Deep learning, network pharmacology and computational simulation reveal multi-target therapeutic mechanisms of Vitis vinifera (Grapes) compounds against Alzheimer’s disease
Md. Mainuddin Hossain, Juthi Adhikari, Abu Zaffar Shibly · In Silico Research in Biomedicine · 2025
Vitis vinifera ( V. vinifera ) has shown therapeutic promise in Alzheimer’s disease (AD) due to its bioactive compounds. However, the molecular mechanisms underlying these effects remain poorly understood. This study aims to elucidate these mechanisms through an integrative approach combining deep learning, network pharmacology, and molecular docking analyses. Bioactive compounds of V. vinifera were retrieved from PubChem, and their potential targets predicted using SwissTargetPrediction. AD-related genes were obtained from GeneCards, and overlapping targets were used to construct a protein–protein interaction (PPI) network via STRING and Cytoscape. Hub genes were identified using CytoHubba and validated using a GCNConv-based graph deep learning framework. Functional enrichment (GO and KEGG) analyses were conducted using DAVID, and molecular docking was performed to assess compound-target binding affinities. 6 pharmacologically active compounds (flavylium, jasmonic acid, methyl anthranilate, dehydrovomifoliol, methyl vanillate, and 2,3-dimethoxyphenol) were identified with favorable pharmacokinetic properties. From 42 overlapping AD-related targets, 7 hub genes were validated via the GCNConv model, which showed strong predictive performance (R²: 0.9858 training, 0.9677 validation, 0.9575 testing). Among these, 5 key targets (TNF, APP, IL6, PPARG, and GSK3B) were functionally linked to neuroinflammatory and neurodegenerative signaling pathways, including IL-17 and NF-κB. Molecular docking and dynamic simulation revealed that flavylium exhibited strong and stable binding with all 5 key targets compared to memantine (control). This study suggests that V. vinifera exerts therapeutic effects against AD through multitarget regulation of inflammation and neuronal survival pathways. Flavylium, in particular, emerges as a promising candidate and offers mechanistic insights into V. vinifera -based AD therapy.