Journal of Chemical Information and Modeling 最新文献_第10页

ConfRank+: Extending Conformer Ranking to Charged Molecules ConfRank+：将构象排序扩展到带电分子。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-07 DOI: 10.1021/acs.jcim.5c01259

Rick Oerder, Christian Hölzer and Jan Hamaekers*,

{"title":"ConfRank+: Extending Conformer Ranking to Charged Molecules","authors":"Rick Oerder, Christian Hölzer and Jan Hamaekers*, ","doi":"10.1021/acs.jcim.5c01259","DOIUrl":"10.1021/acs.jcim.5c01259","url":null,"abstract":"We present a machine learning model for high-throughput energetic ranking of charged molecular conformers. Based on the ConfRank (Hölzer et al. J. Chem. Inf. Model. 2024 64, 8909–8925) approach, the model is trained in a pairwise fashion to predict energy differences for pairs of conformers. By conditioning the model on data set embedding vectors, we are able to train our model on two different reference levels simultaneously, allowing for a larger training data set and to emulate multiple reference methods. In particular, we train our model on a large subset of the SPICE 2.0.1 data set with ωB97M-D3(BJ)/def2-TZVPPD range-separated hybrid meta-GGA DFT reference computations and a self-developed conformer data set based on the GEOM data set including r2SCAN-3c references. The result is a single multifidelity model that can reproduce both reference levels up to ML-typical model errors for small- and medium-sized molecules including the following elements: H, Li, B, C, N, O, F, Na, Mg, Si, P, S, Cl, K, Ca, Br, and I. By including partial atomic charges obtained from the electronegativity equilibration charge model, our model incorporates information about the charge distribution in a molecule, allowing the treatment of charged closed-shell species and explicit treatment of electrostatic interactions. We test the ranking capability of the model on various data sets, paying special attention to molecular charges of −1, 0, 1. Throughout all tests, we find our model to be as accurate as current AIMNet2 and MACE-OFF23(L) models, while requiring an order of magnitude fewer parameters and matching the robustness of the state-of-the-art semiempirical quantum method GFN2-xTB.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8664–8678"},"PeriodicalIF":5.3,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144792957","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

HMBVIP: A Novel Hierarchical Multi-Bio-View Intelligent Prediction Networks for Drug–Target Interaction Prediction HMBVIP：一种用于药物-靶标相互作用预测的新型分层多生物视图智能预测网络。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-07 DOI: 10.1021/acs.jcim.5c01142

Hailong Yang, Qiao Ning, Ze Song, Yue Chen, Guanjin Wang, Zhaohong Deng*, Yun Zuo*, Yuxi Ge* and Shudong Hu,

{"title":"HMBVIP: A Novel Hierarchical Multi-Bio-View Intelligent Prediction Networks for Drug–Target Interaction Prediction","authors":"Hailong Yang, Qiao Ning, Ze Song, Yue Chen, Guanjin Wang, Zhaohong Deng*, Yun Zuo*, Yuxi Ge* and Shudong Hu, ","doi":"10.1021/acs.jcim.5c01142","DOIUrl":"10.1021/acs.jcim.5c01142","url":null,"abstract":"Drug–target interaction (DTI) prediction is crucial in drug discovery. Recent advances in multiview learning have made it possible to automatically extract complex features from multiple perspectives. Multiview models, which integrate diverse biological data sources, have demonstrated improved prediction accuracy and robustness. However, current approaches still face major limitations: (1) reliance on single-scale sequence tokenizers that fail to capture biological information across different granularities and (2) shallow, single-layer integration of data views that overlook the hierarchical nature of biological systems. To tackle these challenges, we propose the concept of “bio-token” and design a multiscale biological tokenizer that captures biological features at varying resolutions. We also introduce a novel hierarchical multi-bio-view learning (HMBV) approach, implemented in an end-to-end DTI prediction network termed HMBVIP. The hierarchical multiview fusion enriches hidden representations with multidimensional biological context, thereby enhancing both prediction accuracy and biologically meaningful interpretability. The results on benchmark data sets demonstrate that HMBVIP consistently outperforms current state-of-the-art models.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8871–8888"},"PeriodicalIF":5.3,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144792188","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

MPLBind: Predicting the Effect of Binding Site Point Mutations on Protein–Ligand Binding Affinity Using Protein Large Language Models MPLBind：利用蛋白质大语言模型预测结合位点点突变对蛋白质-配体结合亲和力的影响。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-07 DOI: 10.1021/acs.jcim.5c00838

Jiahuan Jin, and , Jie Li*,

{"title":"MPLBind: Predicting the Effect of Binding Site Point Mutations on Protein–Ligand Binding Affinity Using Protein Large Language Models","authors":"Jiahuan Jin,  and , Jie Li*, ","doi":"10.1021/acs.jcim.5c00838","DOIUrl":"10.1021/acs.jcim.5c00838","url":null,"abstract":"Protein–ligand binding affinity may be affected by protein mutations, especially point mutations occurring within the binding site, which may indirectly contribute to interindividual differences in drug response. In recent years, several methods have been proposed to predict the effect of binding site mutations on protein–ligand binding affinity. However, the impact of mutations is difficult to predict accurately. In this study, a method named MPLBind is proposed to predict the effect of mutations on protein–ligand binding affinity, which effectively utilizes ligand descriptors and fingerprints, mutant residues’ local environment changes, and large protein language model features, which contain context, evolutionary information, conservation, and functional information on protein sequences. The use of the large protein language model and the fusion strategy of ligand and mutation features significantly improved the prediction performance. Experimental results show that MPLBind has better performance against competing baseline models, not only in predicting protein–ligand binding affinity but also in predicting the effect of mutations on protein–ligand binding affinity.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8720–8729"},"PeriodicalIF":5.3,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144792193","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

Rescaling of Point Charges as a Way to Improve the Simple-to-Use Electrostatic Embedding Scheme Developed to Explore Enzyme Activity with QM-Oriented Software 点电荷的重新缩放作为一种改进简单易用的静电嵌入方案的方法，开发了用于探索酶活性的qm导向软件。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-07 DOI: 10.1021/acs.jcim.5c01235

Andrzej J. Kałka, Aleš Novotný and Jernej Stare*,

{"title":"Rescaling of Point Charges as a Way to Improve the Simple-to-Use Electrostatic Embedding Scheme Developed to Explore Enzyme Activity with QM-Oriented Software","authors":"Andrzej J. Kałka, Aleš Novotný and Jernej Stare*, ","doi":"10.1021/acs.jcim.5c01235","DOIUrl":"10.1021/acs.jcim.5c01235","url":null,"abstract":"Computer-aided exploration of enzymatic reactions, which still leaves many important questions open, calls for robust and accurate techniques of molecular modeling. One of the most intriguing issues related to enzymatic reactions is the role of electrostatic interactions established between the reacting moiety and its enzymatic environment. In order to evaluate these interactions, we previously devised a QM/MM scheme based on electrostatic embedding of the reaction kernel, treated by quantum chemistry, into the enzymatic surroundings represented by point charges [A. Prah et al., ACS Catal. 2019, 9, 1231.]. The method features remarkable simplicity and reliably predicts the effect of electrostatics on enzyme catalysis. Yet, this simplified approach has pitfalls; in particular, it tends to overestimate the attracting force between the electrons and the surrounding point charges─an effect named electron spill-out─impairing the accuracy of evaluated electrostatic interactions. Herein, by using statistical methods together with reference quantum calculations, we critically assess the impact of this pitfall and propose a very simple but effective correction based on attenuation of point charges near the QM–MM boundary depending on their distance from the quantum subsystem. We demonstrate that the proposed correction can significantly improve the accuracy of computed energies of electrostatic interactions between the reaction kernel and its enzyme surroundings, thereby representing an important methodological advance of our electrostatic embedding approach. Noteworthily, the optimal attenuation scheme can vary among the considered systems─in particular, it is sensitive to the net charge of the reaction kernel─suggesting the scheme be tuned individually for each considered enzymatic reaction following the presented workflow.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8653–8663"},"PeriodicalIF":5.3,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.jcim.5c01235","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144796719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Autogenerating a Domain-Specific Question-Answering Data Set from a Thermoelectric Materials Database to Enable High-Performing BERT Models 从热电材料数据库自动生成特定领域的问答数据集，以实现高性能BERT模型。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-07 DOI: 10.1021/acs.jcim.5c00840

Odysseas Sierepeklis, and , Jacqueline M. Cole*,

{"title":"Autogenerating a Domain-Specific Question-Answering Data Set from a Thermoelectric Materials Database to Enable High-Performing BERT Models","authors":"Odysseas Sierepeklis,  and , Jacqueline M. Cole*, ","doi":"10.1021/acs.jcim.5c00840","DOIUrl":"10.1021/acs.jcim.5c00840","url":null,"abstract":"We present a method for autogenerating a large domain-specific question-answering (QA) dataset from a thermoelectric materials database. We show that a small language model, BERT, once fine-tuned on this automatically generated dataset of 99,757 QA pairs about thermoelectric materials, affords better performance in the field of thermoelectric materials compared to a BERT model fine-tuned on the generic English-language QA data set, SQuAD-v2. We further show that mixing the two data sets (ours and SQuAD-v2), which have significantly different syntactic and semantic scopes, allows the BERT model to achieve even better performance. The best-performing BERT model fine-tuned on the mixed data set outperforms the models fine-tuned on the other two data sets by scoring an exact match of 67.93% and an F1 score of 72.29% when evaluated on our test data set. This has important implications as it demonstrates the ability to realize high-performing small language models, with modest computational resources, empowered by domain-specific materials data sets which can be generated according to our method.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8579–8592"},"PeriodicalIF":5.3,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.jcim.5c00840","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144796745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrating Physics-Based Simulations with Data-Driven Deep Learning Represents a Robust Strategy for Developing Inhibitors Targeting the Main Protease 将基于物理的模拟与数据驱动的深度学习相结合，为开发针对主要蛋白酶的抑制剂提供了一种强大的策略。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-06 DOI: 10.1021/acs.jcim.5c01307

Yanqing Yang, Yangwei Jiang, Dong Zhang, Leili Zhang* and Ruhong Zhou*,

引用次数: 0

Decoding Protein Stabilization: Impact on Aggregation, Solubility, and Unfolding Mechanisms 解码蛋白质稳定性：对聚集、溶解度和展开机制的影响。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-06 DOI: 10.1021/acs.jcim.5c00611

Martin Havlásek, Sérgio M. Marques, Veronika Szotkowská, Antonín Kunka, Petra Babková, Jiří Damborský, Zbyněk Prokop* and David Bednář*,

{"title":"Decoding Protein Stabilization: Impact on Aggregation, Solubility, and Unfolding Mechanisms","authors":"Martin Havlásek, Sérgio M. Marques, Veronika Szotkowská, Antonín Kunka, Petra Babková, Jiří Damborský, Zbyněk Prokop* and David Bednář*, ","doi":"10.1021/acs.jcim.5c00611","DOIUrl":"10.1021/acs.jcim.5c00611","url":null,"abstract":"Modern computational tools can predict the mutational effects on protein stability, sometimes at the expense of activity or solubility. Here, we investigate two homologous computationally stabilized haloalkane dehalogenases: (i) the soluble thermostable DhaA115 (Tmapp = 74 °C) and (ii) the poorly soluble and aggregating thermostable LinB116 (Tmapp = 65 °C), together with their respective wild-type variants. The intriguing difference in the solubility of these highly homologous proteins has remained unexplained for three decades. We combined experimental and in-silico techniques and examined the effects of stabilization on solubility and aggregation propensity. A detailed analysis of the unfolding mechanisms in the context of aggregation explained the negative consequences of stabilization observed in LinB116. With the aid of molecular dynamics simulations, we identified regions exposed during the unfolding of LinB116 that were later found to exhibit aggregation propensity. Our analysis identified cryptic aggregation-prone regions and increased surface hydrophobicity as key factors contributing to the reduced solubility of LinB116. This study reveals novel molecular mechanisms of unfolding for hyperstabilized dehalogenases and highlights the importance of contextual information in protein engineering to avoid the negative effects of stabilizing mutations on protein solubility.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8688–8701"},"PeriodicalIF":5.3,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.jcim.5c00611","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144787026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy Landscapes and Structural Plasticity of Intrinsically Disordered Histones 内在无序组蛋白的能量景观和结构可塑性。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-06 DOI: 10.1021/acs.jcim.4c02269

Rafael G. Viegas, Hao Wu, Murilo N. Sanches, Garegin A. Papoian and Vitor B.P. Leite*,

{"title":"Energy Landscapes and Structural Plasticity of Intrinsically Disordered Histones","authors":"Rafael G. Viegas, Hao Wu, Murilo N. Sanches, Garegin A. Papoian and Vitor B.P. Leite*, ","doi":"10.1021/acs.jcim.4c02269","DOIUrl":"10.1021/acs.jcim.4c02269","url":null,"abstract":"Intrinsically disordered proteins (IDPs) are characterized by their lack of a stable 3D structure, enabling them to adopt multiple conformations and participate in various cellular processes. This study investigates the conformational dynamics of histone tails, specifically the H4 tail and the linker histone H1, focusing on the effects of post-translational modifications (PTMs) such as acetylation. Utilizing the energy landscape visualization method (ELViM), we projected the conformational space of wild type and acetylated forms of the H4 tail, revealing significant insights into their structural heterogeneity and preferential ensembles. This approach demonstrated that acetylation reduces the conformational heterogeneity of the H4 tail and introduces regions within the conformational space uniquely occupied by each form, which may correlate with specific biological functions. Furthermore, the conformational space of the linker histone H1 was analyzed, illustrating how its structural heterogeneity is influenced by nucleosome binding modes. This work highlights the critical role of conformational plasticity and PTMs in regulating the multifunctionality of IDPs, thereby enhancing our understanding of their contributions to chromatin dynamics and cellular regulation.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8679–8687"},"PeriodicalIF":5.3,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.jcim.4c02269","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144787027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ReaxANA: Analysis of Reactive Dynamics Trajectories for Reaction Network Generation ReaxANA：反应网络生成的反应动力学轨迹分析。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-06 DOI: 10.1021/acs.jcim.5c00521

Hong Zhu, Xin Chen* and Jiali Gao*,

{"title":"ReaxANA: Analysis of Reactive Dynamics Trajectories for Reaction Network Generation","authors":"Hong Zhu, Xin Chen* and Jiali Gao*, ","doi":"10.1021/acs.jcim.5c00521","DOIUrl":"10.1021/acs.jcim.5c00521","url":null,"abstract":"In reactive molecular dynamics (MD) simulations, such as those used to model combustion, filtering noisy data from reactive trajectories is crucial for accurately constructing reaction networks and elucidating macroscopic mechanisms. To address this challenge, we introduce a graph algorithm-based explicit denoising approach that defines user-controlled operations for removing oscillatory reaction patterns, including combination and separation, isomerization, and node contraction. This algorithm is implemented in ReaxANA, a parallel Python package designed to extract reaction mechanisms from both heterogeneous and homogeneous reactive MD trajectories. ReaxANA operates solely on atomic position data, enabling its easy integration with various simulation platforms. We demonstrate its capabilities through the analysis of the TNT (trinitrotoluene) explosion system generated by using molecular dynamics simulations with the ReaxFF force field. ReaxANA effectively distinguishes structural isomers, facilitating a comprehensive examination of reaction networks. Our findings reveal that the primary decomposition pathway of TNT involves pyrolysis of the ortho nitro group (-NO2), followed by further decomposition that leads to a five-membered ring compound. ReaxANA is an open-source software and packaged in a Docker container for cross-platform compatibility, providing insights and advanced analytical capabilities.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8549–8562"},"PeriodicalIF":5.3,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144787029","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

Site-of-Metabolism Prediction with Aleatoric and Epistemic Uncertainty Quantification 代谢位点预测与任意和认知不确定性量化。

IF 5.3 2区化学

Journal of Chemical Information and Modeling Pub Date : 2025-08-06 DOI: 10.1021/acs.jcim.5c00762

Roxane Axel Jacob, Oliver Wieder, Ya Chen, Angelica Mazzolari, Andreas Bergner, Klaus-Juergen Schleifer and Johannes Kirchmair*,

{"title":"Site-of-Metabolism Prediction with Aleatoric and Epistemic Uncertainty Quantification","authors":"Roxane Axel Jacob, Oliver Wieder, Ya Chen, Angelica Mazzolari, Andreas Bergner, Klaus-Juergen Schleifer and Johannes Kirchmair*, ","doi":"10.1021/acs.jcim.5c00762","DOIUrl":"10.1021/acs.jcim.5c00762","url":null,"abstract":"In silico metabolism prediction models have become indispensable tools to optimize the metabolic properties of xenobiotics while preserving their intended biological activity. Among these, site-of-metabolism (SOM) prediction models are particularly valuable for pinpointing metabolically labile atomic positions. However, the practical utility of these models depends not only on their ability to deliver accurate predictions but also on their capacity to provide reliable estimates of predictive uncertainty. In this work, we introduce aweSOM, a graph neural network (GNN)-based SOM prediction model that leverages deep ensembling to model the total predictive accuracy and partition it into its aleatoric and epistemic components. We conduct a comprehensive evaluation of aweSOM’s uncertainty estimates on a high-quality data set, identifying key challenges that currently constrain the performance of SOM prediction models. Based on these findings, we propose actionable insights to drive progress in the field of metabolism prediction.","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 16","pages":"8462–8474"},"PeriodicalIF":5.3,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/pdf/10.1021/acs.jcim.5c00762","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144787391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0