Exploring the mechanism of autophagy and inflammation in osteoarthritis through bioinformatics and experimental...

Osteoarthritis (OA) is a chronic joint disorder characterized by cartilage degradation and accompanied by inflammatory processes. Inflammation and autophagy dysfunction have been associated with the pathogenesis of OA. However, the interplay between autophagy- and inflammation-related genes in the development of OA remains poorly understood. RNA sequencing data of OA samples and autophagy-related genes (ARGs) were obtained from public databases, while inflammation-related genes were extracted from published literature. Differentially expressed genes (DEGs) in the OA dataset were identified using the limma R package. Using WGCNA, we identified autophagy-related gene modules (WGCNA-ARGs), and differential expression analysis was conducted to detect differentially expressed inflammation-related genes (IDEGs). Subsequently, candidate genes were identified by intersecting DEGs, WGCNA-ARGs, and IDEGs. Four machine learning algorithms-LASSO, SVM, RF, and XGBoost-were employed to screen for biomarkers, with diagnostic performance evaluated by ROC curve analysis. The biological functions and therapeutic potential of the biomarkers were further explored through multiple approaches: enrichment analysis, immune infiltration analysis, chromosomal localization, regulatory network construction, and drug prediction. Finally, the expression levels of the biomarkers were validated in clinical samples using RT-qPCR.