Journal of Chemical Information and Modeling 最新文献

{"title":"T-ALPHA: A Hierarchical Transformer-Based Deep Neural Network for Protein-Ligand Binding Affinity Prediction with Uncertainty-Aware Self-Learning for Protein-Specific Alignment.","authors":"Gregory W Kyro, Anthony M Smaldone, Yu Shee, Chuzhi Xu, Victor S Batista","doi":"10.1021/acs.jcim.4c02332","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02332","url":null,"abstract":"<p><p>There is significant interest in targeting disease-causing proteins with small molecule inhibitors to restore healthy cellular states. The ability to accurately predict the binding affinity of small molecules to a protein target in silico enables the rapid identification of candidate inhibitors and facilitates the optimization of on-target potency. In this work, we present T-ALPHA, a novel deep learning model that enhances protein-ligand binding affinity prediction by integrating multimodal feature representations within a hierarchical transformer framework to capture information critical to accurately predicting binding affinity. T-ALPHA outperforms all existing models reported in the literature on multiple benchmarks designed to evaluate protein-ligand binding affinity scoring functions. Remarkably, T-ALPHA maintains state-of-the-art performance when utilizing predicted structures rather than crystal structures, a powerful capability in real-world drug discovery applications where experimentally determined structures are often unavailable or incomplete. Additionally, we present an uncertainty-aware self-learning method for protein-specific alignment that does not require additional experimental data and demonstrate that it improves T-ALPHA's ability to rank compounds by binding affinity to biologically significant targets such as the SARS-CoV-2 main protease and the epidermal growth factor receptor. To facilitate implementation of T-ALPHA and reproducibility of all results presented in this paper, we made all of our software available at https://github.com/gregory-kyro/T-ALPHA.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143447353","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":"A Machine-Learned \"Chemical Intuition\" to Overcome Spectroscopic Data Scarcity.","authors":"Cailum M K Stienstra, Teun van Wieringen, Liam Hebert, Patrick Thomas, Kas J Houthuijs, Giel Berden, Jos Oomens, Jonathan Martens, W Scott Hopkins","doi":"10.1021/acs.jcim.4c02329","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02329","url":null,"abstract":"<p><p>Machine learning models for predicting IR spectra of molecular ions (infrared ion spectroscopy, IRIS) have yet to be reported owing to the relatively sparse experimental data sets available. To overcome this limitation, we employ the Graphormer-IR model for neutral molecules as a knowledgeable starting point and then employ transfer learning to refine the model to predict the spectra of gaseous ions. A library of 10,336 computed spectra and a small data set of 312 experimental IRIS spectra is used for model fine-tuning. Nonspecific global graph encodings that describe the molecular charge state (<i>i.e</i>., (de)protonation, sodiation), combined with an additional transfer learning step that considers computed spectra for ions, improved model performance. The resulting Graphormer-IRIS model yields spectra that are 21% more accurate than those produced by commonly employed DFT quantum chemical models, while capturing subtle phenomena such as spectral red-shifts due to sodiation. Dimensionality reduction of model embeddings demonstrates derived \"chemical intuition\" of functional groups, trends in molecular electron density, and the location of charge sites. Our approach will enable fast IRIS predictions for determining the structures of unknown small molecule analytes (<i>e.g</i>., metabolites, lipids) present in biological samples.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143439322","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":"PbImpute: Precise Zero Discrimination and Balanced Imputation in Single-Cell RNA Sequencing Data.","authors":"Yi Zhang, Yin Wang, Xinyuan Liu, Xi Feng","doi":"10.1021/acs.jcim.4c02125","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02125","url":null,"abstract":"<p><p>Single-cell RNA sequencing (scRNA-seq) has emerged as a transformative technology for elucidating cellular heterogeneity at unprecedented resolution. However, technical limitations such as limited sequencing depth and mRNA capture efficiency often result in zero counts, commonly referred to as \"dropout zeros\" in scRNA-seq data. These zeros pose significant challenges to downstream analysis, as they can distort the interpretation of cellular transcriptomes. While numerous computational methods have been developed to address this challenge, existing approaches frequently suffer from either insufficient imputation of zeros (under-imputation) or excessive modification of zeros (over-imputation). Here, we propose a precisely balanced imputation (PbImpute) method designed to achieve optimal equilibrium between dropout recovery and biological zero preservation in scRNA-seq data. PbImpute employs a multistage approach: (1) Initial discrimination between technical dropouts and biological zeros through parameter optimization of a new zero-inflated negative binomial (ZINB) distribution model, followed by initial imputation; (2) Application of a uniquely designed static repair algorithm to enhance data fidelity; (3) Secondary dropout identification based on gene expression frequency and partition-specific coefficient of variation; (4) Graph-embedding neural network-based imputation; and (5) Implementation of a uniquely designed dynamic repair mechanism to mitigate over-imputation effects. PbImpute distinguishes itself by uniquely integrating ZINB modeling with static and dynamic repair. This advantageous combined approach achieves a balance between over- and under-imputation, while simultaneously preserving true biological zeros and reducing signal distortion. Comprehensive evaluation using both simulated and real scRNA-seq data sets demonstrated that PbImpute achieves superior performance (F1 Score = 0.88 at 83% dropout rate, ARI = 0.78 on PBMC) in discriminating between technical dropouts and biological zeros compared to state-of-the-art methods. The method significantly improves gene-gene and cell-cell correlation structures, enhances differential expression analysis sensitivity, optimizes clustering resolution and dimensional reduction visualization, and facilitates more accurate trajectory inference. Ablation studies confirmed the essential contribution of both the imputation and repair modules to the method's performance. The code is available at https://github.com/WyBioTeam/PbImpute. By enhancing the accuracy of scRNA-seq data imputation, PbImpute can improve the identification of cell subpopulations and the detection of differentially expressed genes, thereby facilitating more precise analyses of cellular heterogeneity and advancing disease research.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143431966","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":"AI-Augmented R-Group Exploration in Medicinal Chemistry.","authors":"Hongtao Zhao, Karolina Kwapień, Eva Nittinger, Christian Tyrchan, Magnus Nilsson, Susanne Berglund, Werngard Czechtizky","doi":"10.1021/acs.jcim.4c02326","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02326","url":null,"abstract":"<p><p>Efficient R-group exploration in the vast chemical space, enabled by increasingly available building blocks or generative AI, remains an open challenge. Here, we developed an enhanced Free-Wilson QSAR model embedding R-groups by atom-centric pharmacophoric features. Regioisomers of R-groups can be distinguished by explicitly accounting for the atomic positions. Good predictivity is observed consistently across 12 public data sets. Integrated into an open-source program, we showcase its application in performing Free-Wilson analysis as well as R-group exploration in an uncharted chemical space.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143439341","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":"Accurate Prediction of ωB97X-D/6-31G* Equilibrium Geometries from a Neural Net Starting from Merck Molecular Force Field (MMFF) Molecular Mechanics Geometries.","authors":"Thomas Hehre, Philip Klunzinger, Bernard Deppmeier, William Ohlinger, Warren Hehre","doi":"10.1021/acs.jcim.4c01898","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01898","url":null,"abstract":"<p><p>Starting from Merck Molecular Force Field (MMFF) geometries, a neural-net based model has been formulated to closely reproduce ωB97X-D/6-31G* equilibrium geometries for organic molecules. The model involves training to >6 million energy and force calculations for molecules with molecular weights ranging from 200 to 600 amu, corresponding to both ωB97X-D/6-31G* and MMFF equilibrium geometries as well as small deviations away from these geometries. 422 natural products not involved in training with molecular weights ranging from 200 to 691 amu have been used to assess the neural net model against changes in bond lengths, bond angles, and dihedral angles, as well as against changes in proton and ¹³C chemical shifts resulting from using equilibrium geometries from the neural-net in lieu of geometries from ωB97X-D/6-31G*. The neural net reduces calculation times by two or more orders of magnitude.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143439337","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-Throughput Prediction of Metal-Embedded Complex Properties with a New GNN-Based Metal Attention Framework.","authors":"Xiayi Zhao, Bao Wang, Kun Zhou, Jiangjiexing Wu, Kai Song","doi":"10.1021/acs.jcim.4c02163","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02163","url":null,"abstract":"<p><p>Metal-embedded complexes (MECs), including transition metal complexes (TMCs) and metal-organic frameworks (MOFs), are important in catalysis, materials science, and molecular devices due to their unique metal atom centrality and complex coordination environments. However, modeling and predicting their properties accurately is challenging. A new metal attention (MA) framework for graph neural networks (GNNs) was proposed to address the limitations of traditional methods, which fail to differentiate core coordination structures from ordinary covalent bonds. This MA framework converts heterogeneous graphs of complexes into homogeneous ones with distinct metal features by highlighting key metal-feature coordination through hierarchical pooling and a metal cross-attention. To assess its performance, 11 widely used GNN algorithms, three of which are heterogeneous, were compared. Experimental results indicate significant improvements in accuracy: an average of 32.07% for predicting TMC properties and up to 23.01% for MOF CO₂ absorption. Moreover, tests on the framework's robustness regarding data set size variation and comparison with a larger non-MA model show that the enhanced performance stems from the architecture, not merely increasing model capacity. The MA framework's potential in predicting metal complex properties offers a potent statistical tool for optimizing and designing new materials like catalysts and gas storage systems.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143412358","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":"Dynamic Electronic Structure Fluctuations in the De Novo Peptide ACC-Dimer Revealed by First-Principles Theory and Machine Learning.","authors":"Peter Mastracco, Luke Nambi Mohanam, Giacomo Nagaro, Sangram Prusty, Younghoon Oh, Ruqian Wu, Qiang Cui, Allon I Hochbaum, Stacy M Copp, Sahar Sharifzadeh","doi":"10.1021/acs.jcim.4c01979","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01979","url":null,"abstract":"<p><p>Recent studies have reported long-range charge transport in peptide- and protein-based fibers and wires, rendering this class of materials as promising charge-conducting interfaces between biological systems and electronic devices. In the complex molecular environment of biomolecular building blocks, however, it is unclear which chemical and structural dynamic features support electronic conductivity. Here, we investigate the role of finite temperature fluctuations on the electronic structure and its implications for conductivity in a peptide-based fiber material composed of an antiparallel coiled coil hexamer, ACC-Hex, building block. All-atom classical molecular dynamics (MD) and first-principles density functional theory (DFT) are combined with interpretable machine learning (ML) to understand the relationship between physical and electronic structure of the peptide dimer subunit of ACC-Hex. For 1101 unique MD \"snapshots\" of the ACC peptide dimer, hybrid DFT calculations predict a significant variation of near-gap orbital energies among snapshots, with an increase in the predicted number of nearly degenerate states near the highest occupied molecular orbital (HOMO), which suggests improved conductivity. Interpretable ML is then used to investigate which nuclear conformations increase the number of nearly degenerate states. We find that molecular conformation descriptors of interphenylalanine distance and orientation are, as expected, highly correlated with increased state density near the HOMO. Unexpectedly, we also find that descriptors of tightly coiled peptide backbones, as well as those describing the change in the electrostatic environment around the peptide dimer, are important for predicting the number of hole-accessible states near the HOMO. Our study illustrates the utility of interpretable ML as a tool for understanding complex trends in large-scale ab initio simulations.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143412357","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":"The Need for Continuing Blinded Pose- and Activity Prediction Benchmarks.","authors":"Christian Kramer, John Chodera, Kelly L Damm-Ganamet, Michael K Gilson, Judith Günther, Uta Lessel, Richard A Lewis, David Mobley, Eva Nittinger, Adam Pecina, Matthieu Schapira, W Patrick Walters","doi":"10.1021/acs.jcim.4c02296","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02296","url":null,"abstract":"<p><p>Computational tools for structure-based drug design (SBDD) are widely used in drug discovery and can provide valuable insights to advance projects in an efficient and cost-effective manner. However, despite the importance of SBDD to the field, the underlying methodologies and techniques have many limitations. In particular, binding pose and activity predictions (P-AP) are still not consistently reliable. We strongly believe that a limiting factor is the lack of a widely accepted and established community benchmarking process that independently assesses the performance and drives the development of methods, similar to the CASP benchmarking challenge for protein structure prediction. Here, we provide an overview of P-AP, unblinded benchmarking data sets, and blinded benchmarking initiatives (concluded and ongoing) and offer a perspective on learnings and the future of the field. To accelerate a breakthrough on the development of novel P-AP methods, it is necessary for the community to establish and support a long-term benchmark challenge that provides nonbiased training/test/validation sets, a systematic independent validation, and a forum for scientific discussions.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":" ","pages":""},"PeriodicalIF":5.6,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143416786","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":"metaCDA: A Novel Framework for CircRNA-Driven Drug Discovery Utilizing Adaptive Aggregation and Meta-Knowledge Learning","authors":"Li Peng,&nbsp;Huaping Li,&nbsp;Sisi Yuan,&nbsp;Tao Meng,&nbsp;Yifan Chen,&nbsp;Xiangzheng Fu* and Dongsheng Cao*,&nbsp;","doi":"10.1021/acs.jcim.4c0219310.1021/acs.jcim.4c02193","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c02193https://doi.org/10.1021/acs.jcim.4c02193","url":null,"abstract":"<p >In the emerging field of RNA drugs, circular RNA (circRNA) has attracted much attention as a novel multifunctional therapeutic target. Delving deeper into the intricate interactions between circRNA and disease is critical for driving drug discovery efforts centered around circRNAs. Current computational methods face two significant limitations: a lack of aggregate information in heterogeneous graph networks and a lack of higher-order fusion information. To this end, we present a novel approach, metaCDA, which utilizes meta-knowledge and adaptive aggregate learning to improve the accuracy of circRNA and disease association predictions and addresses the limitations of both. We calculate multiple similarity measures between disease and circRNA, construct a heterogeneous graph based on these, and apply meta-networks to extract meta-knowledge from the heterogeneous graph, so that the constructed heterogeneous maps have adaptive contrast enhancement information. Then, we construct a nodal adaptive attention aggregation system, which integrates a multihead attention mechanism and a nodal adaptive attention aggregation mechanism, so as to achieve accurate capture of higher-order fusion information. We conducted extensive experiments, and the results show that metaCDA outperforms existing state-of-the-art models and can effectively predict disease-associated circRNA, opening up new prospects for circRNA-driven drug discovery.</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 4","pages":"2129–2144 2129–2144"},"PeriodicalIF":5.6,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143473619","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":"AGDIFF: Attention-Enhanced Diffusion for Molecular Geometry Prediction","authors":"André Brasil Vieira Wyzykowski,&nbsp;Fatemeh Fathi Niazi and Alex Dickson*,&nbsp;","doi":"10.1021/acs.jcim.4c0189610.1021/acs.jcim.4c01896","DOIUrl":"https://doi.org/10.1021/acs.jcim.4c01896https://doi.org/10.1021/acs.jcim.4c01896","url":null,"abstract":"<p >Accurate prediction of molecular geometries is crucial for drug discovery and materials science. Existing fast conformer prediction algorithms often rely on approximate empirical energy functions, resulting in low accuracy. More accurate methods like ab initio molecular dynamics and Markov chain Monte Carlo can be computationally expensive due to the need for evaluating quantum mechanical energy functions. To address this, we introduce AGDIFF, a novel machine learning framework that utilizes diffusion models for efficient and accurate molecular structure prediction. AGDIFF extends previous models (such as GeoDiff) by enhancing the global, local, and edge encoders with attention mechanisms, an improved SchNet architecture, batch normalization, and feature expansion techniques. AGDIFF outperforms GeoDiff on both the GEOM-QM9 and GEOM-Drugs data sets. For GEOM-QM9, with a threshold (δ) of 0.5 Å, AGDIFF achieves a mean COV-R of 93.08% and a mean MAT-R of 0.1965 Å. On the more complex GEOM-Drugs data set, using δ = 1.25 Å, AGDIFF attains a median COV-R of 100.00% and a mean MAT-R of 0.8237 Å. These findings demonstrate AGDIFF’s potential to advance molecular modeling techniques, enabling more efficient and accurate prediction of molecular geometries, thus contributing to computational chemistry, drug discovery, and materials design. https://github.com/ADicksonLab/AGDIFF</p>","PeriodicalId":44,"journal":{"name":"Journal of Chemical Information and Modeling ","volume":"65 4","pages":"1798–1811 1798–1811"},"PeriodicalIF":5.6,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://pubs.acs.org/doi/epdf/10.1021/acs.jcim.4c01896","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143473795","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}