Molecular Diversity最新文献_第10页

Palladium-catalyzed borylative coupling of hypervalent iodine diazo reagents with boranes for the synthesis of organoboron carboxylate. 钯催化高价碘重氮试剂与硼烷的硼酸偶联合成有机羧酸硼。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-20 DOI: 10.1007/s11030-026-11703-2

Haoxuan Yuan, Siyu Wu, Xinke Zhang, Abdulla Yusuf, Wenhao Hu, Yu Qian

引用次数: 0

Design, synthesis, and biological evaluation of novel trifluoromethyl-substituted quinolinone derivatives targeting triple-negative breast cancer. 针对三阴性乳腺癌的新型三氟甲基取代喹啉酮衍生物的设计、合成和生物学评价。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-20 DOI: 10.1007/s11030-026-11701-4

Mei Zhang, Man Chen, Yimin Yang, Sirui Qiao, Qiuli Yao, Jia Jia, Xueyu Man, Chun-Yang He

{"title":"Design, synthesis, and biological evaluation of novel trifluoromethyl-substituted quinolinone derivatives targeting triple-negative breast cancer.","authors":"Mei Zhang, Man Chen, Yimin Yang, Sirui Qiao, Qiuli Yao, Jia Jia, Xueyu Man, Chun-Yang He","doi":"10.1007/s11030-026-11701-4","DOIUrl":"https://doi.org/10.1007/s11030-026-11701-4","url":null,"abstract":"To develop more effective chemotherapeutic agents capable of simultaneously suppressing the growth and metastasis of triple-negative breast cancer (TNBC), a series of novel trifluoromethyl-substituted quinolinone derivatives were synthesized via a visible-light-promoted coupling reaction between coumarin/quinolinone precursors and α-trifluoromethylalkyl bromides. All newly synthesized compounds were fully characterized by 1H NMR, 13C NMR, 19F NMR and high-resolution mass spectrometry. The lead candidate (3a) was obtained, which exhibited potent cytotoxicity to TNBC cell line MDA-MB-231. Furthermore, 3a potently suppressed colony formation and three-dimensional spheroid growth, while markedly inhibited MDA-MB-231 migratory and invasive capacities. Notably, 3a exhibited superior antitumor and anti-metastatic activity in vivo than Paclitaxel. Mechanistic studies revealed that 3a induced apoptosis by regulating the BCL-2/Caspase-3 pathway and suppressed metastasis through downregulating the expression of VEGFR-2, MMP-2, and MMP-9.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148786751","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

Cloud-based ligand-guided virtual screening with deep-learning-enhanced docking identifies a micromolar CXCR4 antagonist. 基于云的配体引导虚拟筛选与深度学习增强对接识别微摩尔CXCR4拮抗剂。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-20 DOI: 10.1007/s11030-026-11695-z

Nonthaneth Nalinratana, Monsin Sangsawat, Kian Chee Chong, Yichun Xu, Junsong Han, Yanting Ding, Zhai Cai, Worathat Thitikornpong, Hua Zhu, Opa Vajragupta, Pornchai Rojsitthisak

{"title":"Cloud-based ligand-guided virtual screening with deep-learning-enhanced docking identifies a micromolar CXCR4 antagonist.","authors":"Nonthaneth Nalinratana, Monsin Sangsawat, Kian Chee Chong, Yichun Xu, Junsong Han, Yanting Ding, Zhai Cai, Worathat Thitikornpong, Hua Zhu, Opa Vajragupta, Pornchai Rojsitthisak","doi":"10.1007/s11030-026-11695-z","DOIUrl":"https://doi.org/10.1007/s11030-026-11695-z","url":null,"abstract":"We report a cloud-executed, ligand-guided virtual screening workflow for practical and rapid experimental testing of CXCR4 antagonists. Using the Rush platform, an Enamine purchasable library (274,092 compounds) was standardized and pre-filtered to remove assay-interfering substructures, yielding 264,953 molecules for screening. A set of 18,378 compounds was identified through AMD070-based similarity filtering (Tanimoto ≥ 0.2, ECFP4). These candidates were then docked into the IT1t-bound CXCR4 structure using Gnina. A CNN pose-score cutoff (> 0.8) was applied as a pose-confidence filter, and the retained compounds were ranked by docking affinity (kcal/mol). The top 50 were triaged by pharmacokinetic-focused in silico assessment (SwissADME) to prioritize lead-like profiles and exclude structural alerts, resulting in 9 purchasable candidates for biological testing. Flow cytometry-based 12G5 competitive binding in Jurkat cells identified CUEN-837 as the initial hit (IC50 = 8.50 μM, 52.31% inhibition at 10 μM). CUEN-837 also inhibited CXCL12-driven chemotaxis (32.8% inhibition at 10 μM). It reduced the viability of three colorectal cancer organoid strains in a dose-dependent manner (IC50 7.97-11.16 µM), providing supportive phenotypic evidence of activity in patient-derived organoid models. Overall, this end-to-end workflow efficiently compresses large libraries into a confirmed micromolar CXCR4 hit while lowering computational and infrastructure barriers for early-stage discovery.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148786821","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

Discovery of novel RIPK3 inhibitors by virtual screening and bioactivity evaluation. 通过虚拟筛选和生物活性评价发现新的RIPK3抑制剂。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-20 DOI: 10.1007/s11030-026-11680-6

Chenggong Fu, Qin Li, Yuwei Yang, Shukai Gu, Wei Zhao, Xiaojun Yao, Huanxiang Liu

{"title":"Discovery of novel RIPK3 inhibitors by virtual screening and bioactivity evaluation.","authors":"Chenggong Fu, Qin Li, Yuwei Yang, Shukai Gu, Wei Zhao, Xiaojun Yao, Huanxiang Liu","doi":"10.1007/s11030-026-11680-6","DOIUrl":"https://doi.org/10.1007/s11030-026-11680-6","url":null,"abstract":"As the most common chronic neurodegenerative disease, Alzheimer's disease (AD) has limited effective treatment options, highlighting the critical urgency of developing new therapeutic drugs. Receptor-interacting protein kinase 3 (RIPK3) is a key regulatory protein in the process of cell necroptosis and is also a potential target for AD. Existing inhibitors are not suitable for clinical treatment of AD due to problems such as insufficient activity, high toxicity, and poor selectivity. Therefore, the discovery of novel RIPK3 inhibitors is necessary and urgent. This study proposes and applies an integrated virtual screening strategy based on molecular fingerprint similarity, pharmacophore models, molecular docking, and molecular dynamics (MD) simulation. Using this workflow, we selected 54 compounds from approximately 9.7 million compounds for in vitro biological validation. In vitro kinase activity assays showed that three compounds (8019 - 0682, STK930700, and T72029) exhibited good inhibitory activity against RIPK3, with IC50 values of 1.684 µM, 3.002 µM, and 7.592 µM. Among them, 8019 - 0682 and T72029 demonstrated good cytoprotective effects in the TNF-α-induced L929 cell necroptosis model, with EC50 values of 2.135 µM and 15.891 µM. Finally, MD simulations and binding free energy calculations for the three compounds revealed a crucial hydrogen bond interaction with the M97 residue in the RIPK3 hinge region. In summary, this study proposes a combined virtual screening strategy and offers three promising lead compounds targeting RIPK3 for AD treatment.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148786747","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

Umamifusion: attention-enhanced multimodal fusion of graph and sequence features for umami peptide identification. 鲜味肽融合：用于鲜味肽识别的图形和序列特征的注意力增强多模态融合。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-20 DOI: 10.1007/s11030-026-11699-9

Pu Wang, Weihao Liu, Bo Hang

{"title":"Umamifusion: attention-enhanced multimodal fusion of graph and sequence features for umami peptide identification.","authors":"Pu Wang, Weihao Liu, Bo Hang","doi":"10.1007/s11030-026-11699-9","DOIUrl":"https://doi.org/10.1007/s11030-026-11699-9","url":null,"abstract":"Umami peptides are a class of functional biomolecules that significantly enhance food flavor, holding substantial application value in the food industry and nutritional science. Traditional prediction methods rely heavily on manual feature engineering, failing to adequately capture the complex semantic information and spatial structural characteristics of peptide sequences. This study presents UmamiFusion, a multimodal deep learning framework that integrates sequence local patterns with spatial structural features through attention-enhanced fusion to overcome the limitations of single-modal modeling approaches. Specifically, the protein language model ESM2 is employed to generate graph-structured representations of peptide molecules (with nodes representing amino acid residues and edges defining spatial contact relationships), combined with graph convolutional networks to extract high-order topological features. Simultaneously, one-dimensional convolutional neural networks (1D-CNNs) capture local contextual information from amino acid sequences. A bidirectional cross- attention mechanism enables adaptive interaction between graph nodes and sequence positions before the features are concatenated and fed into a multilayer perceptron for end-to-end classification. Experimental results on two meticulously constructed datasets, UMP442 and UMP614, demonstrate that UmamiFusion achieves accuracies of 96.27% and 98.24%, respectively, with MCC scores of 0.8214 and 0.8529, and AUC values of 0.9711 and 0.9879, significantly surpassing current state-of-the-art models. This research validates the critical role of attention-enhanced multimodal feature fusion in umami peptide prediction, providing an efficient computational tool for high-throughput screening of functional peptides and food flavor optimization.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148786799","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

Free energy perturbation and machine learning-assisted identification of potential MAP3K8 hit molecules: a comprehensive structure- and ligand-based studies. 自由能微扰和机器学习辅助识别潜在的MAP3K8击中分子：基于结构和配体的综合研究。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-12 DOI: 10.1007/s11030-026-11696-y

Md Ataul Islam, Aaherasha Iqbal, Mohammad Ajmal Ali, Rupesh Chikhale, Md Lutful Islam, Mohammad Abul Farah

{"title":"Free energy perturbation and machine learning-assisted identification of potential MAP3K8 hit molecules: a comprehensive structure- and ligand-based studies.","authors":"Md Ataul Islam, Aaherasha Iqbal, Mohammad Ajmal Ali, Rupesh Chikhale, Md Lutful Islam, Mohammad Abul Farah","doi":"10.1007/s11030-026-11696-y","DOIUrl":"10.1007/s11030-026-11696-y","url":null,"abstract":"MAP3K8, also known as COT or Tpl2, is a serine/threonine kinase that plays a pivotal role in regulating the MAPK signaling pathway and pro-inflammatory cytokine production, making it an attractive therapeutic target for inflammatory disorders and cancer. In this study, an integrated ligand- and structure-based virtual screening workflow was used to identify potential MAP3K8 inhibitors. A pharmacophore model derived from the co-crystal ligand bound to the COT kinase domain was used to screen the ChEMBL, PubChem, and ZINC databases, yielding 20,566 unique molecules. Sequential filtering through RMSD and binding-score thresholds, RDKit-based similarity searching, AutoDock Vina molecular docking, ADMET evaluation, and ChemMASTER-guided clustering reduced the dataset to four structurally diverse lead candidates designated as PM1, PM2, PM3, and PM4. All four molecules preserved the canonical hinge-binding pharmacophore of the reference ligand and exhibited binding energies between - 9.0 and - 10.2 kcal/mol, ranking more favourably than the co-crystal ligand with - 8.6 kcal/mol. The 200 ns molecular dynamics (MD) simulations confirmed the structural stability of the complexes, supported by favourable RMSD, RMSF, RoG, SASA, hydrogen-bond, PCA, and free energy landscape profiles. MM-GBSA calculations revealed that the binding free energies of PM2, PM3, and PM4 were - 42.12, - 38.46, and - 36.96 kcal/mol, respectively, outperformed the co-crystal ligand with - 34.36 kcal/mol, while FEP analysis identified PM3 with - 14.59 kcal/mol as the most thermodynamically favourable test compound, followed by PM2 and PM4. The convergence of docking, dynamics, and free-energy results prioritized PM2, PM3, and PM4 as promising MAP3K8 hit candidates, which require further experimental validation and lead optimization.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719502","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

The JAK/STAT pathway: a pivotal therapeutic target in disease and advances in natural product modulation. JAK/STAT通路：疾病的关键治疗靶点和天然产物调节的进展

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-12 DOI: 10.1007/s11030-026-11679-z

Ming-Yue Li, Yu-Yang Lin, Chun-Ying Wang, Yu-Shan Qi, Jun-Sheng Zhang, Di Ge, Hai-Na Yu, Hua Zhang, Jie Bao

{"title":"The JAK/STAT pathway: a pivotal therapeutic target in disease and advances in natural product modulation.","authors":"Ming-Yue Li, Yu-Yang Lin, Chun-Ying Wang, Yu-Shan Qi, Jun-Sheng Zhang, Di Ge, Hai-Na Yu, Hua Zhang, Jie Bao","doi":"10.1007/s11030-026-11679-z","DOIUrl":"10.1007/s11030-026-11679-z","url":null,"abstract":"The JAK/STAT (Janus kinase/signal transducer and activator of transcription) signaling pathway was discovered during the investigations on cellular responses to interferons (IFNs) in the early 1990s, and its components and functions were gradually elucidated and refined in the subsequent years. Serving as a core mediator of extracellular signal transduction such as cytokines and growth factors, the JAK/STAT signaling pathway plays a pivotal role in various diseases, representing an attractive therapeutic target. Although synthetic agents targeting this pathway have demonstrated clinical efficacy, long-term administration is frequently associated with adverse effects such as immunosuppression and infection risks. In recent years, natural products have emerged as novel candidates for modulating JAK/STAT signaling pathway, owing to their structural diversity, multi-target mechanisms and low-toxicity profiles. This review offers a comprehensive elucidation of the JAK/STAT signaling pathway, covering its discovery history, core components and transmembrane signal transduction mechanisms, and underscores its essential functions in the progression of diverse diseases. Furthermore, this review focuses on summarizing the research progress on natural products that modulate the JAK/STAT pathway from 2021 to 2025, and traces the research history of related natural products. This review aims to comprehensively demonstrate the development potential of natural products as drug lead compounds targeting the JAK/STAT pathway, thereby providing a theoretical foundation for subsequent studies.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719906","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

Deep learning-assisted virtual screening of a large chemical library for selective GSK3β inhibitors. 深度学习辅助的GSK3β抑制剂大型化学文库的虚拟筛选

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-12 DOI: 10.1007/s11030-026-11697-x

Tanmaykumar Varma, Pradnya Kamble, Prabha Garg

{"title":"Deep learning-assisted virtual screening of a large chemical library for selective GSK3β inhibitors.","authors":"Tanmaykumar Varma, Pradnya Kamble, Prabha Garg","doi":"10.1007/s11030-026-11697-x","DOIUrl":"10.1007/s11030-026-11697-x","url":null,"abstract":"Glycogen synthase kinase-3β (GSK3β) is a serine/threonine kinase involved in neurodegenerative, neuropsychiatric, and oncological disorders. The development of selective inhibitors remains challenging due to high kinase conservation, off-target effects, and suboptimal pharmacokinetic properties. A curated dataset of GSK3β molecules was obtained from PubChem, and molecular descriptors were calculated using PaDEL. GSK3BMTPred, a multitask deep neural network (DNN) model, was developed for simultaneous prediction of inhibitor classification and inhibitory potency. The optimized model achieved a training accuracy of 0.9809 for the classification task and a training coefficient of determination (R2) of 0.8692 for the regression task. The multitask framework improved generalization by learning shared molecular representations across both tasks, supported by SHapley Additive exPlanations (SHAP)-based interpretation of key physicochemical features. The model was integrated into a virtual screening workflow followed by molecular docking, molecular dynamics (MD) simulations, and Mechanics/Generalized Born Surface Area (MM/GBSA) analysis. This approach identified compounds showing stable interactions with key Adenosine Triphosphate (ATP)-binding residues and favorable predicted absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties, including blood-brain barrier (BBB) permeability and GSK3α selectivity. Mp7 was identified as a selective GSK3β inhibitor, whereas Mp8 demonstrated dual activity toward both GSK3β and GSK3α isoforms. Overall, this study presents a scalable computational framework for the discovery and prioritization of selective GSK3β inhibitors. The GSK3BMTPred model is freely available at: https://github.com/PGlab-NIPER/GSK3BMTPred.git .","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719706","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

MSCA-PLA: multi-scale cross-attention with differentiable pooling for protein-ligand binding affinity prediction. MSCA-PLA：多尺度交叉关注与可微分池的蛋白质配体结合亲和力预测。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-12 DOI: 10.1007/s11030-026-11698-w

Jiale Pan, Xiaosong Wang, Qingyong Wang

{"title":"MSCA-PLA: multi-scale cross-attention with differentiable pooling for protein-ligand binding affinity prediction.","authors":"Jiale Pan, Xiaosong Wang, Qingyong Wang","doi":"10.1007/s11030-026-11698-w","DOIUrl":"10.1007/s11030-026-11698-w","url":null,"abstract":"Accurately predicting protein-ligand binding affinity is essential for drug discovery. However, protein-ligand binding involves not only pairwise atomic contacts but also synergistic effects between atomic clusters; meanwhile, adaptively identifying which atoms can form effective atomic clusters to drive protein-ligand interactions remains a key challenge. To address this challenge, we propose MSCA-PLA, a multi-scale cross-attention framework with differentiable pooling for protein-ligand binding affinity prediction. Specifically, MSCA-PLA adaptively aggregates atom-level embeddings into binding-relevant atom clusters through differentiable pooling, and performs cluster-level cross-attention between protein and ligand clusters to model cooperative interactions among atom groups. Meanwhile, an atom-level bidirectional cross-attention module is employed to preserve fine-grained pairwise atomic contacts at the binding interface. Furthermore, to effectively integrate interaction information from different scales, we design a staged gated fusion strategy. Experiments on multiple benchmark datasets, diverse-protein settings, virtual-screening tasks, and a small external structural subset support the effectiveness of MSCA-PLA across the evaluated settings.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719895","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

Multimodal deep learning with a joint uncertainty quantification scheme for drug-target interaction prediction. 基于联合不确定性量化方案的多模态深度学习药物-靶标相互作用预测。

IF 4.3 2区化学

Molecular Diversity Pub Date : 2026-08-12 DOI: 10.1007/s11030-026-11692-2

Xingyu Xu, Huilin Xie, Yun Chen

{"title":"Multimodal deep learning with a joint uncertainty quantification scheme for drug-target interaction prediction.","authors":"Xingyu Xu, Huilin Xie, Yun Chen","doi":"10.1007/s11030-026-11692-2","DOIUrl":"10.1007/s11030-026-11692-2","url":null,"abstract":"AI-driven prediction of drug-target interaction (DTI) has emerged as a critical component in modern drug discovery and development. However, this approach is constrained by model and data uncertainties, which substantially affect its reliability and accuracy in DTI prediction. To overcome these limitations, we introduce EUQTri-DTI, a novel evidence-guided uncertainty quantification-based multimodal deep learning framework for DTI prediction. Specifically, EUQTri-DTI is designed to integrate three modality-specific networks to encode 1D protein sequences, 2D molecular images, and 3D drug structures. Here, a bidirectional cross-attention mechanism is employed to facilitate information exchange between different modalities. Additionally, we develop a joint uncertainty quantification scheme by calculating the weight summation of evidential uncertainty and prediction entropy from the aforementioned output, enabling a more comprehensive and nuanced assessment of uncertainties. Experiments are presented to demonstrate that EUQTri-DTI achieves stable and competitive performance through multiple evaluation metrics on three benchmark datasets, compared to baseline and state-of-the-art methods. Specifically, EUQTri-DTI achieves ROC-AUC and PR-AUC values of 85.28% and 84.53% on DrugBank, 93.45% and 83.02% on KIBA, and 92.41% and 85.32% on Davis, respectively. Moreover, the uncertainty analyses indicate that misclassified samples generally exhibit higher uncertainty, supporting sample-level confidence assessment and reliability-aware decision making. Overall, EUQTri-DTI combines competitive predictive performance with sample-level reliability assessment and shows potential for uncertainty-aware virtual screening.","PeriodicalId":708,"journal":{"name":"Molecular Diversity","volume":" ","pages":""},"PeriodicalIF":4.3,"publicationDate":"2026-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148719773","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