Identification of biomarkers and immune infiltration characterization of lipid metabolism associated genes in osteoarthritis based on machine learning algorithms

Osteoarthritis (OA) is a prevalent degenerative condition commonly observed in the elderly, leading to consequential disability. Despite notable advancements made in clinical strategies for OA, its pathogenesis remains uncertain. The intricate association between OA and metabolic processes has yet to receive comprehensive exploration. In our investigation, we leveraged public databases and applied machine learning algorithms, including WGCNA, LASSO, RF, immune infiltration analysis, and pathway enrichment analysis, to scrutinize the role of lipid metabolism-associated genes (LAGs) in the OA. Our findings identified three distinct biomarkers, and evaluated their expression to assess their diagnostic value in the OA patients. The exploration of immune infiltration in these patients revealed an intricate relationship between immune cells and the identified biomarkers. In addition, in vitro experiments, including qRT-PCR, Western blot, chondrocyte lipid droplets detection and mitochondrial fatty acid oxidation measurement, further verified abnormal expressions of selected LAGs in OA cartilage and confirmed the correlation between lipid metabolism and OA.