Journal of Chemical Information and Modeling 最新文献

筛选
英文 中文
DBNAMP: A Chemically Curated Database of Noncanonical Antimicrobial Peptides with Standardized Molecular Representations DBNAMP:具有标准化分子表征的非规范抗菌肽的化学策划数据库
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-27 DOI: 10.1021/acs.jcim.6c02717
Tian Liang, Bai Li, Xiaokun Hong, Lin Yang, Jinyu Li
{"title":"DBNAMP: A Chemically Curated Database of Noncanonical Antimicrobial Peptides with Standardized Molecular Representations","authors":"Tian Liang,&nbsp;Bai Li,&nbsp;Xiaokun Hong,&nbsp;Lin Yang,&nbsp;Jinyu Li","doi":"10.1021/acs.jcim.6c02717","DOIUrl":"10.1021/acs.jcim.6c02717","url":null,"abstract":"<p>Antimicrobial peptides (AMPs) are promising anti-infectiveagents,yet conventional sequence-centered resources often inadequately representnoncanonical residues, stereochemical variations, terminal modifications,cyclization, lipidation, stapling, and peptidomimetic scaffolds, limitingchemically meaningful annotation, comparison, and structure–activityrelationship analysis. Here, we present DBNAMP, a manually curatedand chemically explicit database of AMP-like entities containing residue-levelnoncanonical features, molecular-level covalent modifications, orboth; it currently comprises 1,859 curated entries, including 1,421entries with computed representative three-dimensional conformersand 129 standardized controlled-vocabulary records for noncanonicalresidues and peptide modifications. Among 537 source-level ambiguity-auditrecords, 84 remained unresolved after review, of which 82 retainedat least one uppercase X character. Only one such record was includedin the final release. Each DBNAMP entry integrates sequence, chemicalannotation, antimicrobial activity, target organisms, physicochemicaldescriptors, molecular representations, structural information, andliterature or patent provenance. In addition to browsing, filtering,and data download, DBNAMP enables chemically informed explorationof related peptide entities through standardized molecular representationsand incorporates a lightweight SMILES-based prioritization modulefor preliminary ranking of candidate molecules. By bridging sequence-levelpeptide annotation and chemically explicit molecular representation,DBNAMP supports chemoinformatics analysis, structure–activityrelationship studies, virtual screening, and computer-aided designof chemically diverse AMP candidates and is freely accessible at https://dbnamp.cn/.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"13974–13985"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149251104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
C-MAGE: Automated Extraction and Organization Engine for Molecular Representations in Chemical Documents C-MAGE:化学文件中分子表示的自动提取和组织引擎
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-27 DOI: 10.1021/acs.jcim.6c02975
Alexander Taylor, Jeehyun Hwang, Mihir Surve, Wei Wang, Olaf Wiest
{"title":"C-MAGE: Automated Extraction and Organization Engine for Molecular Representations in Chemical Documents","authors":"Alexander Taylor,&nbsp;Jeehyun Hwang,&nbsp;Mihir Surve,&nbsp;Wei Wang,&nbsp;Olaf Wiest","doi":"10.1021/acs.jcim.6c02975","DOIUrl":"10.1021/acs.jcim.6c02975","url":null,"abstract":"<p>To exploit the datapresent in chemical literature for AI applicationsin chemistry, an expanded and reliable extraction pipeline is needed.AI-ready chemical databases broaden chemical space coverage and providebenchmarking and training sets for machine learning model developmentto drive innovation within the field. To curate data from chemicaldocuments, a multimodal approach is needed. C-MAGE is an open-sourcemolecule extraction engine that processes figures and tables in PDFfiles to provide machine-readable molecule data sets of the structuresas CXSMILES strings, an encompassing representation that evades heuristictranslation of superatoms to SMILES subunits. This led to a 7.1% increasein accurate molecular representations generated. All data extractedundergo confidence-based sorting to ensure high precision (77.6%)in data set collection and remove the need for manually sorting datapoint validity. Both the CXSMILES and their confidence values arereproducible and are generated efficiently (1.7 s per molecule). C-MAGE’suse is demonstrated by the generation of an annotated data set containing1454 CXSMILES strings from 38 chemical documents. C-MAGE thus enableslarge-scale information extraction of chemical structures in a consistentand reproducible pipeline, complementing the more common text extractions.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"13986–13995"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149275054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An Antisymmetric Framework with Cross-Modal Information Decoupling for Predicting Missense Mutation-Induced Changes in Protein–Ligand Binding Affinity 一个跨模态信息解耦的反对称框架用于预测错义突变诱导的蛋白质-配体结合亲和力的变化
IF 5.6 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 DOI: 10.1021/acs.jcim.6c02822
Henghui Fan,Chuanmei Bi,Dayu Tan,Fei Wang,Junfeng Xia
{"title":"An Antisymmetric Framework with Cross-Modal Information Decoupling for Predicting Missense Mutation-Induced Changes in Protein–Ligand Binding Affinity","authors":"Henghui Fan,Chuanmei Bi,Dayu Tan,Fei Wang,Junfeng Xia","doi":"10.1021/acs.jcim.6c02822","DOIUrl":"https://doi.org/10.1021/acs.jcim.6c02822","url":null,"abstract":"Accurately predicting changes in protein–ligand binding affinity induced by missense mutations is important for understanding disease mechanisms, identifying drug resistance mutations, and supporting precision medicine. However, existing computational methods often directly integrate protein sequence and structure information, which may introduce background information, structural uncertainty, and redundant cross-modal information, limiting the identification of transferable mutation-related features. In addition, most existing approaches do not explicitly enforce the thermodynamic consistency of binding free energy changes upon mutation reversal. Here, we present MutBindPred, a cross-modal deep learning framework for predicting missense mutation-induced changes in protein–ligand binding affinity. MutBindPred jointly models wild-type and mutant protein–ligand complexes and introduces a cross-modal information decoupling module to separate modality-shared background information from modality-specific mutation-related signals. The framework further captures directional differences between wild-type and mutant states and incorporates an antisymmetric constraint to improve the thermodynamic consistency of predicted ΔΔG values. Under rigorous protein and ligand cold-start settings, MutBindPred achieves improved generalization performance compared with existing baseline methods. Further analyses show that sequence-derived heterogeneous representations provide the strongest transferable predictive signals. Structural representations show limited standalone predictive utility but contribute as an auxiliary modality for guiding cross-modal representation learning. Overall, MutBindPred provides a thermodynamically consistent and generalizable framework for predicting missense mutation-induced changes in protein–ligand binding affinity.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149368316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
From Random Splits to Unseen Solvents: Quantum-Chemistry-Derived Surface Descriptors Improve Solubility Prediction 从随机分裂到看不见的溶剂:量子化学衍生的表面描述符改善溶解度预测
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-23 DOI: 10.1021/acs.jcim.6c01879
Ziqi Liu, Guanhao Qi, Guojun Jiang, Hua Zhao, Jingpeng Zhou, Fengshan Zhang, Shanshan Gao, Xiaoming Song
{"title":"From Random Splits to Unseen Solvents: Quantum-Chemistry-Derived Surface Descriptors Improve Solubility Prediction","authors":"Ziqi Liu,&nbsp;Guanhao Qi,&nbsp;Guojun Jiang,&nbsp;Hua Zhao,&nbsp;Jingpeng Zhou,&nbsp;Fengshan Zhang,&nbsp;Shanshan Gao,&nbsp;Xiaoming Song","doi":"10.1021/acs.jcim.6c01879","DOIUrl":"10.1021/acs.jcim.6c01879","url":null,"abstract":"<p>Solventscreening often requires predictions for solventsthatare absent from the training data. Using 103,262 aligned BigSolDBv2.1 records, we constructed a fixed solvent-level OOD benchmark inwhich five solvent identities were excluded from model training, validation,early stopping, SHAP ranking, and feature selection. A compact RDKittwo-dimensional baseline reaches an IID <em>R</em><sup>2</sup> = 0.954 ± 0.001 under random splitting but falls to an OOD <em>R</em><sup>2</sup> = 0.541 ± 0.019 on the held-out solvents.Under the same benchmark, adding Gasteiger charge proxies and monomer-levelGFN2-xTB/COSMO σ-moment surface descriptors raises the OOD <em>R</em><sup>2</sup> to 0.756 ± 0.007. A validation-selected47-feature representation that also includes deterministic solute–solventcompatibility features reaches an OOD <em>R</em><sup>2</sup> = 0.788 ± 0.008, an MAE = 0.413, and an RMSE = 0.574 log<sub>10</sub> <em>S</em> unit under the 20-seed protocol.The improvement depends on solvent identity: DMSO, NMP, and cyclohexaneshow the largest error reductions, ethylene glycol changes little,and <em>n</em>-propyl acetate is slightly worse than the electrostaticbaseline. Learner controls and pair-xTB calculations are used as supportingchecks; optimized 1:1 dimers and small 1:2/2:1 clusters do not improvethe fixed-OOD result. MICA v0.2 distributes the frozen 20-seed ensemblefor prepared-descriptor inference with ensemble-disagreement summaries,applicability-domain flags, and solvent-ranking reports.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"13996–14011"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149241656","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An Informatics Approach to Recommender Systems for PFAS Molecule Replacement PFAS分子置换推荐系统的信息学方法
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-29 DOI: 10.1021/acs.jcim.6c02203
Marcus Schwarting, Krishna Rajan
{"title":"An Informatics Approach to Recommender Systems for PFAS Molecule Replacement","authors":"Marcus Schwarting,&nbsp;Krishna Rajan","doi":"10.1021/acs.jcim.6c02203","DOIUrl":"10.1021/acs.jcim.6c02203","url":null,"abstract":"<p>Per- and polyfluoroalkyl substances (PFAS) are persistentenvironmentalcontaminants used across an enormous range of industrial and consumerapplications, and identifying safe, nonfluorinated replacements remainsa slow, largely trial-and-error process. In this work, we constructa recommender engine for suggesting PFAS alternative chemistries basedon past replacements. We present a data-driven recommender frameworkthat learns molecular substitution relationships from experimentallydocumented PFAS-to-non-PFAS pairings, and we demonstrate its viabilitythrough two case studies. In the first, a recommender approach groundedin known substitution chemistry and validated by first-principlesphysics is shown to accelerate the identification of over 500 potentialnonregrettable PFAS alternatives across diverse application domains.In the second, we consider substitutions with silicone molecules,performing ab initio calculations on 95 unique PFAS molecules thathave been replaced by silicones and 12 unique silicone molecules thathave served as replacements, and we then use our recommender engineto identify chemistries that are prime candidates for replacement.Together, these results demonstrate that a data-driven recommenderapproach can meaningfully navigate the vast space of possible PFASreplacements and surface chemically plausible, synthesizable candidatesfor experimental evaluation.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"13857–13869"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149281857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
AlphaFold3 Pseudo-Ensembles Capture β-Sheet Organization and Structural Diversity of Short Amyloidogenic Peptides AlphaFold3伪集成捕获β-薄片组织和短淀粉样蛋白肽的结构多样性
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-22 DOI: 10.1021/acs.jcim.6c02489
Francesca Peccati
{"title":"AlphaFold3 Pseudo-Ensembles Capture β-Sheet Organization and Structural Diversity of Short Amyloidogenic Peptides","authors":"Francesca Peccati","doi":"10.1021/acs.jcim.6c02489","DOIUrl":"10.1021/acs.jcim.6c02489","url":null,"abstract":"<p>Beyond predicting unique native globular protein structures,AlphaFoldhas been used to sample conformational diversity in globular proteins.Here, we asked whether this capacity can be transferred to amyloidfibrils formed by short peptides, despite the fundamental difficultiesposed by their polymorphism, sensitivity to environmental variablessuch as pH and cofactors, limited evolutionary constraints, and underrepresentationof amyloids in structural databases. Using a comprehensive set ofshort amyloidogenic peptides, we show that sets of AlphaFold3 (AF3)models (pseudo-ensembles) are accurate predictors of parallel versusantiparallel β-sheet organization, correctly assigning 112 ofthe 117 peptides for which a clear preference is predicted. Moreover,the simultaneous sampling of both organizations provides a signatureof structural diversity, as shown for the polymorphic peptides MVGGVVIAand LYIQWL. Modeling of the precise steric-zipper architecture isless reliable and biased toward compact arrangements, while predictionin the presence of cosolvents and metal ions does not reproduce solvent-mediatedinterfaces or noncanonical coordination geometries. Thus, althoughAF3 does not fully capture the physicochemical complexity of amyloidassemblies, its pseudo-ensembles provide a rapid and useful approachfor predicting β-sheet organization and identifying amyloidogenicsequences with potential structural diversity.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"14089–14101"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149229971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Insights into Composition-Induced Structural Shifts in Lipid Nanoparticles 洞察组成诱导的结构转变在脂质纳米颗粒
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-26 DOI: 10.1021/acs.jcim.6c02185
Rongmei Judy Wei, Shadi Fuladi, Doug J. Grzetic, Mayank Misra, Thomas F. Hughes, Dave J. Giesen, Febin Tom, Neha Chopra, Kunal Paul, Tati Reddy Vadicherla, Mee Y. Shelley, Andrea R. Browning, Naveen K. Thakral, Peter J. Skrdla, Shiva Sekharan, John C. Shelley
{"title":"Insights into Composition-Induced Structural Shifts in Lipid Nanoparticles","authors":"Rongmei Judy Wei,&nbsp;Shadi Fuladi,&nbsp;Doug J. Grzetic,&nbsp;Mayank Misra,&nbsp;Thomas F. Hughes,&nbsp;Dave J. Giesen,&nbsp;Febin Tom,&nbsp;Neha Chopra,&nbsp;Kunal Paul,&nbsp;Tati Reddy Vadicherla,&nbsp;Mee Y. Shelley,&nbsp;Andrea R. Browning,&nbsp;Naveen K. Thakral,&nbsp;Peter J. Skrdla,&nbsp;Shiva Sekharan,&nbsp;John C. Shelley","doi":"10.1021/acs.jcim.6c02185","DOIUrl":"10.1021/acs.jcim.6c02185","url":null,"abstract":"<p>Lipid nanoparticles(LNPs) are a leading platform fornucleic aciddelivery, yet current formulations typically suffer from rapid <em>in vivo</em> clearance and limited extrahepatic delivery. Recentexperiments demonstrate that increasing the helper lipid content (e.g.,to 40% 1,2-distearoyl-<em>sn</em>-glycero-3-phosphocholine[DSPC]) can significantly improve mRNA stability, prolong circulation,and enhance transfection efficiency. In this study, we used coarse-grainedmolecular dynamics simulations to provide mechanistic insight intothe structural changes driving these improvements. Our simulationsconfirm that an elevated helper lipid content fundamentally altersthe internal morphology of the LNP. Consistent with experimental observations,at high DSPC levels, the core transitions from a compact, disordered,predominantly lipid state to an extended structure with distinct lamellar-likeelements encompassing lipid-rich and water-rich regions. This structuralreorganization can be used to rationalize the enhanced stability anddelivery properties reported in the experiments. These findings providea molecular-level understanding of the composition-morphology relationshipand suggest that coarse-grained modeling can inform the design ofnovel LNP formulations.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"14200–14214"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149248653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
LESS-PseU: A Language Model-Enhanced Semistacking Framework for RNA Pseudouridine Site Prediction LESS-PseU:一种用于RNA伪尿嘧啶位点预测的语言模型增强半堆叠框架
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-27 DOI: 10.1021/acs.jcim.6c02817
Hao Yang, Zhiwei Cheng, Sisi He, Yong Zhang, Xiaolei Zhu
{"title":"LESS-PseU: A Language Model-Enhanced Semistacking Framework for RNA Pseudouridine Site Prediction","authors":"Hao Yang,&nbsp;Zhiwei Cheng,&nbsp;Sisi He,&nbsp;Yong Zhang,&nbsp;Xiaolei Zhu","doi":"10.1021/acs.jcim.6c02817","DOIUrl":"10.1021/acs.jcim.6c02817","url":null,"abstract":"<p>Accurate prediction of RNA pseudouridine (Ψ) sitesremainschallenging because existing models often struggle to integrate interpretablehandcrafted features with high-dimensional contextual representationsfrom pretrained genomic language models. To address this issue, wepropose LESS-PseU, a language model-enhanced semistacking frameworkthat fuses the predicted probability features of primary learnerswith DNABERT2 embeddings for multiple species Ψ site prediction.Specifically, we first built three primary learners based on threekinds of hand-crafted features, including one-hot, k-mer, and nucleotidechemical property features. Instead of training a primary learnerbased on the embeddings extracted from DNABERT2, the embeddings weredirectly concatenated with the predicted probabilities of the threeprimary learners as the input features for a Logistic Regression meta-learnerfor final prediction. Under a leakage-aware cross-validation setting,our model was validated on the training sets of three species. Furthermore,on independent test sets from human, mouse, and yeast, LESS-PseU achievedAUC values of 0.9069, 0.8833, and 0.8820, respectively, which werehigher than those of the compared models. A Friedman test indicatedan overall difference (<em>p</em> = 0.029). Paired DeLongtests further showed that the AUC improvements over Meta-PseU weresignificant for the human and mouse data sets, while those over PseU-STwere significant for all three species. Ablation analysis, fusion-strategycomparison, and pretrained language-model comparison further demonstratethe effectiveness of the proposed framework. Additional evaluationson m5C and m7G benchmarks provided preliminary evidence that the frameworkmay be applicable to other RNA modification prediction tasks.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"14144–14158"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149250821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Machine Learning Predictions of Experimental Mössbauer Properties of Fe-Containing Molecular Complexes and Protein Models 机器学习预测实验Mössbauer含铁分子复合物和蛋白质模型的性质
IF 6.4 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-09 Epub Date: 2026-10-28 DOI: 10.1021/acs.jcim.6c02183
Ioannis Skoulidas, Somayeh Tavasolikejani, Daisy Morgan, Jia-Min Chu, Yong Zhang
{"title":"Machine Learning Predictions of Experimental Mössbauer Properties of Fe-Containing Molecular Complexes and Protein Models","authors":"Ioannis Skoulidas,&nbsp;Somayeh Tavasolikejani,&nbsp;Daisy Morgan,&nbsp;Jia-Min Chu,&nbsp;Yong Zhang","doi":"10.1021/acs.jcim.6c02183","DOIUrl":"10.1021/acs.jcim.6c02183","url":null,"abstract":"<p>Mössbauer spectroscopy is among the best techniquesforprobing molecular structures of iron complexes and proteins. Extractingstructural insights from experimental spectra often necessitates accurateDFT predictions of isomer shift (IS) and quadrupole splitting (QS)parameters. However, substantial computational cost and sensitivityto the choice of functional and basis set limit their utility forhigh-throughput structural screening. To address this gap, varioustraditional machine learning (ML) and deep learning (DL) models wereinvestigated, using a curated experimental data set of 522 diversecompounds. After engineering models and descriptors, the DL modelChemprop has the highest coefficient of determination (<em>R</em><sup>2</sup>) for the test data set (0.92 for IS; 0.73 for QS), with <em>R</em><sup>2</sup> = 0.98/0.91 for IS/QS in the overall dataset. Validation of this model corroborates its high, close-to-DFTaccuracy for IS predictions, and also shows a significant improvementroom for QS predictions. Its predictions of several iron sites inboth heme and nonheme proteins and both first- and second-coordinationsphere models with challenging and/or unique novel structural featuresreplicated previous DFT structural predictions that are in excellentagreement with experiment, which is a significant application giventhat no protein data was used to train the ML model. This approachachieved a 218-fold reduction in computational time compared to DFTcalculations with same geometries. Results from this first ML workon Mössbauer data of molecular iron complexes will facilitaterapid structural screening previously impractical for large sets ofiron complexes and iron-protein binding site structures.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"66 21","pages":"13846–13856"},"PeriodicalIF":6.4,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149275053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Property-Defining Feature Extraction in Porous Crystalline Materials Using Attention-Guided Cooperative Game Theory 基于注意引导合作博弈论的多孔晶体材料属性定义特征提取
IF 5.6 2区 化学
Journal of Chemical Information and Modeling Pub Date : 2026-11-08 DOI: 10.1021/acs.jcim.6c02337
Kwabena Koranteng Asiedu,Luke E. K. Achenie,Nana Yaw Asiedu
{"title":"Property-Defining Feature Extraction in Porous Crystalline Materials Using Attention-Guided Cooperative Game Theory","authors":"Kwabena Koranteng Asiedu,Luke E. K. Achenie,Nana Yaw Asiedu","doi":"10.1021/acs.jcim.6c02337","DOIUrl":"https://doi.org/10.1021/acs.jcim.6c02337","url":null,"abstract":"Porous crystalline materials assembled from molecular building blocks offer vast potential across applications such as gas storage, separation, and catalysis, yet their structural tunability creates an expansive design space that demands the systematic identification of their property-defining features. To help researchers identify these features rapidly and inform next generation design, we introduce the Explainable Crystal Transformer (ExCT); an attention-guided cooperative game-theoretic workflow for extracting property-defining structural features in porous crystalline materials by combining attention-based representation learning, Shapley-style attribution, and local surrogate-fidelity screening. Designed for broader use across reticular frameworks, ExCT is demonstrated here with metal–organic frameworks (MOFs) as a representative case study of H2, CH4, and CO2 adsorption: cooperative game-theoretic attributions highlight chemically plausible local regions in selected hypothetical MOFs and qualitatively correspond to reported adsorption environments in experimentally studied cases, whereas attention identifies broader property-associated regions. Strict held-out surrogate-fidelity screening retains only a subset of high-fidelity explanations, more frequently for H2 than CH4 and less frequently for CO2.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"1 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149360537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书