Shan Lu, Nicholas J Huls, Koushiki Basu, Tonglei Li
{"title":"Deep Learning of CYP450 Binding of Small Molecules by Quantum Information.","authors":"Shan Lu, Nicholas J Huls, Koushiki Basu, Tonglei Li","doi":"10.1021/acs.jcim.4c01735","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01735","url":null,"abstract":"<p><p>Drug-drug interaction can lead to diminished therapeutic effects or increased toxicity, posing significant risks, especially in polypharmacy, and cytochrome P450 plays an indispensable role in this interaction. Cytochrome P450, responsible for the metabolism and detoxification of most drugs, metabolizes about 90% of Food and Drug Administration-approved drugs, making early detection of potential drug-drug interactions. Over the years, in-silico modeling has become a valuable tool for predicting drug-drug interactions. Still, conventional molecular descriptors focusing on the structural properties of drugs often overlook complex electronic interactions critical for accurate predictions. To address this, we implemented the Manifold Embedding of Molecular Surface (MEMS) approach, which retains the quantum mechanical characteristics of molecules. MEMS-generated electronic attributes were embedded and featurized for deep learning using the DeepSets architecture, where our models achieved high accuracy, particularly for cytochrome P450 enzyme 1A2 (CYP1A2), with F1 scores reaching up to 0.866. This study highlights the potential of integrating detailed electronic properties with deep learning to improve predictive models for drug-drug interactions, addressing the limitations of traditional molecular descriptors and machine-learning techniques.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143044927","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}
Rui Zhang, Dylan Ye, Anit Gurung, Ralf Warmuth, Daniel G Kuroda, Lu Wang
{"title":"p<i>K</i><sub>a</sub> Matching Enables Quantum Proton Delocalization in Acid-1-Methylimidazole Binary Mixtures.","authors":"Rui Zhang, Dylan Ye, Anit Gurung, Ralf Warmuth, Daniel G Kuroda, Lu Wang","doi":"10.1021/acs.jcim.4c02187","DOIUrl":"10.1021/acs.jcim.4c02187","url":null,"abstract":"<p><p>Short hydrogen bonds (SHBs), characterized by donor-acceptor heteroatom separations below 2.7 Å, are prevalent in condensed-phase systems. Recently, we identified SHBs in nonaqueous binary mixtures of acetic acid and 1-methylimidazole (MIm), where electronic and nuclear quantum effects facilitate extensive proton delocalization. In this work, we explore the conditions favoring SHB formation in binary acid-base mixtures and propose that the difference in p<i>K</i><sub>a</sub> values between the acid and base, measured in a nonaqueous, aprotic solvent like DMSO, is a key determinant. Using MIm as a model base, we perform electronic structure calculations to systematically analyze p<i>K</i><sub>a</sub> matching across 97 acid-MIm pairs in DMSO solutions. Through a combination of first-principles simulations and infrared spectroscopy, we confirm the formation of SHBs and the delocalization of protons in benzoic acid-MIm and salicylic acid-MIm binary mixtures. Our results demonstrate that p<i>K</i><sub>a</sub> matching can significantly alter proton behavior in nonaqueous systems, transforming acid-base interactions from conventional proton transfer to quantum mechanical proton delocalization. This work establishes DMSO as a valuable alternative to water for assessing p<i>K</i><sub>a</sub> matching and highlights the importance of hydrogen bond networks in modulating these conditions. By elucidating the impact of electronic and nuclear quantum effects, our results provides insights for designing organic mixtures that leverage SHBs for advanced material applications.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"798-810"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142941368","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}
Mehrnoosh Khodam Hazrati, Lukáš Sukeník, Robert Vácha
{"title":"Split Membrane: A New Model to Accelerate All-Atom MD Simulation of Phospholipid Bilayers.","authors":"Mehrnoosh Khodam Hazrati, Lukáš Sukeník, Robert Vácha","doi":"10.1021/acs.jcim.4c01664","DOIUrl":"10.1021/acs.jcim.4c01664","url":null,"abstract":"<p><p>All-atom molecular dynamics simulations are powerful tools for studying cell membranes and their interactions with proteins and other molecules. However, these processes occur on time scales determined by the diffusion rate of phospholipids, which are challenging to achieve in all-atom models. Here, we present a new all-atom model that accelerates lipid diffusion by splitting phospholipid molecules into head and tail groups. The bilayer structure is maintained by using external lateral potentials, which compensate for the lipid split. This split model enhances lateral lipid diffusion more than ten times, allowing faster and cheaper equilibration of large systems with different phospholipid types. The current model has been tested on membranes containing PSM, POPC, POPS, POPE, POPA, and cholesterol. We have also evaluated the interaction of the split model membranes with the Disheveled DEP domain and amphiphilic helix motif of the transcriptional repressor Opi1 as representative of peripheral proteins as well as the dimeric fragment of the epidermal growth factor receptor transmembrane domain and the Human A2A Adenosine of G protein-coupled receptors as representative of transmembrane proteins. The split model can predict the interaction sites of proteins and their preferred phospholipid type. Thus, the model could be used to identify lipid binding sites and equilibrate large membranes at an affordable computational cost.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"845-856"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142941372","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":"PACKMOL-GUI: An All-In-One VMD Interface for Efficient Molecular Packing.","authors":"Jian Huang, Chenchen Wu, Xiner Yang, Zaixing Yang, Shengtang Liu, Gang Yu","doi":"10.1021/acs.jcim.4c01639","DOIUrl":"10.1021/acs.jcim.4c01639","url":null,"abstract":"<p><p>PACKMOL is a widely utilized molecular modeling tool within the computational chemistry community. However, its tremendous advantages have been impeded by the longstanding lack of a robust open-source graphical user interface (GUI) that integrates parameter settings with the visualization of molecular and geometric constraints. To address this limitation, we have developed PACKMOL-GUI, a VMD plugin that leverages the dynamic extensibility of the Tcl/Tk toolkit. This GUI enables the configuration of all PACKMOL parameters through an intuitive user panel, while also facilitating the visualization of molecular structures and geometric constraints, including cubes, boxes, and spheres, among others via the VMD software. The seamless interaction between the VMD and PACKMOL provides an intuitive and efficient all-in-one platform for the packing of complex molecular systems.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"778-784"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142963269","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}
Maria Laura De Sciscio, Fabio Centola, Simona Saporiti, Marco D'Abramo
{"title":"Dissecting Methionine Oxidation by Hydrogen Peroxide in Proteins: Thermodynamics, Kinetics, and Susceptibility Descriptors.","authors":"Maria Laura De Sciscio, Fabio Centola, Simona Saporiti, Marco D'Abramo","doi":"10.1021/acs.jcim.4c01617","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01617","url":null,"abstract":"<p><p>The oxidation of Met residues in proteins is a complex process, where protein-specific structural and dynamical features play a relevant role in determining the reaction kinetics. Aiming to a full-side perspective, we report here a comprehensive characterization of Met oxidation kinetics by hydrogen peroxide in a leptin protein case study. To do that, we estimated the reaction-free energy profile of the Met oxidation via a QM/MM approach, while the kinetics of the formation of the reactive species were calculated using classical molecular dynamics (MD) simulations. Our data, validated against the available experimental data on the Met oxidation in this protein, indicated that the protein's local and global motion represent the primary discriminating factor among residues' oxidation rates. Moreover, assuming that the free energy profile is independent of the specific protein system, the different reactivities of Met residues within five proteins (hGCSF, IL-1ra, leptin, somatotropin, and RNase) were qualitatively analyzed in terms of well-known structural/dynamic features, which can affect the kinetics of the whole process. The comprehensive analysis of the reaction thermodynamics and kinetics fingerprint enabled the identification of additional descriptors, helpful in assessing the susceptibility of protein-bound Met residues to oxidation.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 2","pages":"749-761"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143044940","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}
Zuolong Zhang, Fang Liu, Xiaonan Shang, Shengbo Chen, Fang Zuo, Yi Wu, Dazhi Long
{"title":"ComNet: A Multiview Deep Learning Model for Predicting Drug Combination Side Effects.","authors":"Zuolong Zhang, Fang Liu, Xiaonan Shang, Shengbo Chen, Fang Zuo, Yi Wu, Dazhi Long","doi":"10.1021/acs.jcim.4c01737","DOIUrl":"10.1021/acs.jcim.4c01737","url":null,"abstract":"<p><p>As combination therapy becomes more common in clinical applications, predicting adverse effects of combination medications is a challenging task. However, there are three limitations of the existing prediction models. First, they rely on a single view of the drug and cannot fully utilize multiview information, resulting in limited performance when capturing complex structures. Second, they ignore subgraph information at different scales, which limits the ability to model interactions between subgraphs. Third, there has been limited research on effectively integrating multiview features of molecules. Therefore, we propose ComNet, a deep learning model that improves the accuracy of side effect prediction by integrating multiview features of drugs. First, to capture diverse features of drugs, a multiview feature extraction module is proposed, which not only uses molecular fingerprints but also extracts semantic information on SMILES and spatial information on 3D conformations. Second, to enhance the modeling ability of complex structures, a multiscale subgraph fusion mechanism is proposed, which can fuse local and global graph structures of drugs. Finally, a multiview feature fusion mechanism is proposed, which uses an attention mechanism to adaptively adjust the weights of different views to achieve multiview data fusion. Experiments on several publicly available data sets show that ComNet performs better than existing methods in various complex scenarios, especially in cold-start scenarios. Ablation experiments show that each core structure in ComNet contributes to the overall performance. Further analysis shows that ComNet not only converges rapidly and has good generalization ability but also identifies different substructures in the molecule. Finally, a case study on a self-collected data set validates the superior performance of ComNet in practical applications.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"626-639"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142918653","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":"High-Accuracy Identification and Structure-Activity Analysis of Antioxidant Peptides via Deep Learning and Quantum Chemistry.","authors":"Wanxing Li, Xuejing Liu, Yuanfa Liu, Zhaojun Zheng","doi":"10.1021/acs.jcim.4c01713","DOIUrl":"10.1021/acs.jcim.4c01713","url":null,"abstract":"<p><p>Antioxidant peptides (AOPs) hold great promise for mitigating oxidative-stress-related diseases, but their discovery is hindered by inefficient and time-consuming traditional methods. To address this, we developed an innovative framework combining machine learning and quantum chemistry to accelerate AOP identification and analyze structure-activity relationships. A Bi-LSTM-based model, AOPP, achieved superior performance with accuracies of 0.9043 and 0.9267, precisions of 0.9767 and 0.9848, and Matthews correlation coefficients (MCCs) of 0.818 and 0.859 on two data sets, outperforming existing methods. Compared with XGBoost and LightGBM, AOPP demonstrated a 4.67% improvement in accuracy. Feature fusion significantly enhanced classification, as validated by UMAP visualization. Experimental validation of ten peptides confirmed the antioxidant activity, with LLA exhibiting the highest DPPH and ABTS scavenging rates (0.108 and 0.437 mmol/g, respectively). Quantum chemical calculations identified LLA's lowest HOMO-LUMO gap (Δ<i>E</i> = 0.26 eV) and C<sub>3</sub>-H<sub>26</sub> as the key active site contributing to its superior antioxidant potential. This study highlights the synergy of machine learning and quantum chemistry, offering an efficient framework for AOP discovery with broad applications in therapeutics and functional foods.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"603-612"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142941363","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}
Yang Yue, Yihua Cheng, Céline Marquet, Chenguang Xiao, Jingjing Guo, Shu Li, Shan He
{"title":"Meta-Learning Enables Complex Cluster-Specific Few-Shot Binding Affinity Prediction for Protein-Protein Interactions.","authors":"Yang Yue, Yihua Cheng, Céline Marquet, Chenguang Xiao, Jingjing Guo, Shu Li, Shan He","doi":"10.1021/acs.jcim.4c01607","DOIUrl":"10.1021/acs.jcim.4c01607","url":null,"abstract":"<p><p>Predicting protein-protein interaction (PPI) binding affinities in unseen protein complex clusters is essential for elucidating complex protein interactions and for the targeted screening of peptide- or protein-based drugs. We introduce MCGLPPI++, a meta-learning framework designed to improve the adaptability of pretrained geometric models in such scenarios. To effectively boost the meta-learning optimization by injecting prior intersample distribution knowledge, three specially designed training sample cluster splitting patterns based on protein interaction interfaces are introduced. Additionally, MCGLPPI++ is equipped with an independent energy component which explicitly models interface nonbonded interaction energies closely related to the strengths of PPIs. To validate our approach, we curate a new data set featuring a challenging test cluster of T-cell receptors binding to antigenic peptide-MHC molecules (TCR-pMHC). Experimental results show that geometric models enhanced by the MCGLPPI++ framework achieve significantly more robust binding affinity predictions after fine-tuning on a few samples from this novel cluster compared to their vanilla counterparts, which demonstrates the effectiveness of the framework.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"580-588"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142941365","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":"FlowBack: A Generalized Flow-Matching Approach for Biomolecular Backmapping.","authors":"Michael S Jones, Smayan Khanna, Andrew L Ferguson","doi":"10.1021/acs.jcim.4c02046","DOIUrl":"10.1021/acs.jcim.4c02046","url":null,"abstract":"<p><p>Coarse-grained models have become ubiquitous in biomolecular modeling tasks aimed at studying slow dynamical processes such as protein folding and DNA hybridization. These models can considerably accelerate sampling but it remains challenging to accurately and efficiently restore all-atom detail to the coarse-grained trajectory, which can be vital for detailed understanding of molecular mechanisms and calculation of observables contingent on all-atom coordinates. In this work, we introduce FlowBack as a deep generative model employing a flow-matching objective to map samples from a coarse-grained prior distribution to an all-atom data distribution. We construct our prior distribution to be agnostic to the coarse-grained map and molecular type. A protein-specific model trained on ∼65k structures from the Protein Data Bank achieves state-of-the-art performance on structural metrics compared to previous generative and rules-based approaches in applications to static PDB structures, all-atom simulations of fast-folding proteins, and coarse-grained trajectories generated by a machine-learned force field. A DNA-protein model trained on ∼1.5k DNA-protein complexes achieves excellent reconstruction and generative capabilities on static DNA-protein complexes from the Protein Data Bank as well as on out-of-distribution coarse-grained dynamical simulations of DNA-protein complexation. FlowBack offers an accurate, efficient, and easy-to-use tool to recover all-atom structures from coarse-grained molecular simulations with higher robustness and fewer steric clashes than previous approaches. We make FlowBack freely available to the community as an open source Python package.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"672-692"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142941460","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}
Lucas Modesto da Costa, Reed Nieman, Adelia J A Aquino, Yehia Mechref, Hans Lischka
{"title":"The Use of Glycan Ensemble Structures and Nonpolar Surface Area Distributions for Correlating with Liquid Chromatography Retention Times.","authors":"Lucas Modesto da Costa, Reed Nieman, Adelia J A Aquino, Yehia Mechref, Hans Lischka","doi":"10.1021/acs.jcim.4c01920","DOIUrl":"10.1021/acs.jcim.4c01920","url":null,"abstract":"<p><p>The separation and structural identification of glycans are of great bioanalytical importance. To obtain a good understanding of the structural flexibility of glycans, replica exchange molecular dynamics (REMD) simulations were used based on AMBER force field calculations to create ensembles of glycan structures. Nonpolar surface area (NPSA) calculations based on continuum solvation (CS) models (Dhakal, R., et al. <i>Int. J. Mass Spectrom.</i> <b>2021</b>, <i>461</i>, 116495) were used to quantitatively characterize the polarity of the glycans. Retention times determined by tandem liquid chromatography-mass spectrometry (LC-MS) were correlated with CS-NPSA results obtained from analysis of the investigated glycan ensembles. Three classes of glycans with increasingly complex structures were investigated: linear glycans, fucosylated and sialylated biantennary glycans, and sialylated triantennary glycans. The linear and biantennary structures displayed bimodal distributions in their energies and CS-NPSA values, suggesting two sets of structures, while the more complex triantennary glycans displayed only a single distribution. The peak values of the CS-NPSA distributions (histogram structures) were selected as representatives to be correlated with the experimental retention times. For comparison, the most stable ensemble structures and those obtained from straightforward geometry optimizations were considered, as well. Overall, the histogram structures were found to correlate well with the retention times. In the case of the linear glycans, the CS-NPSA values for all three structural choices correlated very well with the retention times. For the biantennary glycans, the histogram data missed the retention-time ordering in one case but predicted the correct ordering for the triantennary case. Principal component analysis was performed to characterize the main glycan modes of the molecular dynamics.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":"882-895"},"PeriodicalIF":5.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142963276","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}