98%
921
2 minutes
20
Drug repositioning (DR) presents a cost-effective strategy for drug development by identifying novel therapeutic applications for existing drugs. Current computational approaches remain constrained by their inability to synergize localized substructure patterns with global network semantics, leading to overreliance on data augmentation to mitigate latent drug-disease association (DDA) information gaps. To address these limitations, we present multi-view stacked graph convolutional network (MVSGDR), a novel DR framework featuring three technical innovations: (i) multi-view stacked module that enables depth-wise feature enhancement through hierarchical aggregation of multi-hop neighborhood interactions across distinct graph convolutional layers; (ii) bi-level subgraph transformer module that decomposes DDAs into METIS (a graph partitioning tool) informative subgraphs for breadth-wise analysis of external and internal subgraph drug-disease relationships; and (iii) negative sampling balancing strategy that mitigates sample imbalance through negative sample synthesis. Extensive 10-fold cross-validation experiments across four benchmark datasets confirm MVSGDR's superior performance, demonstrating its statistically significant improvements over existing methods. Moreover, case studies further validate MVSGDR's potential utility through identification of previously unreported DDAs with supporting literature evidence.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12403086 | PMC |
http://dx.doi.org/10.1093/bib/bbaf396 | DOI Listing |
Brief Bioinform
August 2025
Department of Thoracic Surgery, The First Affiliated Hospital of Sun Yat-sen University, Zhongshan Second Road 58, Guangzhou, 510080 Guangdong, China.
Drug repositioning (DR) presents a cost-effective strategy for drug development by identifying novel therapeutic applications for existing drugs. Current computational approaches remain constrained by their inability to synergize localized substructure patterns with global network semantics, leading to overreliance on data augmentation to mitigate latent drug-disease association (DDA) information gaps. To address these limitations, we present multi-view stacked graph convolutional network (MVSGDR), a novel DR framework featuring three technical innovations: (i) multi-view stacked module that enables depth-wise feature enhancement through hierarchical aggregation of multi-hop neighborhood interactions across distinct graph convolutional layers; (ii) bi-level subgraph transformer module that decomposes DDAs into METIS (a graph partitioning tool) informative subgraphs for breadth-wise analysis of external and internal subgraph drug-disease relationships; and (iii) negative sampling balancing strategy that mitigates sample imbalance through negative sample synthesis.
View Article and Find Full Text PDFMed Image Anal
August 2025
School of Computation, Information and Technology, Technical University of Munich, Munich, Germany; School of Medicine, Klinikum Rechts der Isar, Technical University of Munich, Munich, Germany. Electronic address:
Cardiac magnetic resonance (CMR) imaging is the gold standard for non-invasive cardiac assessment, offering rich spatio-temporal views of the heart's anatomy and physiology. Patient-level health factors, such as demographics, metabolic, and lifestyle, are known to substantially influence cardiovascular health and disease risk, yet remain uncaptured by CMR alone. To holistically understand cardiac health and to enable the best possible interpretation of an individual's disease risk, CMR and patient-level factors must be jointly exploited within an integrated framework.
View Article and Find Full Text PDFDue to the properties of optical lenses, usual imaging devices suffer from a limited depth of field (DoF), and objects outside this area are blurred. To overcome the limited DoF, a common method is to continuously adjust the focal length or focal plane of the imaging system to capture a set of multi-focus images, and then fuse them into an all-in-focus image. However, such imaging mechanisms cannot capture multi-focus images simultaneously, thus failing to achieve all-in-focus imaging for each frame in dynamic scenes.
View Article and Find Full Text PDFBiophys Rev
April 2025
Division of Advanced Materials, Instituto Potosino de Investigación Científica y Tecnológica, Camino a la Presa San José 2055, 78216 San Luis Potosí, Mexico.
Despite its long history and widespread use, conventional bright-field optical microscopy has received recent attention as an excellent option to perform accurate, label-free, imaging of biological objects. As with any imaging system, bright-field produces an ill-defined representation of the specimen, in this case characterized by intertwined phase and amplitude in image formation, invisibility of phase objects at exact focus, and both positive and negative contrast present in images. These drawbacks have prevented the application of bright-field to the accurate imaging of unlabeled specimens.
View Article and Find Full Text PDFBMC Bioinformatics
April 2025
Faculty of Medical Technology, Center for Research Innovation and Biomedical Informatics, Mahidol University, Bangkok, 10700, Thailand.
Accelerating drug discovery for glucocorticoid receptor (GR)-related disorders, including innovative machine learning (ML)-based approaches, holds promise in advancing therapeutic development, optimizing treatment efficacy, and mitigating adverse effects. While experimental methods can accurately identify GR antagonists, they are often not cost-effective for large-scale drug discovery. Thus, computational approaches leveraging SMILES information for precise in silico identification of GR antagonists are crucial, enabling efficient and scalable drug discovery.
View Article and Find Full Text PDF