{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T14:07:04Z","timestamp":1790086024486,"version":"4.0.1"},"reference-count":58,"publisher":"American Chemical Society (ACS)","issue":"18","license":[{"start":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T00:00:00Z","timestamp":1788134400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T00:00:00Z","timestamp":1788134400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T00:00:00Z","timestamp":1788134400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100001665","name":"Agence Nationale de la Recherche","doi-asserted-by":"publisher","award":["ANR-22-CPJ1-0093-01"],"award-info":[{"award-number":["ANR-22-CPJ1-0093-01"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001665","name":"Agence Nationale de la Recherche","doi-asserted-by":"publisher","award":["ANR-24-CE29-5745"],"award-info":[{"award-number":["ANR-24-CE29-5745"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004794","name":"Centre National de la Recherche Scientifique","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100004794","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,22]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>This work presents thermoMLIP, a machine-learning-interatomic-potential-based workflow for efficient conformer optimization and thermochemistry prediction across the organic chemical space. By combining MLIP-driven geometry optimization with robust thermochemical corrections, thermoMLIP achieves near-DFT accuracy at a fraction of the computational cost. We further show that MLIP-based conformer searches systematically identify lower-energy structures than conventional approaches, improving thermochemical predictions across diverse data sets. Although thermoMLIP substantially accelerates thermochemistry calculations, its computational cost remains prohibitive for the exhaustive screening of very large chemical spaces. We therefore employ an uncertainty-aware active-learning strategy based on evidential graph neural networks to efficiently generate a large, uniform thermodynamic data set, which is subsequently used to train thermoGNN, a surrogate capable of near-instantaneous prediction of molecular thermodynamic properties. Together, thermoMLIP and thermoGNN establish a hierarchical framework that bridges accurate MLIP-based thermochemistry and ultralarge-scale molecular screening, enabling applications such as reaction-network exploration and high-throughput reaction discovery.<\/jats:p>","DOI":"10.1021\/acs.jctc.6c00936","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T04:50:44Z","timestamp":1788411044000},"page":"9526-9537","source":"Crossref","is-referenced-by-count":0,"title":["MLIP-Enhanced Thermochemistry\nPredictions across Organic\nChemical Space"],"prefix":"10.1021","volume":"22","author":[{"given":"Bowen","family":"Deng","sequence":"first","affiliation":[{"name":"Ecole Nationale Supe\u0301rieure de Chimie de Paris, Universite\u0301 PSL, CNRS, Institute of Chemistry for Life and Health Sciences , ,","place":["Paris, France, 75 005"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8322-0572","authenticated-orcid":true,"given":"Thijs","family":"Stuyver","sequence":"additional","affiliation":[{"name":"Ecole Nationale Supe\u0301rieure de Chimie de Paris, Universite\u0301 PSL, CNRS, Institute of Chemistry for Life and Health Sciences , ,","place":["Paris, France, 75 005"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,31]]},"reference":[{"key":"2026092209071963500_cit1","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1038\/s41586-018-0337-2","article-title":"Machine learning\nfor molecular and materials science","volume":"559","author":"Butler","year":"2018","journal-title":"Nature"},{"key":"2026092209071963500_cit2","article-title":"Guest editorial:\nSpecial topic on data-enabled theoretical chemistry","author":"Rupp","year":"2018","journal-title":"arXiv"},{"key":"2026092209071963500_cit3","doi-asserted-by":"publisher","first-page":"5826","DOI":"10.1021\/acs.jpca.9b04195","article-title":"Accurate thermochemistry with small\ndata sets: A bond additivity correction and transfer learning approach","volume":"123","author":"Grambow","year":"2019","journal-title":"J. Phys. Chem. A"},{"key":"2026092209071963500_cit4","doi-asserted-by":"publisher","first-page":"649","DOI":"10.1021\/acs.jcim.4c02014","article-title":"Graph-Based Deep Learning Models\nfor Thermodynamic\nProperty Prediction: The Interplay between Target Definition, Data\nDistribution, Featurization, and Model Architecture","volume":"65","author":"Deng","year":"2025","journal-title":"J. Chem. Inf. Model."},{"key":"2026092209071963500_cit5","doi-asserted-by":"publisher","first-page":"146401","DOI":"10.1103\/PhysRevLett.98.146401","article-title":"Generalized neural-network representation\nof high-dimensional\npotential-energy surfaces","volume":"98","author":"Behler","year":"2007","journal-title":"Phys. Rev. Lett."},{"key":"2026092209071963500_cit6","doi-asserted-by":"publisher","first-page":"1902765","DOI":"10.1002\/adma.201902765","article-title":"Machine learning\ninteratomic potentials\nas emerging tools for materials science","volume":"31","author":"Deringer","year":"2019","journal-title":"Adv.\nMater."},{"key":"2026092209071963500_cit7","article-title":"SchNet: A continuous-filter convolutional neural network for modeling\nquantum interactions","author":"Schu\u0308tt","year":"2017","journal-title":"arXiv"},{"key":"2026092209071963500_cit8","doi-asserted-by":"publisher","first-page":"2453","DOI":"10.1038\/s41467-022-29939-5","article-title":"E(3)-equivariant\ngraph neural networks for data-efficient and accurate interatomic\npotentials","volume":"13","author":"Batzner","year":"2022","journal-title":"Nat. Commun."},{"key":"2026092209071963500_cit9","article-title":"Learning Smooth\nand Expressive Interatomic Potentials for Physical Property Prediction","author":"Fu","year":"2025","journal-title":"arXiv"},{"key":"2026092209071963500_cit10","article-title":"Equiformer: Equivariant\nGraph Attention Transformer for 3D Atomistic Graphs","author":"Liao","year":"2023","journal-title":"arXiv"},{"key":"2026092209071963500_cit11","article-title":"MACE: Higher Order\nEquivariant Message Passing Neural Networks for Fast and Accurate\nForce Fields","author":"Batatia","year":"2022","journal-title":"arXiv"},{"key":"2026092209071963500_cit12","doi-asserted-by":"publisher","first-page":"3192","DOI":"10.1039\/C6SC05720A","article-title":"ANI-1:\nan extensible neural network\npotential with DFT accuracy at force field computational cost","volume":"8","author":"Smith","year":"2017","journal-title":"Chem. Sci."},{"key":"2026092209071963500_cit13","doi-asserted-by":"publisher","first-page":"17598","DOI":"10.1021\/jacs.4c07099","article-title":"MACE-OFF: Short-Range Transferable\nMachine Learning Force Fields for Organic Molecules","volume":"147","author":"Kova\u0301cs","year":"2025","journal-title":"J. Am. Chem. Soc."},{"key":"2026092209071963500_cit14","article-title":"Uma: A family\nof universal models for atoms","author":"Wood","year":"2025","journal-title":"arXiv"},{"key":"2026092209071963500_cit15","article-title":"The Open Molecules 2025 (OMol25) Dataset, Evaluations,\nand Models","author":"Levine","year":"2025","journal-title":"arXiv"},{"key":"2026092209071963500_cit16","doi-asserted-by":"publisher","first-page":"273002","DOI":"10.1088\/1361-648X\/aa680e","article-title":"The atomic simulation environment\u2500a Python library\nfor working with atoms","volume":"29","author":"Hjorth Larsen","year":"2017","journal-title":"J. Phys.: Condens. Matter"},{"key":"2026092209071963500_cit17","doi-asserted-by":"publisher","first-page":"291","DOI":"10.12688\/f1000research.22758.1","article-title":"GoodVibes: Automated\nthermochemistry for heterogeneous computational chemistry data","volume":"9","author":"Luchini","year":"2020","journal-title":"F1000research"},{"key":"2026092209071963500_cit18","doi-asserted-by":"publisher","first-page":"140022","DOI":"10.1038\/sdata.2014.22","article-title":"Quantum\nchemistry structures and properties of 134 kilo molecules","volume":"1","author":"Ramakrishnan","year":"2014","journal-title":"Sci. Data"},{"key":"2026092209071963500_cit19","doi-asserted-by":"publisher","first-page":"1376","DOI":"10.1038\/s41597-024-04206-y","article-title":"QM40, Realistic\nQuantum Mechanical\nDataset for Machine Learning in Molecular Science","volume":"11","author":"Madushanka","year":"2024","journal-title":"Sci. Data"},{"key":"2026092209071963500_cit20","doi-asserted-by":"publisher","first-page":"2328","DOI":"10.1038\/s41467-020-16201-z","article-title":"Prediction of organic homolytic bond dissociation enthalpies\nat near chemical accuracy with sub-second computational cost","volume":"11","author":"St. John","year":"2020","journal-title":"Nat. Commun."},{"key":"2026092209071963500_cit21","doi-asserted-by":"publisher","first-page":"1900","DOI":"10.1039\/D3DD00169E","article-title":"Expansion of bond\ndissociation prediction with machine learning to medicinally and environmentally\nrelevant chemical space","volume":"2","author":"Sv","year":"2023","journal-title":"Digital Discovery"},{"key":"2026092209071963500_cit22","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1038\/s41597-022-01390-7","article-title":"QMugs, quantum\nmechanical properties of drug-like molecules","volume":"9","author":"Isert","year":"2022","journal-title":"Sci. Data"},{"key":"2026092209071963500_cit23","doi-asserted-by":"publisher","first-page":"101040","DOI":"10.1016\/j.cdc.2023.101040","article-title":"QMugs 1.1: Quantum mechanical properties\nof organic\ncompounds commonly encountered in reactivity datasets","volume":"46","author":"Neeser","year":"2023","journal-title":"Chem. Data Collect."},{"key":"2026092209071963500_cit24","doi-asserted-by":"publisher","first-page":"6615","DOI":"10.1039\/b810189b","article-title":"Long-range\ncorrected hybrid density functionals with\ndamped atom\u2013atom dispersion corrections","volume":"10","author":"Chai","year":"2008","journal-title":"Phys. Chem. Chem. Phys."},{"key":"2026092209071963500_cit25","unstructured":"RDKit contributors.\n          \n          RDKit: Open-source cheminformatics. 2026; https:\/\/www.rdkit.org."},{"key":"2026092209071963500_cit26","doi-asserted-by":"publisher","first-page":"4266","DOI":"10.1002\/anie.202011941","article-title":"autodE: Automated Calculation\nof Reaction Energy Profiles\u2500Application to Organic and Organometallic\nReactions","volume":"60","author":"Young","year":"2021","journal-title":"Angew. Chem., Int. Ed."},{"key":"2026092209071963500_cit27","doi-asserted-by":"publisher","first-page":"2562","DOI":"10.1021\/acs.jcim.5b00654","article-title":"Better Informed\nDistance Geometry: Using What We Know\nTo Improve Conformation Generation","volume":"55","author":"Riniker","year":"2015","journal-title":"J. Chem.\nInf. Model."},{"key":"2026092209071963500_cit28","doi-asserted-by":"publisher","first-page":"9955","DOI":"10.1002\/chem.201200497","article-title":"Supramolecular\nbinding thermodynamics by dispersion-corrected density functional\ntheory","volume":"18","author":"Grimme","year":"2012","journal-title":"Chem. - Eur. J."},{"key":"2026092209071963500_cit29","doi-asserted-by":"publisher","first-page":"204109","DOI":"10.1063\/5.0152838","article-title":"Analytical\nharmonic vibrational frequencies with VV10-containing density functionals:\nTheory, efficient implementation, and benchmark assessments","volume":"158","author":"Liang","year":"2023","journal-title":"J. Chem. Phys."},{"key":"2026092209071963500_cit30","doi-asserted-by":"publisher","first-page":"214110","DOI":"10.1063\/1.4952647","article-title":"\u03c9B97M-V: A combinatorially optimized,\nrange-separated hybrid, meta-GGA density functional with VV10 nonlocal\ncorrelation","volume":"144","author":"Mardirossian","year":"2016","journal-title":"J. Chem. Phys."},{"key":"2026092209071963500_cit31","doi-asserted-by":"publisher","first-page":"134105","DOI":"10.1063\/1.3484283","article-title":"Property-optimized\nGaussian basis sets for molecular\nresponse calculations","volume":"133","author":"Rappoport","year":"2010","journal-title":"J. Chem. Phys."},{"key":"2026092209071963500_cit32","article-title":"Deep Evidential Regression","author":"Amini","year":"2020","journal-title":"arXiv"},{"key":"2026092209071963500_cit33","doi-asserted-by":"publisher","first-page":"1356","DOI":"10.1021\/acscentsci.1c00546","article-title":"Evidential Deep Learning for Guided Molecular Property\nPrediction and Discovery","volume":"7","author":"Soleimany","year":"2021","journal-title":"ACS Cent. Sci."},{"key":"2026092209071963500_cit34","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1021\/acs.jcim.3c01250","article-title":"Chemprop:\nA Machine Learning Package for Chemical Property Prediction","volume":"64","author":"Heid","year":"2024","journal-title":"J. Chem. Inf. Model."},{"key":"2026092209071963500_cit35","doi-asserted-by":"publisher","first-page":"28","DOI":"10.1021\/acs.jcim.5c02332","article-title":"Chemprop v2: An Efficient, Modular Machine Learning\nPackage for Chemical Property Prediction","volume":"66","author":"Graff","year":"2026","journal-title":"J.\nChem. Inf. Model."},{"key":"2026092209071963500_cit36","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1186\/s13321-023-00682-3","article-title":"Explainable uncertainty\nquantifications for deep learning-based molecular\nproperty prediction","volume":"15","author":"Yang","year":"2023","journal-title":"J. Cheminform."},{"key":"2026092209071963500_cit37","doi-asserted-by":"publisher","first-page":"7548","DOI":"10.1021\/acs.jpca.2c04848","article-title":"Deep learning-based\nincrement theory for formation enthalpy predictions","volume":"126","author":"Chen","year":"2022","journal-title":"J. Phys. Chem. A"},{"key":"2026092209071963500_cit38","doi-asserted-by":"publisher","first-page":"1059","DOI":"10.1021\/je000236i","article-title":"The NIST Chemistry WebBook: A chemical\ndata resource\non the internet","volume":"46","author":"Linstrom","year":"2001","journal-title":"J. Chem. Eng. Data"},{"key":"2026092209071963500_cit39","doi-asserted-by":"publisher","first-page":"919","DOI":"10.1039\/D4DD00043A","article-title":"Repurposing\nquantum chemical descriptor\ndatasets for on-the-fly generation of informative reaction representations:\nApplication to hydrogen atom transfer reactions","volume":"3","author":"Alfonso-Ramos","year":"2024","journal-title":"Digital Discovery"},{"key":"2026092209071963500_cit40","doi-asserted-by":"publisher","first-page":"3400","DOI":"10.1039\/D5DD00340G","article-title":"Screening Diels\u2013Alder\nreaction space to identify candidate reactions for self-healing polymer\napplications","volume":"4","author":"Ferrer","year":"2025","journal-title":"Digital Discovery"},{"key":"2026092209071963500_cit41","doi-asserted-by":"publisher","first-page":"121","DOI":"10.1146\/annurev-physchem-071119-040123","article-title":"The exploration of\nchemical reaction networks","volume":"71","author":"Unsleber","year":"2020","journal-title":"Annu. Rev. Phys.\nChem."},{"key":"2026092209071963500_cit42","doi-asserted-by":"publisher","first-page":"718","DOI":"10.1002\/aic.10655","article-title":"Automatic reaction network generation\nusing RMG for steam cracking of n-hexane","volume":"52","author":"Van Geem","year":"2006","journal-title":"AIChE\nJ."},{"key":"2026092209071963500_cit43","doi-asserted-by":"publisher","first-page":"7447","DOI":"10.1039\/D3SC01202F","article-title":"First principles\nreaction discovery: from the Schrodinger equation to experimental\nprediction for methane pyrolysis","volume":"14","author":"Xu","year":"2023","journal-title":"Chem. Sci."},{"key":"2026092209071963500_cit44","doi-asserted-by":"publisher","first-page":"1044","DOI":"10.1038\/nchem.2099","article-title":"Discovering chemistry with an ab\ninitio nanoreactor","volume":"6","author":"Wang","year":"2014","journal-title":"Nat. Chem."},{"key":"2026092209071963500_cit45","doi-asserted-by":"publisher","first-page":"226","DOI":"10.1039\/D2SC05089G","article-title":"Predictive\nchemistry: machine learning\nfor reaction deployment, reaction development, and reaction discovery","volume":"14","author":"Tu","year":"2023","journal-title":"Chem. Sci."},{"key":"2026092209071963500_cit46","doi-asserted-by":"publisher","first-page":"2686","DOI":"10.1021\/acs.jcim.0c01480","article-title":"Reaction mechanism generator v3. 0: advances\nin automatic mechanism generation","volume":"61","author":"Liu","year":"2021","journal-title":"J. Chem.\nInf. Model."},{"key":"2026092209071963500_cit47","doi-asserted-by":"publisher","DOI":"10.1002\/jcc.70462","article-title":"Reaction Enumeration\nBased on NBO-Informed Molecular\nGraphs","volume":"47","author":"Alfonso-Ramos","year":"2026","journal-title":"J. Comput. Chem."},{"key":"2026092209071963500_cit48","doi-asserted-by":"publisher","first-page":"2199","DOI":"10.1021\/acs.jcim.0c00092","article-title":"Self-consistent component increment theory for predicting\nenthalpy of formation","volume":"60","author":"Zhao","year":"2020","journal-title":"J. Chem. Inf. Model."},{"key":"2026092209071963500_cit49","doi-asserted-by":"publisher","first-page":"10362","DOI":"10.1021\/acs.jctc.5c01161","article-title":"Anticipating the\nselectivity of intramolecular cyclization reaction pathways with neural\nnetwork potentials","volume":"21","author":"Casetti","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026092209071963500_cit50","doi-asserted-by":"publisher","DOI":"10.1002\/chem.202301957","article-title":"Combining molecular\nquantum mechanical modeling and machine learning for accelerated reaction\nscreening and discovery","volume":"29","author":"Casetti","year":"2023","journal-title":"Chem. - Eur. J."},{"key":"2026092209071963500_cit51","doi-asserted-by":"publisher","first-page":"064105","DOI":"10.1063\/5.0059742","article-title":"Toward\nthe design of chemical reactions:\nMachine learning barriers of competing mechanisms in reactant space","volume":"155","author":"Heinen","year":"2021","journal-title":"J. Chem. Phys."},{"key":"2026092209071963500_cit52","doi-asserted-by":"publisher","DOI":"10.1002\/chem.202300387","article-title":"Machine Learning-Guided\nComputational Screening of\nNew Candidate Reactions with High Bioorthogonal Click Potential","volume":"29","author":"Stuyver","year":"2023","journal-title":"Chem. - Eur. J."},{"key":"2026092209071963500_cit53","doi-asserted-by":"publisher","first-page":"66","DOI":"10.1038\/s41597-023-01977-8","article-title":"Reaction profiles for quantum chemistry-computed\n[3 + 2] cycloaddition reactions","volume":"10","author":"Stuyver","year":"2023","journal-title":"Sci. Data"},{"key":"2026092209071963500_cit54","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1098\/rspa.1936.0060","article-title":"The theory\nof reactions involving proton transfers","volume":"154","author":"Bell","year":"1936","journal-title":"Proc.\nR. Soc. London, Ser. A"},{"key":"2026092209071963500_cit55","doi-asserted-by":"publisher","first-page":"1333","DOI":"10.1039\/tf9363201333","article-title":"Further considerations on the thermodynamics of chemical\nequilibria and reaction rates","volume":"32","author":"Evans","year":"1936","journal-title":"Trans. Faraday\nSoc."},{"key":"2026092209071963500_cit56","doi-asserted-by":"publisher","first-page":"4906","DOI":"10.1021\/acs.jcim.2c00965","article-title":"The RMG Database for Chemical Property Prediction","volume":"62","author":"Johnson","year":"2022","journal-title":"J. Chem. Inf. Model."},{"key":"2026092209071963500_cit57","article-title":"Learning 3D Representations\nof Molecular Chirality with Invariance to Bond Rotations","author":"Adams","year":"2022","journal-title":"arXiv"},{"key":"2026092209071963500_cit58","doi-asserted-by":"publisher","first-page":"045005","DOI":"10.1088\/2632-2153\/ac8f1a","article-title":"Physics-based\nrepresentations for machine learning properties of chemical reactions","volume":"3","author":"van Gerwen","year":"2022","journal-title":"Mach. Learn.: Sci. Technol."}],"container-title":["Journal of Chemical Theory\nand Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jctc.6c00936","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jctcce\/article-pdf\/22\/18\/9526\/68020628\/acs.jctc.6c00936.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jctcce\/article-pdf\/22\/18\/9526\/68020628\/acs.jctc.6c00936.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T13:50:40Z","timestamp":1790085040000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jctcce\/article\/22\/18\/9526\/5387570\/MLIP-Enhanced-Thermochemistry-Predictions-across"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,31]]},"references-count":58,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,8,31]]},"published-print":{"date-parts":[[2026,9,22]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jctc.6c00936","relation":{"has-preprint":[{"id-type":"doi","id":"10.26434\/chemrxiv.15003003\/v3","asserted-by":"object"},{"id-type":"doi","id":"10.26434\/chemrxiv.15003003\/v2","asserted-by":"object"}]},"ISSN":["1549-9618","1549-9626"],"issn-type":[{"value":"1549-9618","type":"print"},{"value":"1549-9626","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,31]]}}}