Machine learning-driven discovery of STAT3 as a pivotal target for Wen-Wei-San-Ji formula in chronic atrophic gastritis therapy
Year: 2025
Type: article
Abstract: Background Wen-Wei-San-Ji Formula (WWSJ) is a Traditional Chinese medicine (TCM) formula used in treatment of chronic atrophic gastritis (CAG), demonstrating significant clinical efficacy. However, its specific mechanism of action remains unclear. This study investigates the molecular mechanisms and active components of WWSJ for CAG by employing network pharmacology, machine learning, and molecular docking, complemented by experimental validation. Methods Active compounds and targets of WWSJ were screened using TCMSP and Uniprot databases, while targets for CAG were collected from GeneCards and OMIM databases. Protein-protein-interaction and herb-compound-target networks were constructed using by STRING and Cytoscape. Functional enrichment was performed with DAVID. Machine learning (PCA/LASSO/RF) identified core targets from GEO data. In molecular docking studies, AutoDock Vina was used to evaluate binding activity. In vitro and in vivo experiments were conducted to validate the results. Results A total of 53 active ingredients with (OB)≥30% and (DL)≥0.1 from WWSJ were selected. Targets of the formula (1,657) and CAG disease (923) were screened, resulting in the identification of 207 common targets. Functional analysis of WWSJ by GO and KEGG revealed 742 Biological Processes (BP), 85 Cellular Components (CC), 116 Molecular Functions (MF), and 149 signaling pathways associated with CAG. Additionally, 5 core genes were identified through machine learning and molecular docking. The compounds demonstrated strong binding affinity with the target proteins, especially STAT3. Therefore, we hypothesized that WWSJ can treat CAG through inhibiting STAT3 signaling. Results of cell and animal experiments indicated that WWSJ is effective in the treatment of CAG through the inhibition of STAT3. Conclusion Integrated network pharmacology predictions and machine learning optimization identified STAT3 as the primary therapeutic target of WWSJ. Experimental studies confirmed that WWSJ alleviates CAG by suppressing STAT3 phosphorylation.
Authors: Tingting Huang, Lu Wang, Mengru Dou, Jia Guo, Kaihua Long, Yuan Wang, Yang Liu, Bo Wang, Weijian Zhao, Shanrong Han, Jingyi Bai, Xinli Wen, Ye Li, Yuxi Liu, Hong Zhang
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,550
OpenAlex ID: W4412496160
