Abstract
Background
Osteoarthritis (OA), a common degenerative joint disease, is pathologically characterized by joint pain and functional limitation. Cuprotosis-related genes (CRGs) exert vital biological effects on various diseases, but their functions in OA remain largely unknown. We aimed to explore the potential role of CRGs in OA and to establish a diagnostic classifier.
Methods
The Gene Expression Omnibus database was firstly employed to collect data sets on several controls and OA samples. Batch correction was conducted using RobustRankAggreg and sva package to remove the systematic errors between different batches of sequencing. The limma package was utilized to screen differentially expressed genes, and CRGs were identified through Pearson correlation analysis.
Results
A total of 2,033 CRGs were identified after analyzing several data sets. Through Least Absolute Shrinkage and Selection Operator COX model and support vector machine-recursive feature elimination classifier, 6 crucial CRGs were finally determined, including biglycan, Ephrin-A3, leukemia inhibitory factor, natural killer cell granule protein 7, stimulator of chondrogenesis 1 and tumor necrosis factor, alpha-induced protein 3. The integrated analysis on these genes revealed that they had high prediction performance. The area under the curve was 0.772 in the training set and 0.693 in the validation set. These crucial CRGs exhibited significant correlations with the infiltration of M2 macrophages, resting mast cells and other immune cells.
Conclusions
A diagnostic classifier for OA was successfully constructed based on CRGs, and significant associations are found between crucial CRGs and immune microenvironment in OA.