Identification of ferroptosis-related key genes influencing synovitis development in osteoarthritis based on machine learning

This study aims to explore the potential involvement of ferroptosis-related genes (FRGs) in osteoarthritis (OA) synovitis. Synovial tissue samples from 32 OA patients and 32 normal controls were obtained from the GEO database and analyzed to identify differentially expressed FRGs.Weighted gene co-expression network analysis (WGCNA) identified a gene module significantly associated with OA. To further prioritize candidate genes, three machine learning approaches—Random Forest, Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost)—were applied, among which the SVM model demonstrated the best predictive performance.Through integrated analysis, CARS was identified as a candidate key ferroptosis-related gene associated with OA synovitis. Further immune infiltration analysis revealed a significant positive correlation between CARS expression and follicular helper T cell abundance, suggesting a potential association between ferroptosis-related metabolic regulation and immune microenvironment alterations in OA synovial tissue.In addition, competing endogenous RNA (ceRNA) network analysis indicated that CARS may be regulated by multiple non-coding RNAs, implying a complex post-transcriptional regulatory landscape. Experimental validation using qRT-PCR and Western blot analysis supported the bioinformatics prediction that CARS expression is elevated in OA synovial tissues at both mRNA and protein levels.Overall, this study provides a bioinformatics-based framework for identifying ferroptosis-related gene signatures in OA synovitis and suggests that CARS may serve as a potential candidate biomarker associated with ferroptosis and immune regulation in OA. Further functional studies are required to clarify its mechanistic role.