Analysis of Multi-Target Synergistic Mechanism of Coix Seed Therapy for Herpes Zoster Based on Machine Learning and Network Pharmacology

Zhiqin Song, Lin Yang, Jing He, Yuchao Li, Ningxian Yang, Min Yang, Mingkai Wu

Objective: To explore the efficacy and mechanism of Coix seeds in treating herpes zoster (HZ) using an integrated computational approach. Methods: Network pharmacology, molecular docking, and machine learning were employed. Disease-related targets were collected from multiple databases, and intersection targets with Coix seed were analyzed via PPI, GO, and KEGG enrichment. A “TCM-Ingredient-Target” network was constructed using Cytoscape. Molecular docking and dynamics simulations were performed for validation. Results: Fifty-five overlapping targets were identified, with core targets including TNF, EGF, and GAPDH. Enrichment analysis revealed key pathways such as inflammation and immune regulation. Molecular docking confirmed strong binding affinity between active compounds and targets. Conclusions: This study demonstrates that Coix seed exerts anti-HZ effects through multi-target mechanisms, providing a theoretical basis for developing novel multi-pathway treatment strategies.