{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T23:16:59Z","timestamp":1788218219571,"version":"build-2803163510"},"reference-count":101,"publisher":"American Chemical Society (ACS)","issue":"16","license":[{"start":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T00:00:00Z","timestamp":1786320000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T00:00:00Z","timestamp":1786320000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T00:00:00Z","timestamp":1786320000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/na","name":"the Innovation Program for Quantum Science and Technology","doi-asserted-by":"publisher","award":["2021ZD0303305"],"award-info":[{"award-number":["2021ZD0303305"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["22073040"],"award-info":[{"award-number":["22073040"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["22233003"],"award-info":[{"award-number":["22233003"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["22241302"],"award-info":[{"award-number":["22241302"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["NA"],"award-info":[{"award-number":["NA"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,25]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>To accelerate computationally expensive quantum mechanical (QM) calculations, integrating machine learning potentials (MLPs) with molecular mechanics (MM) has become a promising approach, which can be employed to perform multiscale ML\/MM-MD simulations and estimate reaction free energy for enzymatic catalysis. However, the ML\/MM frameworks face the challenge of intuitive chemical interpretability in QM\/MM electrostatic coupling, which limits their transferability in unlearned MM environments. Therefore, their practical application in enzyme engineering is hindered by the bottleneck of MLP retraining for different enzyme environments. In this work, guided by linear response theory, the MLPs trained in the gas phase were naturally adapted to QM\/MM systems by integrating an MM atom electrostatic field-induced recursively embedded atom neural network (MIREANN) into multiscale coupling. The performance and broad applicability of our model are exhibited by nanosecond-scale ML\/MM-MD on diverse enzymatic systems (e.g., cyclooxygenase, hydrolases), which achieves QM\/MM accuracy at a speed close to molecular mechanics. Since MIREANN provides a physics-based framework for the relationship between the electrostatic energy and the electrostatic field generated by the MM environment, it exhibits powerful transferability in reproducing the free energy barriers predicted by QM\/MM-MD with errors less than 0.5 kcal mol\u20131 for various enzyme variants without the need to retrain the MLP. Our model is expected to be applied to a wider range of chemical and biological systems, such as enzyme or drug design, to promote the in-depth development of related research.<\/jats:p>","DOI":"10.1021\/acs.jctc.6c00900","type":"journal-article","created":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T12:02:14Z","timestamp":1786363334000},"page":"8131-8144","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing Machine-Learned\nElectrostatic Models with\nLinear Response Theory for Multiscale Simulations in Enzymes"],"prefix":"10.1021","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-2672-7812","authenticated-orcid":true,"given":"Xinhu","family":"Sha","sequence":"first","affiliation":[{"name":"Nanjing University , , ,","place":["Nanjing, China, 210023"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuehui","family":"Guo","sequence":"additional","affiliation":[{"name":"Nanjing University , , ,","place":["Nanjing, China, 210023"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziyu","family":"Chen","sequence":"additional","affiliation":[{"name":"Nanjing University , , ,","place":["Nanjing, China, 210023"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7185-7085","authenticated-orcid":true,"given":"Daiqian","family":"Xie","sequence":"additional","affiliation":[{"name":"Nanjing University , , ,","place":["Nanjing, China, 210023"]},{"name":"Hefei National Laboratory , ,","place":["Hefei, China, 230088"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2097-9186","authenticated-orcid":true,"given":"Yanzi","family":"Zhou","sequence":"additional","affiliation":[{"name":"Nanjing University , , ,","place":["Nanjing, China, 210023"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,10]]},"reference":[{"key":"2026083118580777400_cit1","doi-asserted-by":"publisher","DOI":"10.1002\/anie.202205735","article-title":"Best-practice DFT protocols\nfor basic molecular computational chemistry","volume":"61","author":"Bursch","year":"2022","journal-title":"Angew. Chem. Int. Ed."},{"key":"2026083118580777400_cit2","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1038\/s43588-021-00060-9","article-title":"Biomolecular modeling thrives in the age of technology","volume":"1","author":"Schlick","year":"2021","journal-title":"Nat. Comput. Sci."},{"key":"2026083118580777400_cit3","doi-asserted-by":"publisher","first-page":"1813","DOI":"10.1126\/science.1096361","article-title":"The many\nroles of computation in drug discovery","volume":"303","author":"Jorgensen","year":"2004","journal-title":"Science"},{"key":"2026083118580777400_cit4","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1016\/0022-2836(76)90311-9","article-title":"Theoretical studies of enzymic reactions:\nDielectric,\nelectrostatic and steric stabilization of the carbonium ion in the\nreaction of lysozyme","volume":"103","author":"Warshel","year":"1976","journal-title":"J. Mol. Biol."},{"key":"2026083118580777400_cit5","doi-asserted-by":"publisher","first-page":"1198","DOI":"10.1002\/anie.200802019","article-title":"QM\/MM methods for biomolecular\nsystems","volume":"48","author":"Senn","year":"2009","journal-title":"Angew.\nChem. Int. Ed."},{"key":"2026083118580777400_cit6","doi-asserted-by":"publisher","first-page":"142","DOI":"10.1021\/ar900171c","article-title":"Advances\nin quantum and molecular mechanical (QM\/MM)\nsimulations for organic and enzymatic reactions","volume":"43","author":"Acevedo","year":"2010","journal-title":"Acc. Chem. Res."},{"key":"2026083118580777400_cit7","doi-asserted-by":"publisher","first-page":"2660","DOI":"10.3390\/molecules27092660","article-title":"Review on\nthe QM\/MM methodologies\nand their application to metalloproteins","volume":"27","author":"Tzeliou","year":"2022","journal-title":"Molecules"},{"key":"2026083118580777400_cit8","doi-asserted-by":"publisher","first-page":"6217","DOI":"10.1021\/cr500628b","article-title":"Mixed quantum mechanical\/molecular\nmechanical molecular\ndynamics simulations of biological systems in ground and electronically\nexcited states","volume":"115","author":"Brunk","year":"2015","journal-title":"Chem. Rev."},{"key":"2026083118580777400_cit9","doi-asserted-by":"publisher","first-page":"700","DOI":"10.1002\/jcc.540110605","article-title":"A combined quantum mechanical and\nmolecular mechanical potential for molecular dynamics simulations","volume":"11","author":"Field","year":"1990","journal-title":"J. Comput. Chem."},{"key":"2026083118580777400_cit10","doi-asserted-by":"publisher","first-page":"573","DOI":"10.1146\/annurev.physchem.59.032607.093618","article-title":"Free energies of\nchemical reactions in solution and\nin enzymes with ab initio quantum mechanics\/molecular mechanics methods","volume":"59","author":"Hu","year":"2008","journal-title":"Annu. Rev. Phys. Chem."},{"key":"2026083118580777400_cit11","doi-asserted-by":"publisher","first-page":"3806","DOI":"10.1021\/ja075896n","article-title":"How do SET-domain\nprotein lysine\nmethyltransferases achieve the methylation state specificity? Revisited\nby ab initio QM\/MM molecular dynamics simulations","volume":"130","author":"Hu","year":"2008","journal-title":"J. Am. Chem. Soc."},{"key":"2026083118580777400_cit12","doi-asserted-by":"publisher","first-page":"6636","DOI":"10.1021\/acs.chemrev.6b00776","article-title":"Multiscale studies\non ionic liquids","volume":"117","author":"Dong","year":"2017","journal-title":"Chem. Rev."},{"key":"2026083118580777400_cit13","doi-asserted-by":"publisher","first-page":"5131","DOI":"10.1021\/acs.jpcb.9b01989","article-title":"Computational insights\ninto Endo\/Exo\nselectivity of the Diels\u2013Alder  reaction in explicit solvent\nat ab initio quantum mechanical\/molecular mechanical level","volume":"123","author":"Li","year":"2019","journal-title":"J. Phys. Chem. B"},{"key":"2026083118580777400_cit14","doi-asserted-by":"publisher","first-page":"4934","DOI":"10.1021\/acs.jctc.6b00663","article-title":"Multiscale quantum mechanics\/molecular mechanics simulations\nwith neural networks","volume":"12","author":"Shen","year":"2016","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit15","doi-asserted-by":"publisher","first-page":"47044","DOI":"10.1021\/jacs.5c12428","article-title":"General framework\nfor geometric deep learning on tensorial properties of molecules and\ncrystals","volume":"147","author":"Yan","year":"2025","journal-title":"J. Am. Chem. Soc."},{"key":"2026083118580777400_cit16","doi-asserted-by":"publisher","first-page":"040301","DOI":"10.1103\/PhysRevMaterials.6.040301","article-title":"Deep dive into machine\nlearning density functional theory for materials science and chemistry","volume":"6","author":"Fiedler","year":"2022","journal-title":"Phys. Rev. Mater."},{"key":"2026083118580777400_cit17","doi-asserted-by":"publisher","first-page":"872","DOI":"10.1038\/s41467-017-00839-3","article-title":"Bypassing\nthe Kohn-Sham equations\nwith machine learning","volume":"8","author":"Brockherde","year":"2017","journal-title":"Nat. Commun."},{"key":"2026083118580777400_cit18","doi-asserted-by":"publisher","first-page":"241722","DOI":"10.1063\/1.5019779","article-title":"SchNet \u2013 A deep learning architecture\nfor molecules\nand materials","volume":"148","author":"Schu\u0308tt","year":"2018","journal-title":"J. Chem. Phys."},{"key":"2026083118580777400_cit19","doi-asserted-by":"publisher","first-page":"9816","DOI":"10.1021\/acs.chemrev.1c00107","article-title":"Combining machine learning and computational\nchemistry for predictive insights into chemical systems","volume":"121","author":"Keith","year":"2021","journal-title":"Chem. Rev."},{"key":"2026083118580777400_cit20","doi-asserted-by":"publisher","first-page":"6114","DOI":"10.1038\/s41467-024-50418-6","article-title":"Modelling chemical processes in explicit\nsolvents with machine learning potentials","volume":"15","author":"Zhang","year":"2024","journal-title":"Nat.\nCommun."},{"key":"2026083118580777400_cit21","doi-asserted-by":"publisher","first-page":"1019","DOI":"10.1038\/s41586-024-08127-z","article-title":"Ab initio\ncharacterization of protein molecular dynamics with AI2BMD","volume":"635","author":"Wang","year":"2024","journal-title":"Nature"},{"key":"2026083118580777400_cit22","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.adn4397","article-title":"Biomolecular dynamics with machine-learned quantum-mechanical\nforce fields trained on diverse chemical fragments","volume":"10","author":"Unke","year":"2024","journal-title":"Sci. Adv."},{"key":"2026083118580777400_cit23","doi-asserted-by":"publisher","first-page":"15679","DOI":"10.1038\/ncomms15679","article-title":"Universal fragment descriptors for predicting properties\nof inorganic crystals","volume":"8","author":"Isayev","year":"2017","journal-title":"Nat. Commun."},{"key":"2026083118580777400_cit24","doi-asserted-by":"publisher","first-page":"3564","DOI":"10.1021\/acs.chemmater.9b01294","article-title":"Graph networks as\na universal machine learning framework for molecules and crystals","volume":"31","author":"Chen","year":"2019","journal-title":"Chem. Mater."},{"key":"2026083118580777400_cit25","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1038\/s41524-022-00734-6","article-title":"Recent advances and applications\nof deep learning methods\nin materials science","volume":"8","author":"Choudhary","year":"2022","journal-title":"Npj Comput. Mater."},{"key":"2026083118580777400_cit26","doi-asserted-by":"publisher","first-page":"727","DOI":"10.1038\/s41557-023-01427-3","article-title":"Exploring the frontiers\nof condensed-phase\nchemistry with a general reactive machine learning potential","volume":"16","author":"Zhang","year":"2024","journal-title":"Nat. Chem."},{"key":"2026083118580777400_cit27","doi-asserted-by":"publisher","first-page":"579","DOI":"10.1038\/s41467-023-36329-y","article-title":"Learning local equivariant\nrepresentations\nfor large-scale atomistic dynamics","volume":"14","author":"Musaelian","year":"2023","journal-title":"Nat. Commun."},{"key":"2026083118580777400_cit28","doi-asserted-by":"publisher","first-page":"10580","DOI":"10.1021\/jp9536514","article-title":"Hybrid models for combined quantum\nmechanical and molecular mechanical\napproaches","volume":"100","author":"Bakowies","year":"1996","journal-title":"J. Phys. Chem."},{"key":"2026083118580777400_cit29","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1007\/128_2006_084","article-title":"QM\/MM methods\nfor biological systems","volume-title":"Atomistic Approaches\nin Modern Biology: from Quantum Chemistry to Molecular Simulations","author":"Senn","year":"2007"},{"key":"2026083118580777400_cit30","doi-asserted-by":"publisher","first-page":"185","DOI":"10.1007\/s00214-006-0143-z","article-title":"QM\/MM: What have we learned, where are we, and where\ndo we go from here?","volume":"117","author":"Lin","year":"2007","journal-title":"Theor. Chem. Acc."},{"key":"2026083118580777400_cit31","doi-asserted-by":"publisher","first-page":"2501","DOI":"10.1021\/acs.jctc.4c01449","article-title":"Multiscale force field model based\non a graph neural network for complex chemical systems","volume":"21","author":"Xie","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit32","doi-asserted-by":"publisher","first-page":"2641","DOI":"10.1021\/acs.jctc.0c01112","article-title":"Machine learning in QM\/MM molecular\ndynamics simulations of condensed-phase systems","volume":"17","author":"Bo\u0308selt","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit33","doi-asserted-by":"publisher","first-page":"11473","DOI":"10.1039\/D1SC02742E","article-title":"Machine learning of solvent effects\non molecular spectra and reactions","volume":"12","author":"Gastegger","year":"2021","journal-title":"Chem. Sci."},{"key":"2026083118580777400_cit34","doi-asserted-by":"publisher","first-page":"5588","DOI":"10.1021\/acs.jctc.5c00372","article-title":"NepoIP\/MM: Toward accurate biomolecular simulation with a machine\nlearning\/molecular mechanics model incorporating polarization effects","volume":"21","author":"Song","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit35","doi-asserted-by":"publisher","first-page":"214109","DOI":"10.1063\/5.0205123","article-title":"Learning QM\/MM potential using equivariant\nmultiscale model","volume":"160","author":"Lei","year":"2024","journal-title":"J. Chem. Phys."},{"key":"2026083118580777400_cit36","doi-asserted-by":"publisher","first-page":"3812","DOI":"10.1021\/acs.jpclett.2c00654","article-title":"BuRNN: Buffer region neural network approach for polarizable-embedding\nneural network\/molecular mechanics simulations","volume":"13","author":"Lier","year":"2022","journal-title":"J. Phys. Chem. Lett."},{"key":"2026083118580777400_cit37","doi-asserted-by":"publisher","first-page":"5745","DOI":"10.1021\/acs.jctc.1c00565","article-title":"Machine-learning-assisted free energy\nsimulation of solution-phase\nand enzyme reactions","volume":"17","author":"Pan","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit38","doi-asserted-by":"publisher","first-page":"817","DOI":"10.1021\/acs.jctc.4c01487","article-title":"Machine learning\nquantum mechanical\/molecular mechanical potentials: Evaluating transferability\nin dihydrofolate reductase-catalyzed reactions","volume":"21","author":"Arattu Thodika","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit39","doi-asserted-by":"publisher","first-page":"6993","DOI":"10.1021\/acs.jctc.1c00201","article-title":"Development of range-corrected\ndeep learning potentials for fast, accurate quantum mechanical\/molecular\nmechanical simulations of chemical reactions in solution","volume":"17","author":"Zeng","year":"2021","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit40","doi-asserted-by":"publisher","first-page":"5477","DOI":"10.1021\/acs.jpcb.5c02006","article-title":"Transferability of MACE graph neural\nnetwork for range corrected \u0394-machine learning potential QM\/MM\napplications","volume":"129","author":"Giese","year":"2025","journal-title":"J. Phys. Chem. B"},{"key":"2026083118580777400_cit41","doi-asserted-by":"publisher","first-page":"024115","DOI":"10.1063\/5.0038120","article-title":"A simplified\ncharge projection scheme for long-range electrostatics in ab initio\nQM\/MM calculations","volume":"154","author":"Pan","year":"2021","journal-title":"J. Chem. Phys."},{"key":"2026083118580777400_cit42","doi-asserted-by":"publisher","first-page":"2695","DOI":"10.1021\/acs.jctc.4c01393","article-title":"Efficient training of neural network\npotentials for chemical and enzymatic reactions by continual learning","volume":"21","author":"Lei","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit43","doi-asserted-by":"publisher","first-page":"1261","DOI":"10.1021\/acs.jctc.2c01172","article-title":"QD\u03c0: A quantum deep potential\ninteraction model for drug discovery","volume":"19","author":"Zeng","year":"2023","journal-title":"J. Chem.\nTheory Comput."},{"key":"2026083118580777400_cit44","doi-asserted-by":"publisher","first-page":"12709","DOI":"10.1021\/acs.jctc.5c01400","article-title":"A universal\naugmentation framework\nfor long-range electrostatics in machine learning interatomic potentials","volume":"21","author":"Kim","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit45","doi-asserted-by":"publisher","first-page":"5194","DOI":"10.1021\/acs.jctc.4c01792","article-title":"Advancing multiscale\nmolecular modeling with machine learning-derived electrostatics","volume":"21","author":"Semelak","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit46","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1021\/acs.jctc.2c01131","article-title":"Accelerated quantum mechanics\/molecular\nmechanics simulations via neural networks incorporated with mechanical\nembedding scheme","volume":"19","author":"Zhou","year":"2023","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit47","doi-asserted-by":"publisher","first-page":"4514","DOI":"10.1021\/acs.jctc.4c00248","article-title":"emle-engine:\nA flexible electrostatic machine learning\nembedding package for multiscale molecular dynamics simulations","volume":"20","author":"Zinovjev","year":"2024","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit48","doi-asserted-by":"publisher","first-page":"4047","DOI":"10.1021\/acs.jcim.4c00478","article-title":"Assessment of embedding schemes in\na hybrid machine learning\/classical potentials (ML\/MM) approach","volume":"64","author":"Grassano","year":"2024","journal-title":"J. Chem. Inf. Model."},{"key":"2026083118580777400_cit49","doi-asserted-by":"publisher","first-page":"6979","DOI":"10.1021\/acs.jctc.5c00598","article-title":"Accurate free energy calculation\nvia multiscale simulations driven by hybrid machine learning and molecular\nmechanics potentials","volume":"21","author":"Wang","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit50","doi-asserted-by":"publisher","first-page":"341","DOI":"10.1016\/0301-0104(81)85176-2","article-title":"Molecular\npolarizabilities calculated with a modified dipole interaction","volume":"59","author":"Thole","year":"1981","journal-title":"Chem. Phys."},{"key":"2026083118580777400_cit51","doi-asserted-by":"publisher","first-page":"4335","DOI":"10.1021\/acs.jctc.5c00149","article-title":"Modeling enzyme reaction\nand mutation by direct machine learning\/molecular mechanics simulations","volume":"21","author":"Sha","year":"2025","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit52","doi-asserted-by":"publisher","first-page":"5583","DOI":"10.1021\/acs.jctc.8b00571","article-title":"Accelerated computation\nof free energy profile at ab initio quantum mechanical\/molecular mechanics\naccuracy via a semi-empirical reference potential. I. Weighted thermodynamics\nperturbation","volume":"14","author":"Li","year":"2018","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit53","doi-asserted-by":"publisher","first-page":"6814","DOI":"10.1021\/acs.jctc.0c00794","article-title":"Accelerated\ncomputation of free energy\nprofile at ab initio quantum mechanical\/molecular mechanics accuracy\nvia a semiempirical reference potential. 3. Gaussian smoothing on\ndensity-of-states","volume":"16","author":"Hu","year":"2020","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit54","doi-asserted-by":"publisher","first-page":"8763","DOI":"10.1038\/s41467-025-63852-x","article-title":"Machine learning\nof charges and long-range\ninteractions from energies and forces","volume":"16","author":"King","year":"2025","journal-title":"Nat.\nCommun."},{"key":"2026083118580777400_cit55","doi-asserted-by":"publisher","first-page":"631","DOI":"10.1126\/science.1411573","article-title":"A priori evaluation\nof aqueous polarization effects through Monte\nCarlo QM-MM simulations","volume":"258","author":"Gao","year":"1992","journal-title":"Science"},{"key":"2026083118580777400_cit56","doi-asserted-by":"publisher","first-page":"8234","DOI":"10.1021\/acs.jctc.3c00973","article-title":"Free energy\nprofile decomposition analysis for QM\/MM simulations of enzymatic\nreactions","volume":"19","author":"Pan","year":"2023","journal-title":"J. Chem. Theory Comput."},{"key":"2026083118580777400_cit57","doi-asserted-by":"publisher","first-page":"1888","DOI":"10.1021\/acs.jctc.2c00914","article-title":"Electrostatic\nembedding of machine learning potentials","volume":"19","author":"Zinovjev","year":"2023","journal-title":"J.\nChem. Theory Comput."},{"key":"2026083118580777400_cit58","doi-asserted-by":"publisher","first-page":"105","DOI":"10.1016\/bs.mie.2016.05.013","article-title":"Born-Oppenheimer ab\ninitio QM\/MM molecular dynamics simulations of enzyme reactions","volume":"577","author":"Zhou","year":"2016","journal-title":"Methods Enzymol"},{"key":"2026083118580777400_cit59","doi-asserted-by":"publisher","first-page":"156002","DOI":"10.1103\/PhysRevLett.127.156002","article-title":"Physically\nmotivated recursively embedded atom neural\nnetworks: Incorporating local completeness and nonlocality","volume":"127","author":"Zhang","year":"2021","journal-title":"Phys. Rev. Lett."},{"key":"2026083118580777400_cit60","doi-asserted-by":"publisher","first-page":"114801","DOI":"10.1063\/5.0080766","article-title":"REANN: A PyTorch-based\nend-to-end multi-functional\ndeep neural network package for molecular, reactive, and periodic\nsystems","volume":"156","author":"Zhang","year":"2022","journal-title":"J. Chem. Phys."},{"key":"2026083118580777400_cit61","doi-asserted-by":"publisher","first-page":"10089","DOI":"10.1063\/1.464397","article-title":"Particle mesh\nEwald: An N\u00b7log(N)\nmethod for Ewald sums in large systems","volume":"98","author":"Darden","year":"1993","journal-title":"J. Chem.\nPhys."},{"key":"2026083118580777400_cit62","doi-asserted-by":"publisher","first-page":"6424","DOI":"10.1038\/s41467-023-42148-y","article-title":"Universal machine learning\nfor the response of atomistic systems\nto external fields","volume":"14","author":"Zhang","year":"2023","journal-title":"Nat. Commun."},{"key":"2026083118580777400_cit63","doi-asserted-by":"publisher","first-page":"398","DOI":"10.1038\/s41467-020-20427-2","article-title":"A fourth-generation\nhigh-dimensional neural network potential with accurate electrostatics\nincluding non-local charge transfer","volume":"12","author":"Ko","year":"2021","journal-title":"Nat. Commun."},{"key":"2026083118580777400_cit64","unstructured":"Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; et al., Gaussian 16, revision A. 03, Gaussian Inc, Wallingford CT, 2016."},{"key":"2026083118580777400_cit65","doi-asserted-by":"publisher","first-page":"5400","DOI":"10.1021\/acs.jpcb.5c00344","article-title":"QM\/MM-MD studies\non the degradation mechanism and size\neffect of PET by PETase","volume":"129","author":"Guo","year":"2025","journal-title":"J. Phys. Chem. B"},{"key":"2026083118580777400_cit66","doi-asserted-by":"publisher","first-page":"1076","DOI":"10.1007\/s40242-025-5133-6","article-title":"Theoretical studies on the deacylation\nstage of the hydrolysis of PET heptamer by PETase","volume":"41","author":"Guo","year":"2025","journal-title":"Chem. Res. Chin. Univ."},{"key":"2026083118580777400_cit67","doi-asserted-by":"publisher","first-page":"275","DOI":"10.1016\/0010-4655(95)00053-I","article-title":"The calculation\nof the potential of mean force using computer simulations","volume":"91","author":"Roux","year":"1995","journal-title":"Comput. Phys. Commun."},{"key":"2026083118580777400_cit68","doi-asserted-by":"publisher","first-page":"4509","DOI":"10.1021\/j100119a043","article-title":"Constant-temperature\nfree energy surfaces for physical\nand chemical processes","volume":"97","author":"Boczko","year":"1993","journal-title":"J. Phys. Chem."},{"key":"2026083118580777400_cit69","doi-asserted-by":"publisher","first-page":"3172","DOI":"10.1039\/B517914A","article-title":"Advances in methods\nand algorithms in a modern quantum chemistry program package","volume":"8","author":"Shao","year":"2006","journal-title":"Phys. Chem. Chem. Phys."},{"key":"2026083118580777400_cit70","unstructured":"Case, D. A.; Darden, T. A.; Cheatham, T. E., III; Simmerling, C. L.; Wang, J.; Duke, R. E.; Luo, R.; Walker, R. C.; Zhang, W.; Merz, K. M.; et al., AMBER\n12, University of California, San Francisco, 2012."},{"key":"2026083118580777400_cit71","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":"2026083118580777400_cit72","doi-asserted-by":"publisher","first-page":"998","DOI":"10.1126\/science.266.5187.998","article-title":"Simulations\nof gas-phase chemical reactions: Applications to SN2 nucleophilic\nsubstitution","volume":"266","author":"Hase","year":"1994","journal-title":"Science"},{"key":"2026083118580777400_cit73","doi-asserted-by":"publisher","first-page":"1882","DOI":"10.1126\/science.279.5358.1882","article-title":"Gas-phase\nionic\nreactions: Dynamics and mechanism of nucleophilic displacements","volume":"279","author":"Chabinyc","year":"1998","journal-title":"Science"},{"key":"2026083118580777400_cit74","doi-asserted-by":"publisher","first-page":"9461","DOI":"10.1021\/jp046954j","article-title":"Free energy profiles\nfor the identity SN2 reactions Cl\u2013+CH3Cl and NH3+H3BNH3: A constraint\nab initio molecular dynamics study","volume":"108","author":"Yang","year":"2004","journal-title":"J. Phys.\nChem. A"},{"key":"2026083118580777400_cit75","doi-asserted-by":"publisher","first-page":"9054","DOI":"10.1021\/ja049070m","article-title":"Steric retardation\nof SN2 reactions in the gas phase and solution","volume":"126","author":"Vayner","year":"2004","journal-title":"J. Am. Chem. Soc."},{"key":"2026083118580777400_cit76","doi-asserted-by":"publisher","first-page":"22497","DOI":"10.1039\/D2CP02931F","article-title":"Graph-convolutional neural networks\nfor (QM)ML\/MM molecular dynamics simulations","volume":"24","author":"Hofstetter","year":"2022","journal-title":"Phys. Chem. Chem. Phys."},{"key":"2026083118580777400_cit77","doi-asserted-by":"publisher","first-page":"1442","DOI":"10.1021\/acs.jctc.7b01195","article-title":"Molecular dynamics\nsimulations with quantum mechanics\/molecular mechanics\nand adaptive neural networks","volume":"14","author":"Shen","year":"2018","journal-title":"J. Chem. Theory\nComput."},{"key":"2026083118580777400_cit78","doi-asserted-by":"publisher","first-page":"6257","DOI":"10.1021\/acs.jpcb.4c01466","article-title":"Software\ninfrastructure for next-generation\nQM\/MM\u2212\u0394MLP force fields","volume":"128","author":"Giese","year":"2024","journal-title":"J. Phys.\nChem. B"},{"key":"2026083118580777400_cit79","doi-asserted-by":"publisher","first-page":"1213","DOI":"10.1021\/acs.jctc.1c00811","article-title":"Accurate free energies\nfor complex condensed-phase reactions using an artificial neural network\ncorrected DFTB\/MM methodology","volume":"18","author":"Gomez-Flores","year":"2022","journal-title":"J. Chem. Theory\nComput."},{"key":"2026083118580777400_cit80","doi-asserted-by":"publisher","first-page":"8542","DOI":"10.1039\/D6SC01156J","article-title":"Simulating enzyme catalysis with\nelectrostatically embedded machine learning potentials","volume":"17","author":"Gradisteanu","year":"2026","journal-title":"Chem. Sci."},{"key":"2026083118580777400_cit81","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1007\/s00214-005-0008-x","article-title":"Pseudobond\nab initio QM\/MM approach and its applications to enzyme reactions","volume":"116","author":"Zhang","year":"2006","journal-title":"Theor. Chem. Acc."},{"key":"2026083118580777400_cit82","doi-asserted-by":"publisher","first-page":"024114","DOI":"10.1063\/1.1834899","article-title":"Improved\npseudobonds for combined ab initio quantum mechanical\/molecular mechanical\nmethods","volume":"122","author":"Zhang","year":"2005","journal-title":"J. Chem. Phys."},{"key":"2026083118580777400_cit83","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1063\/1.478083","article-title":"A pseudobond\napproach to combining\nquantum mechanical and molecular mechanical methods","volume":"110","author":"Zhang","year":"1999","journal-title":"J. Chem. Phys."},{"key":"2026083118580777400_cit84","doi-asserted-by":"publisher","first-page":"1191","DOI":"10.1126\/science.280.5367.1191","article-title":"Building a\nbetter aspirin","volume":"280","author":"Pennisi","year":"1998","journal-title":"Science"},{"key":"2026083118580777400_cit85","doi-asserted-by":"publisher","first-page":"3228","DOI":"10.1073\/pnas.91.8.3228","article-title":"Selective\ninhibition\nof inducible cyclooxygenase 2 in vivo is antiinflammatory and nonulcerogenic","volume":"91","author":"Masferrer","year":"1994","journal-title":"Proc. Natl. Acad. Sci. U. S. A."},{"key":"2026083118580777400_cit86","doi-asserted-by":"publisher","first-page":"70","DOI":"10.1021\/ja5112964","article-title":"Mechanistic insights\ninto a classic wonder drug\u2500Aspirin","volume":"137","author":"Lei","year":"2015","journal-title":"J.\nAm. Chem. Soc."},{"key":"2026083118580777400_cit87","doi-asserted-by":"publisher","first-page":"768","DOI":"10.1126\/science.1260352","article-title":"Plastic waste inputs\nfrom land into the ocean","volume":"347","author":"Jambeck","year":"2015","journal-title":"Science"},{"key":"2026083118580777400_cit88","doi-asserted-by":"publisher","first-page":"6709","DOI":"10.1021\/acsomega.9b00222","article-title":"Analysis and prevention of microplastics\npollution\nin water: Current perspectives and future directions","volume":"4","author":"Pico\u0301","year":"2019","journal-title":"ACS Omega"},{"key":"2026083118580777400_cit89","doi-asserted-by":"publisher","first-page":"110733","DOI":"10.1016\/j.marpolbul.2019.110733","article-title":"Review on plastic wastes in marine\nenvironment \u2013 Biodegradation and biotechnological solutions","volume":"150","author":"A","year":"2020","journal-title":"Mar. Pollut. Bull."},{"key":"2026083118580777400_cit90","doi-asserted-by":"publisher","first-page":"1196","DOI":"10.1126\/science.aad6359","article-title":"A bacterium that degrades\nand assimilates\npoly(ethylene terephthalate)","volume":"351","author":"Yoshida","year":"2016","journal-title":"Science"},{"key":"2026083118580777400_cit91","doi-asserted-by":"publisher","first-page":"1471","DOI":"10.1002\/cbic.201800097","article-title":"Protein crystallography and site-direct\nmutagenesis\nanalysis of the poly(ethylene terephthalate) hydrolase PETase from\nideonella sakaiensis","volume":"19","author":"Liu","year":"2018","journal-title":"ChemBiochem"},{"key":"2026083118580777400_cit92","doi-asserted-by":"publisher","first-page":"888","DOI":"10.1016\/j.eng.2018.09.007","article-title":"Enhanced poly(ethylene\nterephthalate)\nhydrolase activity by protein engineering","volume":"4","author":"Ma","year":"2018","journal-title":"Engineering"},{"key":"2026083118580777400_cit93","doi-asserted-by":"publisher","first-page":"662","DOI":"10.1038\/s41586-022-04599-z","article-title":"Machine learning-aided engineering of hydrolases for\nPET depolymerization","volume":"604","author":"Lu","year":"2022","journal-title":"Nature"},{"key":"2026083118580777400_cit94","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1718804115","article-title":"Characterization and\nengineering of a plastic-degrading\naromatic polyesterase","volume":"115","author":"Austin","year":"2018","journal-title":"Proc. Natl. Acad. Sci.\nU. S. A."},{"key":"2026083118580777400_cit95","doi-asserted-by":"publisher","first-page":"12771","DOI":"10.1021\/jp953748q","article-title":"Current\nstatus of transition-state\ntheory","volume":"100","author":"Truhlar","year":"1996","journal-title":"J. Phys. Chem."},{"key":"2026083118580777400_cit96","doi-asserted-by":"publisher","first-page":"6835","DOI":"10.1021\/jacs.4c17015","article-title":"Neural network potential with multiresolution approach\nenables accurate prediction of reaction free energies in solution","volume":"147","author":"Pultar","year":"2025","journal-title":"J. Am. Chem. Soc."},{"key":"2026083118580777400_cit97","doi-asserted-by":"publisher","first-page":"9512","DOI":"10.1021\/jp953640a","article-title":"On the validity of electrostatic linear response in\npolar solvents","volume":"100","author":"\u00c5qvist","year":"1996","journal-title":"J. Phys. Chem."},{"key":"2026083118580777400_cit98","doi-asserted-by":"publisher","first-page":"2646","DOI":"10.1021\/j100457a034","article-title":"Solvent effects on the structure and the electronic properties of\nsimple molecules. A MINDO\/3-SCRF-MO study","volume":"84","author":"Tapia","year":"1980","journal-title":"J.\nPhys. Chem."},{"key":"2026083118580777400_cit99","doi-asserted-by":"publisher","first-page":"4320","DOI":"10.1021\/j100347a081","article-title":"Free energy calculations\nof ion hydration: An analysis of the Born model in terms of microscopic\nsimulations","volume":"93","author":"Jayaram","year":"1989","journal-title":"J. Phys. Chem."},{"key":"2026083118580777400_cit100","doi-asserted-by":"publisher","first-page":"891","DOI":"10.1038\/s41929-025-01398-3","article-title":"General reactive element-based machine learning potentials for heterogeneous\ncatalysis","volume":"8","author":"Yang","year":"2025","journal-title":"Nat. Catal"},{"key":"2026083118580777400_cit101","doi-asserted-by":"publisher","first-page":"1701","DOI":"10.1021\/acs.jctc.1c01021","article-title":"Learning atomic multipoles:\nPrediction of the electrostatic potential with equivariant graph neural\nnetworks","volume":"18","author":"Thu\u0308rlemann","year":"2022","journal-title":"J. Chem. Theory Comput."}],"container-title":["Journal of Chemical Theory\nand Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jctc.6c00900","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jctcce\/article-pdf\/22\/16\/8131\/66470346\/acs.jctc.6c00900.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jctcce\/article-pdf\/22\/16\/8131\/66470346\/acs.jctc.6c00900.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T22:58:12Z","timestamp":1788217092000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jctcce\/article\/22\/16\/8131\/5251048\/Enhancing-Machine-Learned-Electrostatic-Models"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,10]]},"references-count":101,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8,10]]},"published-print":{"date-parts":[[2026,8,25]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jctc.6c00900","relation":{},"ISSN":["1549-9618","1549-9626"],"issn-type":[{"value":"1549-9618","type":"print"},{"value":"1549-9626","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,10]]}}}