Journal of Chemical Information and Modeling 最新文献

{"title":"Prediction of Chromatographic Retention Time of a Small Molecule from SMILES Representation Using a Hybrid Transformer-LSTM Model.","authors":"Sargol Mazraedoost, Hadi Sedigh Malekroodi, Petar Žuvela, Myunggi Yi, J Jay Liu","doi":"10.1021/acs.jcim.5c00167","DOIUrl":"https://doi.org/10.1021/acs.jcim.5c00167","url":null,"abstract":"<p><p>Accurate retention time (RT) prediction in liquid chromatography remains a significant consideration in molecular analysis. In this study, we explore the use of a transformer-based language model to predict RTs by treating simplified molecular input line entry system (SMILES) sequences as textual input, an approach that has not been previously utilized in this field. Our architecture combines a pretrained RoBERTa (robustly optimized BERT approach, a variant of BERT) with bidirectional long short-term memory (BiLSTM) networks to predict retention times in reversed-phase high-performance liquid chromatography (RP-HPLC). The METLIN small molecule retention time (SMRT) data set comprising 77,980 small molecules after preprocessing, was encoded using SMILES notation and processed through a tokenizer to enable molecular representation as sequential data. The proposed transformer-LSTM architecture incorporates layer fusion from multiple transformer layers and bidirectional sequence processing, achieving superior performance compared to existing methods with a mean absolute error (MAE) of 26.23 s, a mean absolute percentage error (MAPE) of 3.25%, and <i>R</i>-squared (<i>R</i>²) value of 0.91. The model's explainability was demonstrated through attention visualization, revealing its focus on key molecular features that can influence RT. Furthermore, we evaluated the model's transfer learning capabilities across ten data sets from the PredRet database, demonstrating robust performance across different chromatographic conditions with consistent improvement over previous approaches. Our results suggest that the hybrid model presents a valuable approach for predicting RT in liquid chromatography, with potential applications in metabolomics and small molecule analysis.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143727032","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}

{"title":"Recent Advances in the Modeling of Ionic Liquids Using Artificial Neural Networks.","authors":"Adrian Racki, Kamil Paduszyński","doi":"10.1021/acs.jcim.4c02364","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02364","url":null,"abstract":"<p><p>This paper reviews the recent and most impactful advancements in the application of artificial neural networks in modeling the properties of ionic liquids. As salts that are liquid at temperatures below 100 °C, ionic liquids possess unique properties beneficial for various industrial applications such as carbon capture, catalytic solvents, and lubricant additives. The study emphasizes the challenges in selecting appropriate ILs due to the vast variability in their properties, which depend significantly on their cation and anion structures. The review discusses the advantages of using ANNs, including feed-forward, cascade-forward, convolutional, recurrent, and graph neural networks, over traditional machine learning algorithms for predicting the thermodynamic and physical properties of ILs. The paper also highlights the importance of data preparation, including data collection, feature engineering, and data cleaning, in developing accurate predictive models. Additionally, the review covers the interpretability of these models using techniques such as SHapley Additive exPlanations to understand feature importance. The authors conclude by discussing future opportunities and the potential of combining ANNs with other computational methods to design new ILs with targeted properties.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143717627","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}

{"title":"Stacking Interactions of Druglike Heterocycles with Nucleobases.","authors":"Audrey V Conner, Lauren M Kim, Patrick A Fagan, Drew P Harding, Steven E Wheeler","doi":"10.1021/acs.jcim.4c02420","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02420","url":null,"abstract":"<p><p>Stacking interactions contribute significantly to the interaction of small molecules with RNA, and harnessing the power of these interactions will likely prove important in the development of RNA-targeting inhibitors. To this end, we present a comprehensive computational analysis of stacking interactions between a set of 54 druglike heterocycles and the natural nucleobases. We first show that heterocycle choice can tune the strength of stacking interactions with nucleobases over a large range and that heterocycles favor stacked geometries that cluster around a discrete set of stacking loci characteristic of each nucleobase. Symmetry-adapted perturbation theory results indicate that the strengths of these interactions are modulated primarily by electrostatic and dispersion effects. Based on this, we present a multivariate predictive model of the maximum strength of stacking interactions between a given heterocycle and nucleobase that depends on molecular descriptors derived from the electrostatic potential. These descriptors can be readily computed using density functional theory or predicted directly from atom connectivity (e.g., SMILES). This model is used to predict the maximum possible stacking interactions of a set of 1854 druglike heterocycles with the natural nucleobases. Finally, we show that trivial modifications of standard (fixed-charge) molecular mechanics force fields reduce errors in predicted stacking interaction energies from around 2 kcal/mol to below 1 kcal/mol, providing a pragmatic means of predicting more reliable stacking interaction energies using existing computational workflows. We also analyze the stacking interactions between ribocil and a bacterial riboswitch, showing that two of the three aromatic heterocyclic components engage in near-optimal stacking interactions with binding site nucleobases.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143727083","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}

{"title":"MultiCTox: Empowering Accurate Cardiotoxicity Prediction through Adaptive Multimodal Learning.","authors":"Lin Feng, Xiangzheng Fu, Zhenya Du, Yuting Guo, Linlin Zhuo, Yan Yang, Dongsheng Cao, Xiaojun Yao","doi":"10.1021/acs.jcim.5c00022","DOIUrl":"https://doi.org/10.1021/acs.jcim.5c00022","url":null,"abstract":"<p><p>Cardiotoxicity refers to the inhibitory effects of drugs on cardiac ion channels. Accurate prediction of cardiotoxicity is crucial yet challenging, as it directly impacts the evaluation of cardiac drug efficacy and safety. Numerous methods have been developed to predict cardiotoxicity, yet their performance remains limited. A key limitation is that these methods often rely solely on single-modal data, making multimodal data integration challenging. As a result, we present a multimodal method integrating molecular SMILES, structure, and fingerprint to enhance cardiotoxicity prediction. First, we designed a fusion layer to unify representations from different modalities. During training, the model maximizes intramodal similarity for the same molecule while minimizing intermolecular similarity, ensuring consistent cross-modal representations. This study evaluates the inhibitory effects of candidate drugs on voltage-gated potassium (hERG), sodium (Nav1.5), and calcium (Cav1.2) channels. Experimental results demonstrate that the proposed model significantly outperforms existing state-of-the-art methods in cardiotoxicity prediction. We anticipate that this model will contribute significantly to the development and safety evaluation of cardiac drugs, reducing cardiotoxicity-related risks.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143717620","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}

{"title":"Structural Descriptors for Subunit Interface Regions in Homodimers: Effect of Lipid Membrane and Secondary Structure Type.","authors":"Aslı Yüksek, Batuhan Yıkınç, İrem Nayır, Defne Alnıgeniş, Vahap Gazi Fidan, Tayyip Topuz, Ebru Demet Akten","doi":"10.1021/acs.jcim.4c01233","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01233","url":null,"abstract":"<p><p>A total of 1311 homodimers were collected and analyzed in three different categories to highlight the impact of lipid environment and secondary structure type: 422 cytoplasmic α-helix, 411 cytoplasmic β-strand, and 478 membrane complexes. Structural features of the interface connecting two monomers were investigated and compared to those of the non-interface surface. Every residue on the surface of each monomer was explored based on four attributes: solvent-accessible surface area (SASA), protrusion index (C_<i>x</i>), surface planarity, and surface roughness. SASA and C_<i>x</i> distribution profiles clearly distinguished the interface from the surface in all categories, where the rim of the interface displayed higher SASA and C_<i>x</i> values than the rest of the surface. Surface residues in membrane complexes protruded less than cytoplasmic ones due to the hydrophobic environment, and consequently, the difference between surface and interface residues became less noticeable in that category. Cytoplasmic β-strand complexes displayed markedly lower SASA at the interface core than at the surface. The major distinction between the surface and interface was achieved through surface roughness, which displayed significantly higher values for the interface than the surface, especially in cytoplasmic complexes. Clearly, a surface which is relatively rugged favors the association of two monomers through multiple van der Waals interactions and hydrogen-bond formations. Another structural descriptor with strong distinguishing ability was surface planarity, which was higher at the interface than at the non-interface surface. Surface flatness would eventually facilitate the interconnectedness of an interface with a network of residue pairs bridging two complementary surfaces. Analysis of contact pairs revealed that hydrophobic pairs have the highest frequency of occurrence in the lipid environment of membrane complexes. However, despite the scarcity of polar residues at the interface, the likelihood of observing a contact between polar residues was markedly higher than that of hydrophobic ones.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143717632","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}

{"title":"Energetics of Expanded PAM Readability by Engineered Cas9-NG.","authors":"Shreya Bhattacharya, Priyadarshi Satpati","doi":"10.1021/acs.jcim.5c00011","DOIUrl":"https://doi.org/10.1021/acs.jcim.5c00011","url":null,"abstract":"<p><p>The energetic basis for the enhanced PAM (protospacer adjacent motif) readability in engineered Cas9-NG (a variant of Cas9 from <i>Streptococcus pyogenes</i> (<i>Sp</i>Cas9)) with seven mutations: (R1335V, E1219F, D1135V, L1111R, T1337R, G1218R, and A1322R) remains a fundamental unsolved problem. Utilizing the X-ray structure of the precatalytic complex (<i>Sp</i>Cas9:sgRNA:dsDNA) as a template, we calculated the changes in PAM (TGG, TGA, TGT, or TGC) binding affinity (ΔΔ<i>G</i>) associated with each of the seven mutations in <i>Sp</i>Cas9 through rigorous alchemical simulations (sampling ∼ 53 μs). The underlying thermodynamics (ΔΔ<i>G</i>) accounts for the experimentally observed differences in DNA cleavage activity between <i>Sp</i>Cas9 and Cas9-NG across various DNA substrates. The interaction energies between <i>Sp</i>Cas9 and DNA are significantly influenced by the type and location of the amino acid mutations. Notably, the R1335V mutation disfavors DNA binding by disrupting critical interactions with the PAM. However, the destabilizing effect of the R1335V mutation is mitigated by four advantageous mutations (E1219F, D1135V, L1111R, and T1337R), which primarily introduce nonbase-specific interactions and enhance PAM readability. The hydrophobic substitutions (E1219F and D1135V) are particularly impactful, as they exclude solvent from the PAM binding pocket, strengthening electrostatic interactions in the low dielectric medium and increasing the stability of the noncognate PAM complexes by ∼2-5 kcal/mol. Additionally, L1111R and T1337R facilitate DNA binding by forming direct electrostatic contacts. In contrast, the charge mutations G1218R and A1322R do not effectively promote interactions with the negatively charged DNA, clearly demonstrating that the location of mutations is crucial in shaping these interaction energetics. We demonstrated that stabilization of the Cas9-NG: noncognate PAM complexes enables broader PAM recognition. This is primarily achieved through two mechanisms: (1) the establishment of new nonbase-specific interactions between the protein and nucleotides and (2) the enhancement of electrostatic interactions within a relatively dry and hydrophobic pocket. The findings revealed that mutation-induced desolvation can improve the recognition of noncognate PAMs, paving the way for the rational and innovative design of <i>Sp</i>Cas9 mutants.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143717614","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}

{"title":"DeePMD-GNN: A DeePMD-kit Plugin for External Graph Neural Network Potentials.","authors":"Jinzhe Zeng, Timothy J Giese, Duo Zhang, Han Wang, Darrin M York","doi":"10.1021/acs.jcim.4c02441","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02441","url":null,"abstract":"<p><p>Machine learning potentials (MLPs) have revolutionized molecular simulation by providing efficient and accurate models for predicting atomic interactions. MLPs continue to advance and have had profound impact in applications that include drug discovery, enzyme catalysis, and materials design. The current landscape of MLP software presents challenges due to the limited interoperability between packages, which can lead to inconsistent benchmarking practices and necessitates separate interfaces with molecular dynamics (MD) software. To address these issues, we present DeePMD-GNN, a plugin for the DeePMD-kit framework that extends its capabilities to support external graph neural network (GNN) potentials.DeePMD-GNN enables the seamless integration of popular GNN-based models, such as NequIP and MACE, within the DeePMD-kit ecosystem. Furthermore, the new software infrastructure allows GNN models to be used within combined quantum mechanical/molecular mechanical (QM/MM) applications using the range corrected ΔMLP formalism.We demonstrate the application of DeePMD-GNN by performing benchmark calculations of NequIP, MACE, and DPA-2 models developed under consistent training conditions to ensure fair comparison.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143727026","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}

{"title":"MartiniGlass: a Tool for Enabling Visualization of Coarse-Grained Martini Topologies.","authors":"Christopher Brasnett, Siewert J Marrink","doi":"10.1021/acs.jcim.4c02277","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02277","url":null,"abstract":"<p><p>As molecular modeling gains ever more prominence in understanding cellular processes, high quality visualization of models and dynamics has never been more important. Naturally, much molecular visualization software is written to enable the visualization of atomic level details in structures. While necessary, this means that visualization of increasingly popular coarse-grained (CG) models remains a challenge. Here, we present a Python package, MartiniGlass, that facilitates the visualization of systems simulated with the widely used CG Martini force field using the popular visualization package VMD. MartiniGlass rapidly processes molecular topologies and accounts for important topological features at CG resolution, such as secondary structure restraints, preparing them for easy visualization of simulated trajectories.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143727030","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}

{"title":"Knowledge-Based Artificial Intelligence System for Drug Prioritization.","authors":"Yinchun Su, Jiashuo Wu, Xilong Zhao, Yue Hao, Ziyi Wang, Yongbao Zhang, Yujie Tang, Bingyue Pan, Guangyou Wang, Qingfei Kong, Junwei Han","doi":"10.1021/acs.jcim.5c00027","DOIUrl":"https://doi.org/10.1021/acs.jcim.5c00027","url":null,"abstract":"<p><p><i>In silico</i> drug prioritization may be a promising and time-saving strategy to identify potential drugs, standing as a faster and more cost-effective approach than <i>de novo</i> approaches. In recent years, artificial intelligence has greatly evolved the drug development process. Here, we present a novel computational framework for drug prioritization, <i>labyrinth</i>, designed to simulate human knowledge retrieval and inference to identify potential drug candidates for each disease. With the integration of up-to-date clinical trials, literature co-occurrences, drug-target interactions, and disease similarities, our framework achieves over 90% predictive accuracy across clinical trial phases and strong alignment with clinical practice in TCGA cohorts. We have demonstrated effectiveness across 20 different disease categories with robust ROC-AUC metrics and the balance between predictive accuracy and model interpretability. We further demonstrate its effectiveness at both the population and the individual levels. This study not only demonstrates the capacity for its drug prioritization but underscores the importance of aligning computational models with intuitive human reasoning. We have wrapped the core function into an R package named <i>labyrinth</i>, which is freely available on GitHub under the GPL-v2 license (https://github.com/hanjunwei-lab/labyrinth).</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143727029","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}

{"title":"Investigation of Electrostatic Effects on Enyzme Catalysis: Insights from Computational Simulations of Monoamine Oxidase A Pathological Variants Leading to the Brunner Syndrome.","authors":"Martina Rajić, Jernej Stare","doi":"10.1021/acs.jcim.4c01698","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01698","url":null,"abstract":"<p><p>Brunner syndrome is a rare genetic disorder characterized by impulsive aggressiveness and intellectual disability, which is linked to impaired function of the monoamine oxidase A (MAO-A) enzyme. Patients with specific point mutations in the <i>MAOA</i> gene have been reported to exhibit these symptoms, along with notably elevated serotonin levels, which suggest a decreased catalytic performance of the mutated MAO-A enzymes. In this study, we present multiscale molecular simulations focusing on the rate-limiting step of MAO-A-catalyzed serotonin degradation for the C266F and V244I variants that are reportedly associated with pathologies characteristic of the Brunner syndrome. We found that the C266F mutation causes an approximately 18,000-fold slowdown of enzymatic function, which is equivalent to a <i>MAOA</i> gene knockout. For the V244I mutant, a somewhat smaller, yet still significant 300-fold slowdown has been estimated. Furthermore, we conducted a comprehensive comparison of the impact of enzyme electrostatics on the catalytic function of the wild-type (WT) MAO-A and both aforementioned mutants (C266F and V244I), as well as on the E446K mutant investigated in one of our earlier studies. The results have shown that the mutation induces a noteworthy change in electrostatic interactions between the reacting moiety and its enzymatic surroundings, leading to a decreased catalytic performance in all of the considered MAO-A variants. An analysis of mutation effects supported by geometry comparison of mutants and the wild-type enzyme at a residue level suggests that a principal driving force behind the altered catalytic performance of the mutants is subtle structural changes scattered along the entire enzyme. These shifts in geometry also affect domains most relevant to catalysis, where structural offsets of few tenths of an Å can significantly change contribution to the barrier of the involved residues. These results are in full agreement with the reasoning derived from clinical observations and biochemical data. Our research represents a step forward in the attempts of using fundamental principles of chemical physics in order to explain genetically driven pathologies. In addition, our results support the view that the catalytic function of enzymes is crucially driven by electrostatic interactions.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143707738","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}