【目的】 基于生物信息学探讨佛山市中医院骨九方治疗骨关节炎（OA）的分子机制。【方法】 从TCMSP数据库和HERB 数据库收集骨九方的潜在活性成分及作用靶点；在 GEO数据库获取骨关节炎相关微阵列芯片，进行 WGCNA 分析和差异分 析，获得骨关节炎相关模块基因和差异基因。整合上述生物信息集，进行功能富集分析、蛋白-蛋白互作（PPI）分析，使用 4种机器学习方法（RF、SVM、XGBoost、GLM）确定骨九方治疗骨关节炎的关键基因，对关键基因进行表达分析、单基因 GSEA 分析。最后，对关键活性成分及靶点进行分子模拟验证。【结果】 获得骨九方 660个潜在活性成分和 447个作用靶点， 2 个骨关节炎相关WGCNA模块和1 546个骨关节炎上调差异基因。骨九方与骨关节炎的交集靶点有47个，功能富集显示与 TNF 信号通路、NF-kappa B 信号通路、Hippo 信号通路、PI3K-Akt 信号通路、JAK-STAT 信号通路、p53 信号通路等相关。 RF、SVM机器学习模型具有较好的预测功能，结合PPI分析确定肌动蛋白β（ACTB）为骨九方治疗骨关节炎的关键基因。对 接结果表明，山奈酚、金雀异黄素、木犀草素、汉黄芩素等关键成分与关键靶点具有较好的结合能力。【结论】 骨九方可能 通过多成分、多靶点、多途径抑制骨关节炎炎症水平、抑制软骨细胞凋亡、降低软骨降解、调控成骨细胞和破骨细胞增殖 分化，缓解骨质流失。

Objective To investigate the molecular mechanism of Foshan Hospital of Traditional Chinese Medicine Bone Formula IX in treating osteoarthritis OA） based on bioinformatics. Methods Potential active components and corresponding targets of Bone Formula IX were collected from the TCMSP and HERB databases. OA-related microarray datasets were obtained from the GEO database，and Weighted Gene Co-expression Network Analysis WGCNA） and differential expression analysis were performed to identify OA-related module genes and differentially expressed genes. The integrated bioinformatics datasets were subjected to functional enrichment analysis. Protein-protein interaction PPI） network analysis was conducted，and four machine learning methods RF， SVM， XGBoost， GLM） were employed to identify key genes of Bone Formula IX for OA treatment. Expression analysis and single-sample Gene Set Enrichment Analysis GSEA） of the key genes were performed. Finally，molecular docking simulations were used to validate the key active components and targets. Results A total of 660 potential active components in Bone Formula IX and 447 potential targets were identified，along with 2 OA-related WGCNA modules and 1 546 upregulated differentially expressed genes in OA. Forty-seven overlapping targets were found between Bone Formula IX and OA. Functional enrichment analysis revealed associations with the TNF signaling pathway， NF-kappa B signaling pathway， Hippo signaling pathway， PI3K-Akt signaling pathway，JAK-STAT signaling pathway，and p53 signaling pathway，among others. The RF and SVM machine learning models demonstrated good predictive performance. Combined with PPI analysis，β -actin ACTB） was identified as a key gene for Bone Formula IX in treating OA. Finally，molecular docking results indicated that key components such as kaempferol，genistein，luteolin，and wogonin exhibited strong binding affinity with the key target. Conclusion Bone Formula IX acts through multiple components， targets， and pathways to inhibit OA inflammatory levels， suppress chondrocyte apoptosis and reduce cartilage degradation， and regulate the proliferation and differentiation of osteoblasts and osteoclasts to alleviate bone loss. This study provides a theoretical basis for further clinical research and pharmacological studies on Bone Formula IX for OA treatment.

R285

国家自然科学基金资助项目（编号：82374475）