{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,26]],"date-time":"2026-04-26T09:16:45Z","timestamp":1777195005055,"version":"3.51.4"},"reference-count":51,"publisher":"AIP Publishing","issue":"19","content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2023,5,21]]},"abstract":"<jats:p>The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological activity. While Molecular Dynamics (MD) simulations have emerged as the method of choice for modeling protein-ligand binding events, the high dimensional nature of the MD-derived trajectories often remains a barrier in the statistical elucidation of distinct ligand binding pathways due to the stochasticity inherent in the ligand\u2019s fluctuation in the solution and around the receptor. Here, we demonstrate that an autoencoder based deep neural network, trained using an objective input feature of a large matrix of residue\u2013ligand distances, can efficiently produce an optimal low-dimensional latent space that stores necessary information on the ligand-binding event. In particular, for a system of L99A mutant of T4 lysozyme interacting with its native ligand, benzene, this deep encoder\u2013decoder framework automatically identifies multiple distinct recognition pathways, without requiring user intervention. The intermediates involve the spatially discrete location of the ligand in different helices of the protein before its eventual recognition of native pose. The compressed subspace derived from the autoencoder provides a quantitatively accurate measure of the free energy and kinetics of ligand binding to the native pocket. The investigation also recommends that while a linear dimensional reduction technique, such as time-structured independent component analysis, can do a decent job of state-space decomposition in cases where the intermediates are long-lived, autoencoder is the method of choice in systems where transient, low-populated intermediates can lead to multiple ligand-binding pathways.<\/jats:p>","DOI":"10.1063\/5.0145197","type":"journal-article","created":{"date-parts":[[2023,5,15]],"date-time":"2023-05-15T10:23:01Z","timestamp":1684146181000},"update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":6,"title":["A deep encoder\u2013decoder framework for identifying distinct ligand binding pathways"],"prefix":"10.1063","volume":"158","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0927-3326","authenticated-orcid":false,"given":"Satyabrata","family":"Bandyopadhyay","sequence":"first","affiliation":[{"name":"Tata Institute of Fundamental Research, Center for Interdisciplinary Sciences , Hyderabad 500046, India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1090-5199","authenticated-orcid":false,"given":"Jagannath","family":"Mondal","sequence":"additional","affiliation":[{"name":"Tata Institute of Fundamental Research, Center for Interdisciplinary Sciences , Hyderabad 500046, India"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"317","published-online":{"date-parts":[[2023,5,15]]},"reference":[{"key":"2023081109511448900_c1","doi-asserted-by":"publisher","first-page":"667","DOI":"10.1016\/j.drudis.2013.02.007","article-title":"Molecular determinants of drug\u2013receptor binding kinetics","volume":"18","year":"2013","journal-title":"Drug Discovery Today"},{"key":"2023081109511448900_c2","doi-asserted-by":"publisher","first-page":"0148","DOI":"10.1038\/s41570-018-0148","article-title":"Multiscale methods in drug design bridge chemical and biological complexity in the search for cures","volume":"2","year":"2018","journal-title":"Nat. Rev. Chem."},{"key":"2023081109511448900_c3","doi-asserted-by":"publisher","first-page":"e1006180","DOI":"10.1371\/journal.pcbi.1006180","article-title":"Atomic resolution mechanism of ligand binding to a solvent inaccessible cavity in T4 lysozyme","volume":"14","year":"2018","journal-title":"PLoS Comput. Biol."},{"key":"2023081109511448900_c4","doi-asserted-by":"publisher","first-page":"17743","DOI":"10.1021\/jacs.8b10840","article-title":"Mapping the substrate recognition pathway in cytochrome P450","volume":"140","year":"2018","journal-title":"J. Am. Chem. Soc."},{"key":"2023081109511448900_c5","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1016\/j.sbi.2020.06.022","article-title":"Recent progress in molecular simulation methods for drug binding kinetics","volume":"64","year":"2020","journal-title":"Curr. Opin. Struct. Biol."},{"key":"2023081109511448900_c6","doi-asserted-by":"publisher","first-page":"633","DOI":"10.1021\/acs.jpclett.0c02785","article-title":"An appraisal of computer simulation approaches in elucidating biomolecular recognition pathways","volume":"12","year":"2021","journal-title":"J. Phys. Chem. Lett."},{"key":"2023081109511448900_c7","doi-asserted-by":"publisher","first-page":"9181","DOI":"10.1021\/ja202726y","article-title":"How does a drug molecule find its target binding site?","volume":"133","year":"2011","journal-title":"J. Am. Chem. Soc."},{"key":"2023081109511448900_c8","doi-asserted-by":"publisher","first-page":"13118","DOI":"10.1073\/pnas.1104614108","article-title":"Pathway and mechanism of drug binding to G-protein-coupled receptors","volume":"108","year":"2011","journal-title":"Proc. Natl. Acad. Sci. U. S. A."},{"key":"2023081109511448900_c9","first-page":"41","article-title":"Anton 2: Raising the bar for performance and programmability in a special-purpose molecular dynamics supercomputer","year":"2014"},{"key":"2023081109511448900_c10","article-title":"Anton 3: Twenty microseconds of molecular dynamics simulation before lunch","year":"2021"},{"key":"2023081109511448900_c11","doi-asserted-by":"publisher","first-page":"1990","DOI":"10.1002\/jcc.24030","article-title":"Best bang for your buck: GPU nodes for GROMACS biomolecular simulations","volume":"36","year":"2015","journal-title":"J. Comput. Chem."},{"key":"2023081109511448900_c12","doi-asserted-by":"publisher","first-page":"12562","DOI":"10.1073\/pnas.202427399","article-title":"Escaping free-energy minima","volume":"99","year":"2002","journal-title":"Proc. Natl. Acad. Sci. U. S. A."},{"key":"2023081109511448900_c13","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1146\/annurev-physchem-040215-112229","article-title":"Enhancing important fluctuations: Rare events and metadynamics from a conceptual viewpoint","volume":"67","year":"2016","journal-title":"Annu. Rev. Phys. Chem."},{"key":"2023081109511448900_c14","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1016\/j.sbi.2014.04.002","article-title":"Markov state models of biomolecular conformational dynamics","volume":"25","year":"2014","journal-title":"Curr. Opin. Struct. Biol."},{"key":"2023081109511448900_c15","doi-asserted-by":"crossref","DOI":"10.1007\/978-94-007-7606-7","volume-title":"An Introduction to Markov State Models and Their Application to Long Timescale Molecular Simulation","author":"Bowman","year":"2014"},{"key":"2023081109511448900_c16","doi-asserted-by":"publisher","first-page":"094101","DOI":"10.1063\/1.5041073","article-title":"Assessment and optimization of collective variables for protein conformational landscape: GB1 \u03b2-hairpin as a case study","volume":"149","year":"2018","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c17","doi-asserted-by":"publisher","first-page":"074104","DOI":"10.1063\/1.5139584","article-title":"On the role of solvent in hydrophobic cavity-ligand recognition kinetics","volume":"152","year":"2020","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c18","doi-asserted-by":"publisher","first-page":"114106","DOI":"10.1063\/5.0059965","article-title":"A deep autoencoder framework for discovery of metastable ensembles in biomacromolecules","volume":"155","year":"2021","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c19","doi-asserted-by":"publisher","first-page":"244111","DOI":"10.1063\/1.2746330","article-title":"Dihedral angle principal component analysis of molecular dynamics simulations","volume":"126","year":"2007","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c20","doi-asserted-by":"publisher","first-page":"015102","DOI":"10.1063\/1.4811489","article-title":"Identification of slow molecular order parameters for Markov model construction","volume":"139","year":"2013","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c21","doi-asserted-by":"publisher","first-page":"2000","DOI":"10.1021\/ct300878a","article-title":"Improvements in Markov state model construction reveal many non-native interactions in the folding of NTL9","volume":"9","year":"2013","journal-title":"J. Chem. Theory Comput."},{"key":"2023081109511448900_c22","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1038\/s41467-017-02388-1","article-title":"VAMPnets for deep learning of molecular kinetics","volume":"9","year":"2018","journal-title":"Nat. Commun."},{"key":"2023081109511448900_c23","doi-asserted-by":"publisher","first-page":"214114","DOI":"10.1063\/1.5092521","article-title":"Nonlinear discovery of slow molecular modes using state-free reversible VAMPnets","volume":"150","year":"2019","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c24","doi-asserted-by":"publisher","first-page":"7999","DOI":"10.1021\/acs.jpcb.9b05578","article-title":"High-resolution Markov state models for the dynamics of Trp-cage miniprotein constructed over slow folding modes identified by state-free reversible VAMPnets","volume":"123","year":"2019","journal-title":"J. Phys. Chem. B"},{"key":"2023081109511448900_c25","doi-asserted-by":"publisher","first-page":"3354","DOI":"10.1021\/acs.jctc.9b00118","article-title":"Chasing the full free energy landscape of neuroreceptor\/ligand unbinding by metadynamics simulations","volume":"15","year":"2019","journal-title":"J. Chem. Theory Comput."},{"key":"2023081109511448900_c26","doi-asserted-by":"publisher","first-page":"3495","DOI":"10.1021\/acs.jpclett.9b01183","article-title":"Exhaustive search of ligand binding pathways via volume-based metadynamics","volume":"10","year":"2019","journal-title":"J. Phys. Chem. Lett."},{"key":"2023081109511448900_c27","doi-asserted-by":"publisher","first-page":"072301","DOI":"10.1063\/1.5025487","article-title":"Reweighted autoencoded variational Bayes for enhanced sampling (RAVE)","volume":"149","year":"2018","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c28","doi-asserted-by":"publisher","first-page":"1957","DOI":"10.1021\/acs.jctc.1c01163","article-title":"PathDetect-SOM: A neural network approach for the identification of pathways in ligand binding simulations","volume":"18","year":"2022","journal-title":"J. Chem. Theory Comput."},{"key":"2023081109511448900_c29","doi-asserted-by":"publisher","first-page":"595","DOI":"10.1016\/j.jmb.2008.10.086","article-title":"Halogenated benzenes bound within a non-polar cavity in T4 lysozyme provide examples of I\u22efS and I\u22efSe halogen-bonding","volume":"385","year":"2009","journal-title":"J. Mol. Biol."},{"key":"2023081109511448900_c30","doi-asserted-by":"publisher","first-page":"2674","DOI":"10.1021\/bi00224a016","article-title":"Crystal structures of cytochrome P-450CAM complexed with camphane, thiocamphor, and adamantane: Factors controlling P-450 substrate hydroxylation","volume":"30","year":"1991","journal-title":"Biochemistry"},{"key":"2023081109511448900_c31","doi-asserted-by":"publisher","first-page":"687","DOI":"10.1016\/0022-2836(87)90190-2","article-title":"High-resolution crystal structure of cytochrome P450cam","volume":"195","year":"1987","journal-title":"J. Mol. Biol."},{"key":"2023081109511448900_c32","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","year":"2006","journal-title":"Science"},{"key":"2023081109511448900_c33","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1080\/00268978400101201","article-title":"A molecular dynamics method for simulations in the canonical ensemble","volume":"52","year":"1984","journal-title":"Mol. Phys."},{"key":"2023081109511448900_c34","doi-asserted-by":"publisher","first-page":"1695","DOI":"10.1103\/physreva.31.1695","article-title":"Canonical dynamics: Equilibrium phase-space distributions","volume":"31","year":"1985","journal-title":"Phys. Rev. A"},{"key":"2023081109511448900_c35","doi-asserted-by":"publisher","first-page":"7182","DOI":"10.1063\/1.328693","article-title":"Polymorphic transitions in single crystals: A new molecular dynamics method","volume":"52","year":"1981","journal-title":"J. Appl. Phys."},{"key":"2023081109511448900_c36","doi-asserted-by":"publisher","first-page":"3634","DOI":"10.1103\/physrevlett.72.3634","article-title":"Separation of a mixture of independent signals using time delayed correlations","volume":"72","year":"1994","journal-title":"Phys. Rev. Lett."},{"key":"2023081109511448900_c37","unstructured":"See https:\/\/www.tensorflow.org for tensorflow library."},{"key":"2023081109511448900_c38","unstructured":"See https:\/\/github.com\/keras-team\/keras for keras library."},{"key":"2023081109511448900_c39","article-title":"Adam: A method for stochastic optimization","year":"2017"},{"key":"2023081109511448900_c40","doi-asserted-by":"publisher","first-page":"174105","DOI":"10.1063\/1.3565032","volume":"134","year":"2011","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c41","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1007\/s00332-019-09567-y","article-title":"Variational approach for learning Markov processes from time series data","volume":"30","year":"2019","journal-title":"J. Nonlinear Sci."},{"key":"2023081109511448900_c42","doi-asserted-by":"publisher","first-page":"164120","DOI":"10.1063\/1.5083627","article-title":"Identification of kinetic order parameters for non-equilibrium dynamics","volume":"150","year":"2019","journal-title":"J. Chem. Phys."},{"key":"2023081109511448900_c43","doi-asserted-by":"publisher","first-page":"5525","DOI":"10.1021\/acs.jctc.5b00743","article-title":"PyEMMA 2: A software package for estimation, validation, and analysis of Markov models","volume":"11","year":"2015","journal-title":"J. Chem. Theory Comput."},{"key":"2023081109511448900_c44","doi-asserted-by":"publisher","first-page":"19011","DOI":"10.1073\/pnas.0905466106","article-title":"Constructing the equilibrium ensemble of folding pathways from short off-equilibrium simulations","volume":"106","year":"2009","journal-title":"Proc. Natl. Acad. Sci. U. S. A."},{"key":"2023081109511448900_c45","doi-asserted-by":"publisher","first-page":"1058","DOI":"10.1016\/j.bpj.2018.01.014","article-title":"Escape of a small molecule from inside T4 lysozyme by multiple pathways","volume":"114","year":"2018","journal-title":"Biophys. J."},{"key":"2023081109511448900_c46","doi-asserted-by":"publisher","first-page":"802","DOI":"10.1016\/j.bpj.2023.01.036","article-title":"Long time-step molecular dynamics can retard simulation of protein-ligand recognition process","volume":"122","year":"2023","journal-title":"Biophys. J."},{"key":"2023081109511448900_c47","doi-asserted-by":"publisher","first-page":"8564","DOI":"10.1021\/bi00027a006","article-title":"Energetic origins of specificity of ligand binding in an interior nonpolar cavity of T4 lysozyme","volume":"34","year":"1995","journal-title":"Biochemistry"},{"key":"2023081109511448900_c48","doi-asserted-by":"publisher","first-page":"516","DOI":"10.1038\/nsb0696-516","article-title":"Access of ligands to cavities within the core of a protein is rapid","volume":"3","year":"1996","journal-title":"Nat. Struct. Mol. Biol."},{"key":"2023081109511448900_c49","doi-asserted-by":"publisher","first-page":"179","DOI":"10.1111\/j.1469-1809.1936.tb02137.x","article-title":"The use of multiple measurements in taxonomic problems","volume":"7","year":"1936","journal-title":"Ann. Eugen."},{"key":"2023081109511448900_c50","doi-asserted-by":"publisher","first-page":"708","DOI":"10.1021\/acs.jctc.8b00869","article-title":"Toward achieving efficient and accurate ligand-protein unbinding with deep learning and molecular dynamics through RAVE","volume":"15","year":"2019","journal-title":"J. Chem. Theory Comput."},{"key":"2023081109511448900_c51","doi-asserted-by":"publisher","first-page":"2440","DOI":"10.1021\/acs.jctc.7b00182","article-title":"tICA-metadynamics: accelerating metadynamics by using kinetically selected collective variables","volume":"13","year":"2017","journal-title":"J. Chem. Theory Comput."}],"container-title":["The Journal of Chemical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0145197\/17476507\/194103_1_5.0145197.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/aip\/jcp\/article-pdf\/doi\/10.1063\/5.0145197\/17476507\/194103_1_5.0145197.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,11]],"date-time":"2023-08-11T09:52:35Z","timestamp":1691747555000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/jcp\/article\/158\/19\/194103\/2890463\/A-deep-encoder-decoder-framework-for-identifying"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5,15]]},"references-count":51,"journal-issue":{"issue":"19","published-print":{"date-parts":[[2023,5,21]]}},"URL":"https:\/\/doi.org\/10.1063\/5.0145197","relation":{},"ISSN":["0021-9606","1089-7690"],"issn-type":[{"value":"0021-9606","type":"print"},{"value":"1089-7690","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2023,5,21]]},"published":{"date-parts":[[2023,5,15]]},"article-number":"194103"}}