Journal of Chemical Theory and Computation最新文献

{"title":"Impurity-Preserved Density Matrix Embedding Theory for Local Electronic Excitations","authors":"Teng Zhang,&nbsp;Ze-Wei Li,&nbsp;Zhebin Guan,&nbsp;Hong Jiang","doi":"10.1021/acs.jctc.6c01251","DOIUrl":"10.1021/acs.jctc.6c01251","url":null,"abstract":"<p>Densitymatrix embedding theory (DMET), which is usuallybasedon a Schmidt decomposition of Slater determinants by partitioningthe full system into impurity and environment in terms of local orthogonalorbitals (LOs), has demonstrated considerable promise in electronicstructure studies because it enables the extraction of local propertiesusing a high-level solver within an embedded impurity subsystem withgreatly reduced degrees of freedom, thereby achieving a balance betweenaccuracy and computational cost. However, its application to excitedstates of strongly correlated systems, such as lanthanide complexes,remains challenging because the errors relative to all-electron resultscan still be significant. Motivated by the success of the previouslydeveloped atomic orbitals (AOs) based DMET framework (Ai, Li, andJiang, Phys. Rev. Lett. 2025, 135, 026502.), termed AO-DMET, whichattains improved accuracy by constructing the embedded subspace basedon a non-orthogonal decomposition of the Slater determinant in termsof AOs, we propose a new LO-based partitioning scheme that fully preservesthe impurity space spanned by corresponding AOs and can achieve accuracyclosely matching that of AO-DMET while retaining the orthogonal partitionand its associated computational efficiency. The performance of theproposed method is demonstrated through excitation energy calculationsfor several representative lanthanide complexes. These results establishan efficient and accurate partitioning scheme for describing excitedstates in strongly correlated systems within the DMET framework.</p>","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"22 19","pages":"10126–10136"},"PeriodicalIF":5.8,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148973885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Coupling Quantum Mechanical Modeling and Molecular Dynamics on Heterogeneous Supercomputers for Studying Distal Mutation Effects on Drug Binding in HIV-1","authors":"William Dawson,&nbsp;Louis Beal,&nbsp;Marco Zaccaria,&nbsp;Luigi Genovese","doi":"10.1021/acs.jctc.6c01377","DOIUrl":"10.1021/acs.jctc.6c01377","url":null,"abstract":"<p>Predicting how proteinmutations affect drug bindingremains amajor challenge, particularly when the mutations are distal from thebinding site. In this study, we introduce a coupled simulation workflowthat combines classical molecular dynamics (MD) with high-throughputquantum mechanical (QM) analysis to reveal the electronic structuresignatures of mutation-induced drug resistance in the HIV-1 protease.Our workflow leverages GPU-accelerated MD to generate conformationalensembles, and performs in operando linear-scaling density functionaltheory (DFT) calculations on selected frames parallelized on a coupledpartition of CPU nodes. This design enables efficient, massively parallelquantum analysis of protein–ligand complexes at atomic resolution.Using this approach, we investigate resistance to the antiviral Darunavirin a multimutant HIV-1 protease variant. By mapping the network ofelectronic interactions across the binding interface, our resultshighlight the critical role of conformational sampling and quantuminsight in understanding distal mutation effects, and demonstratea scalable computational strategy for studying complex biophysicalmechanisms of drug resistance. We argue that this type of analysismay pave the way for designing inhibitors that maintain binding stabilityagainst systemic, mutation-induced destabilization.</p>","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"22 19","pages":"10165–10179"},"PeriodicalIF":5.8,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149003486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accounting for Nanofilm Contributions in Interfacial Free Energy Calculations Using Classical Density Functional Theory","authors":"Yafan Yang,&nbsp;Zufeng Zuo,&nbsp;Xingyu Zhao,&nbsp;Shuyu Sun,&nbsp;Denvid Lau","doi":"10.1021/acs.jctc.6c01465","DOIUrl":"10.1021/acs.jctc.6c01465","url":null,"abstract":"<p>Fluidnanofilms play a fundamental role in nanoscale interfacialthermodynamics, yet their interfacial behavior and treatment withinclassical density functional theory (cDFT) warrant further investigation.We investigate nanofilm interfacial properties using a cDFT frameworkbuilt upon the perturbed-chain statistical associating fluid theory(PC-SAFT) for both fluid–fluid and fluid–solid interfacialsystems. Comparison with molecular simulations shows that the long-standingnumerical discrepancies regarding free-standing nanofilms persistin the PC-SAFT functional predictions, supporting the view that theystem from thermal capillary-wave fluctuations rather than deficienciesof the density functional itself, as such fluctuations are inherentlyneglected in the mean-field approximation adopted by standard cDFT.Importantly, we establish a thermodynamically consistent frameworkfor interfacial free energy (IFE) calculation, which explicitly incorporatesnanofilm thermodynamic contributions and redefines the effective fluidvolume by excluding the solid-phase region. The proposed method exhibitsexcellent consistency with another method based on the relationshipbetween IFE and disjoining pressure for fluid systems, while bothfluid–fluid and fluid–solid IFEs differ substantiallyfrom those predicted by the conventional method. We find that neglectingnanofilm contributions induces prominent size-dependent variationsin the IFE and contact angle of hemicylindrical droplets, which isinconsistent with the extensive literature. Different methods alsolead to opposite signs of the line tension for a hemispherical argonnanodroplet on a strongly lyophobic surface. The proposed frameworkprovides a unified molecular-level basis for understanding interfacialprocesses involving nanofilms, including wetting, nucleation, adsorption,and other phenomena that rely on the accurate evaluation of IFEs.</p>","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"22 19","pages":"10403–10418"},"PeriodicalIF":5.8,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149049730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Flexible On-the-Fly Active Learning for Molecular Infrared Spectra","authors":"Giulio Benedini,&nbsp;Matti Hellström,&nbsp;Lucas Visscher","doi":"10.1021/acs.jctc.6c00756","DOIUrl":"10.1021/acs.jctc.6c00756","url":null,"abstract":"<p>Infrared spectroscopyserves as an indispensable analytical toolacross materials science, pharmaceutical development, and the lifesciences. Computational insights derived from first-principles methodsenhance the interpretation of vibrational spectra, yet their applicationremains hindered by high computational cost. Machine-learned interatomicpotentials offer a route to accelerate these predictions by multipleorders of magnitude without sacrificing ab initio accuracy. Here,we present an on-the-fly active learning framework for the automatedtraining of such potentials specifically tailored to infrared spectralprediction. Our approach employs a multitask acquisition strategy,in which model performance is evaluated iteratively on designatedobjectives, avoiding the dependence on fixed training sets. We demonstratethat this multitask scheme creates interatomic potentials that exhibitrobustness and fidelity in complex workflows for calculating infraredspectra.</p>","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"22 19","pages":"10309–10323"},"PeriodicalIF":5.8,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Progress toward a Better BOCS: Systematic Coarse-Graining with Local Density Potentials","authors":"Maria C. Lesniewski,Michael R. DeLyser,W. G. Noid","doi":"10.1021/acs.jctc.6c01228","DOIUrl":"https://doi.org/10.1021/acs.jctc.6c01228","url":null,"abstract":"We introduce version 5.0 of the bottom-up open-source coarse-graining software (BOCS) package. BOCS employs the force-matching variational principle to parametrize potentials for coarse-grained (CG) models directly from atomically detailed simulations. BOCS version 5.0 significantly extends previous versions by treating potentials that depend upon the local density (LD) around each particle as well as potentials that depend upon the square gradient (SG) of this local density. We also describe a new package, PKG-BOCS, for simulating these potentials with LAMMPS. This software treats complex molecular topologies and provides considerable flexibility for defining the local density as well as the LD and SG potentials. We present numerical calculations that provide physical insight into these potentials and their dependence on the definition of the local density. These calculations also demonstrate the accuracy and reliability of our implementation. Finally, we demonstrate that LD potentials can improve the structural fidelity, thermodynamic properties, and transferability of CG models for water.","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"12 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149152092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Vertical Excitation Energies in Solution via a Two-Time-Scale Solvation Model Coupled to Pair-Density Functional Theory","authors":"Arpan Choudhury,Matthew R. Hermes,Donald G. Truhlar,Laura Gagliardi","doi":"10.1021/acs.jctc.6c01690","DOIUrl":"https://doi.org/10.1021/acs.jctc.6c01690","url":null,"abstract":"Calculations of many electronically excited states in solution require both a multiconfigurational treatment of the solute and an efficient treatment of solvent effects. Multiconfiguration pair-density functional theory (MC-PDFT) and its multistate, linearized extension (L-PDFT) can recover correlation beyond state-averaged complete active space self-consistent field theory (SA-CASSCF) at a fraction of the cost of multireference perturbation theory (e.g., CASPT2) or multireference configuration interaction. Here we combine MC-PDFT and L-PDFT with a nonequilibrium-treatment of the electrostatic solvation effect using the solvation model based on density (SMD). The resulting models are called MC-NSCRF-PDFT and L-NSCRF-PDFT, where NSCRF denotes the key state-specific excited-state nonequilibrium self-consistent reaction field step. We apply the new models to the solvatochromic shifts of the lowest valence excitations of formaldehyde, acetone, acrolein, and methylenecyclopropene and to the aqueous shifts of the lowest π → π* and n → π* transitions of cytosine. For vertical excitation energies in solution, MC-NSCRF-PDFT and L-NSCRF-PDFT give results that are in agreement with the far-more-expensive NSCRF model based on multistate CASPT2. These results show that MC-NSCRF-PDFT and L-NSCRF-PDFT can be used as accurate and computationally efficient methods for electronic excitation energies in the condensed phase.","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"69 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149152066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multireference Density Matrix Embedding for Spin-Phonon Relaxation","authors":"Shreya Verma,Antonio L. Mariano,Matthew R. Hermes,Alessandro Lunghi,Giulia Galli,Laura Gagliardi","doi":"10.1021/acs.jctc.6c01503","DOIUrl":"https://doi.org/10.1021/acs.jctc.6c01503","url":null,"abstract":"Spin-phonon coupling governs magnetic relaxation in numerous systems, including single-molecule magnets and molecular spin qubits. In most cases, the accurate prediction of spin relaxation rates requires multireference electronic structure methods, but their computational cost has largely restricted such calculations to isolated molecules. Here we show that spin-phonon relaxation rates can be computed within a multireference density matrix embedding framework. We apply the approach to three cobalt- and two dysprosium-based single-molecule magnets and to a cobalt-based molecular crystal. Across all systems, treating only the first coordination sphere of the magnetic center at the multireference level reproduces spin relaxation rates in good agreement with non-embedded CASSCF calculations, while reducing the correlated problem to 10–68% of the total basis functions. Periodic calculations further demonstrate that spin relaxation rates can be computed for a molecular crystal using an embedded active space of only 259 basis functions out of a total of 2569. These results show that multireference density matrix embedding extends quantitative spin-phonon relaxation calculations from isolated molecules to molecular crystals.","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"16 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149152102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ΔNAzyme: A Foundational QM/MM Δ-Machine Learning Potential for Nucleic Acid Enzyme Catalysis","authors":"Erika McCarthy,Timothy J. Giese,Şölen Ekesan,Julie Puyo-Fourtine,Lauren Lerew,Andreas W. Götz,Darrin M. York","doi":"10.1021/acs.jctc.6c01635","DOIUrl":"https://doi.org/10.1021/acs.jctc.6c01635","url":null,"abstract":"Nucleic acid enzymes (NAzymes), including ribozymes, DNAzymes, and XNAzymes, are programmable biocatalysts that catalyze diverse chemical reactions and provide unique platforms for investigating catalysis and engineering molecular function. Their central roles in biology and evolution, together with emerging applications in biotechnology, synthetic biology, and medicine, have created an urgent need for predictive computational design frameworks. Although multiscale computational enzymology integrating ab initio quantum mechanics can, in principle, provide predictive insight into these systems, current approaches remain computationally prohibitive for large, complex NAzymes and high-throughput design workflows. Machine learning potentials have the potential to enhance the accuracy and efficiency of multiscale modeling approaches, but QM/MM contexts in which the MM environment is highly polarizing remain challenging. Nucleic acid enzymes present a particularly demanding case: experimentally relevant predictions require an accuracy that semiempirical QM/MM methods alone cannot reliably deliver in their often large, highly charged, metal-dependent active sites. Building on the range-corrected QM/MM Δ-machine learning potential (ΔMLP) framework, we develop ΔNAzyme, a foundational ΔMLP for nucleic acid enzyme reactions. We generate a training set of QM/MM simulations spanning diverse nucleic acid enzyme classes, with energies and forces computed at both the reference (DFTB3) and target (PBE0/6–31G*) levels of theory, yielding a correction on top of DFTB3 that recovers PBE0 quality results. We train models using both the Deep Potential and MACE architectures. The MACE model exhibits enhanced transferability, which we attribute to the equivariant, many-body message-passing framework. Free energy profiles from case-study reactions validate ΔNAzyme against target-level results and show that the model supports efficient fine-tuning to new nucleic acid enzyme systems with minimal additional data.","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"57 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149152112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ConSolv: Solvent-Conditional Machine Learning Implicit Solvent Potential","authors":"Linying Zhang,Julija Zavadlav","doi":"10.1021/acs.jctc.6c01253","DOIUrl":"https://doi.org/10.1021/acs.jctc.6c01253","url":null,"abstract":"Implicit solvent machine learning potentials (MLPs) offer a powerful route to bridging the gap between accuracy and efficiency in molecular simulations. However, existing models have largely focused on aqueous environments, overlooking the diverse and important roles of nonaqueous solvents in areas such as organic synthesis and battery technology. Here, we present ConSolv, a solvent-conditional MLP architecture that explicitly incorporates solvent effects on solute interactions through an attention-based solvent embedding block. By combining experimental solvation free energy data with ab initio data, we train a single implicit solvent MLP that is transferable across 66 common organic solvents. ConSolv outperforms classical explicit solvent methods and selected ab initio implicit solvent approaches across multiple solvation free energy benchmarks, and demonstrates generalization to unseen solvents. Beyond solvation free energies, the model shows close agreement with experimental nuclear magnetic resonance (NMR) data for γ-fluorohydrin molecules in chloroform. ConSolv’s architecture is readily extensible to broader chemical spaces and alternative training strategies, while its attention-based design supports explainable artificial intelligence (AI) analysis that can help elucidate complex, solvent-dependent molecular interactions.","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"53 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149152067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"JustEvo: A Prior-Guided Adaptive Mixture-of-Experts\u0000Framework for Multifunctional Protein Directed Evolution","authors":"Zhenyu Chen,Tianxi Sun,Zicheng Ma,Xiaohan Lin,Shizhe Tang,Jia Zheng,Yi Qin Gao","doi":"10.1021/acs.jctc.6c01288","DOIUrl":"https://doi.org/10.1021/acs.jctc.6c01288","url":null,"abstract":"Abstract Machine-learning-guided directed evolution can reduce the experimental burden of protein engineering by prioritizing variants before they are tested. In practice, however, many campaigns begin with few or no target-specific measurements, face choices about how to use heterogeneous protein-model priors, and require functional improvement to be balanced with developability properties such as expression and stability. Here, we present JustEvo, a closed-loop framework for low-data and multifunctional protein evolution. Before experimental measurement data are available, JustEvo nominates an informative first experimental batch using a mutation-prior model or a property-specific predictor. After measurements are obtained, it trains surrogate experts on multiple protein representations and adaptively combines their predictions according to their agreement with the current experimental fitness ranking. We evaluated these components using an EVOLVEpro-derived MoE validation data set and a ProteinGym-derived DMS collection split into a cold-start validation data set and a reserved benchmark data set. On the 81-landscape benchmark data set, JustEvo achieved higher final-round top 5 average fitness than EVOLVEpro and RelaVDEP under the same finite-budget acquisition protocol, with normalized scores of 0.862 for JustEvo, 0.849 for EVOLVEpro, and 0.818 for RelaVDEP. We further adapted JustEvo for fluorescent-protein optimization as a multiobjective design task. In this setting, objective-specific surrogate models were used to rank candidates by their predicted joint performance across coupled properties. We tested this extension on four in silico GFP landscapes and an experimental EYFP closed-loop campaign, where JustEvo expanded the fluorescence-stability Pareto front and increased Pareto hypervolume by 872.1%. These results show that coordinating prior-guided initialization, adaptive mixture-of-experts integration and objective-aware candidate ranking can improve protein variant selection under limited assay budgets.","PeriodicalId":45,"journal":{"name":"Journal of Chemical Theory and Computation","volume":"10 1","pages":""},"PeriodicalIF":5.5,"publicationDate":"2026-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149151225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}