467-023-38491-9","article-title":"Computational design of dynamic receptor-peptide signaling complexes applied to chemotaxis","volume":"14","author":"RE Jefferson","year":"2023","journal-title":"Nat Commun"},{"key":"pbio.3004002.ref026","unstructured":"Dynamic neural relational inference for forecasting trajectories | IEEE Conference Publication | IEEE Xplore. Available from: https:\/\/ieeexplore.ieee.org\/document\/9150845"},{"issue":"3","key":"pbio.3004002.ref027","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1162\/netn_a_00252","article-title":"Forecasting brain activity based on models of spatiotemporal brain dynamics: a comparison of graph neural network architectures","volume":"6","author":"S Wein","year":"2022","journal-title":"Netw Neurosci"},{"key":"pbio.3004002.ref028","doi-asserted-by":"crossref","first-page":"57","DOI":"10.1016\/j.aiopen.2021.01.001","article-title":"Graph neural networks: a review of methods and applications","volume":"1","author":"J Zhou","year":"2020","journal-title":"AI Open"},{"key":"pbio.3004002.ref029","article-title":"A deep graph neural network architecture for modelling spatio-temporal dynamics in resting-stating functional MRI data","author":"T Azevedo","year":"2020","journal-title":"bioRxiv"},{"key":"pbio.3004002.ref030","unstructured":"Kipf T, Fetaya E, Wang KC, Welling M, Zemel R. Neural relational inference for interacting systems. 2018. Available from: https:\/\/doi.org\/10.48550\/arXiv.1802.04687"},{"key":"pbio.3004002.ref031","article-title":"Neural relational inference to learn allosteric long-range interactions in proteins from molecular dynamics simulations","author":"J Zhu","year":"2021"},{"issue":"7475","key":"pbio.3004002.ref032","doi-asserted-by":"crossref","first-page":"295","DOI":"10.1038\/nature12595","article-title":"Structural basis for modulation of a G-protein-coupled receptor by allosteric drugs","volume":"503","author":"RO Dror","year":"2013","journal-title":"Nature"},{"key":"pbio.3004002.ref033","doi-asserted-by":"crossref","first-page":"1070328","DOI":"10.3389\/fmolb.2022.1070328","article-title":"Recent applications of computational methods to allosteric drug discovery","volume":"9","author":"RG Govindaraj","year":"2023","journal-title":"Front Mol Biosci"},{"issue":"14","key":"pbio.3004002.ref034","doi-asserted-by":"crossref","first-page":"7123","DOI":"10.1073\/pnas.1811066116","article-title":"Molecular dynamics-guided discovery of an ago-allosteric modulator for GPR40\/FFAR1","volume":"116","author":"M L\u00fcckmann","year":"2019","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pbio.3004002.ref035","author":"H Zhao","year":"2021"},{"issue":"9","key":"pbio.3004002.ref036","doi-asserted-by":"crossref","first-page":"3848","DOI":"10.1109\/TITS.2019.2935152","article-title":"T-GCN: a temporal graph convolutional network for traffic prediction","volume":"21","author":"L Zhao","year":"2020","journal-title":"IEEE Trans Intell Transport Syst"},{"key":"pbio.3004002.ref037","doi-asserted-by":"crossref","unstructured":"Liu H, Zhu C, Zhang D, Li Q. Attention-based spatial-temporal graph convolutional recurrent networks for traffic forecasting. In: 2023. https:\/\/doi.org\/10.48550\/arXiv.2302.12973","DOI":"10.1007\/978-3-031-46661-8_42"},{"issue":"38","key":"pbio.3004002.ref038","doi-asserted-by":"crossref","first-page":"10095","DOI":"10.1073\/pnas.1708810114","article-title":"Stachel-independent modulation of GPR56\/ADGRG1 signaling by synthetic ligands directed to its extracellular region","volume":"114","author":"GS Salzman","year":"2017","journal-title":"Proc Natl Acad Sci U S A"},{"key":"pbio.3004002.ref039","author":"SP Kordon","year":"2024"},{"key":"pbio.3004002.ref040","article-title":"Uncovering and engineering the mechanical properties of the adhesion GPCR ADGRG1 GAIN domain","author":"L Dumas","year":"2023","journal-title":"bioRxiv"},{"issue":"6383","key":"pbio.3004002.ref041","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1126\/science.aar2094","article-title":"Molecular mechanism of extreme mechanostability in a pathogen adhesin","volume":"359","author":"LF Milles","year":"2018","journal-title":"Science"},{"issue":"7819","key":"pbio.3004002.ref042","doi-asserted-by":"crossref","first-page":"125","DOI":"10.1038\/s41586-020-2379-5","article-title":"Structure of a D2 dopamine receptor-G-protein complex in a lipid membrane","volume":"584","author":"J Yin","year":"2020","journal-title":"Nature"},{"key":"pbio.3004002.ref043","unstructured":"PDZ Domain Proteins: Plug and Play! | Science\u2019s STKE. Available from: https:\/\/www.science.org\/doi\/10.1126\/stke.2003.179.re7"},{"issue":"1","key":"pbio.3004002.ref044","doi-asserted-by":"crossref","first-page":"6232","DOI":"10.1038\/s41467-022-33687-x","article-title":"Atomic resolution protein allostery from the multi-state structure of a PDZ domain","volume":"13","author":"D Ashkinadze","year":"2022","journal-title":"Nat Commun"},{"issue":"7","key":"pbio.3004002.ref045","doi-asserted-by":"crossref","first-page":"1564","DOI":"10.1016\/j.cell.2011.10.049","article-title":"Hot spots for allosteric regulation on protein surfaces","volume":"147","author":"KA Reynolds","year":"2011","journal-title":"Cell"},{"issue":"7061","key":"pbio.3004002.ref046","doi-asserted-by":"crossref","first-page":"1053","DOI":"10.1038\/nature04054","article-title":"Solution structure of a protein denatured state and folding intermediate","volume":"437","author":"TL Religa","year":"2005","journal-title":"Nature"},{"issue":"2","key":"pbio.3004002.ref047","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbae006","article-title":"MoDAFold: a strategy for predicting the structure of missense mutant protein based on AlphaFold2 and molecular dynamics","volume":"25","author":"L Zheng","year":"2024","journal-title":"Brief Bioinform"},{"issue":"1","key":"pbio.3004002.ref048","doi-asserted-by":"crossref","first-page":"4546","DOI":"10.1038\/s41467-022-32340-x","article-title":"Ligand-specific conformational change drives interdomain allostery in Pin1","volume":"13","author":"A Born","year":"2022","journal-title":"Nat Commun"},{"key":"pbio.3004002.ref049","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1016\/j.softx.2015.06.001","article-title":"GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers","author":"MJ Abraham","year":"2015","journal-title":"SoftwareX"},{"issue":"1","key":"pbio.3004002.ref050","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1038\/nmeth.4067","article-title":"CHARMM36m: an improved force field for folded and intrinsically disordered proteins","volume":"14","author":"J Huang","year":"2017","journal-title":"Nat Methods"},{"issue":"11","key":"pbio.3004002.ref051","doi-asserted-by":"crossref","first-page":"1859","DOI":"10.1002\/jcc.20945","article-title":"CHARMM-GUI: a web-based graphical user interface for CHARMM","volume":"29","author":"S Jo","year":"2008","journal-title":"J Comput Chem"},{"key":"pbio.3004002.ref052","doi-asserted-by":"crossref","first-page":"643","DOI":"10.1007\/978-1-4939-7808-3_41","article-title":"Virus-based cancer therapeutics for targeted photodynamic therapy.","author":"B Cao","year":"2018","journal-title":"Virus-derived nanoparticles for advanced technologies: methods and protocols. New York, NY: Springer"}],"container-title":["PLOS Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pbio.3004002","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T17:49:52Z","timestamp":1790012992000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pbio.3004002"}},"subtitle":[],"editor":[{"given":"Chaitan","family":"Khosla","sequence":"first","affiliation":[{"name":"Stanford University","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":52,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,21]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pbio.3004002","relation":{},"ISSN":["1545-7885"],"issn-type":[{"value":"1545-7885","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]}}}