Construction of a glycolysis-related diagnostic model for osteoarthritis through integrated bioinformatics analysis and machine learning.

BACKGROUND

Osteoarthritis (OA) is a prevalent degenerative joint disease that significantly contributes to global disability. Glycolysis, a fundamental process in cellular energy metabolism, is particularly vital for chondrocytes in OA. This study aims to explore the intrinsic relationship between glycolysis-related genes (GRGs) and OA.

METHODS

We incorporated three publicly available datasets from the Gene Expression Omnibus (GEO) database, which included 64 OA samples and 34 normal controls. By utilizing differential expression analysis, weighted gene co-expression network analysis, protein-protein interaction networks, and machine learning methods, we identified three diagnostic biomarkers of OA patients. The expression levels of these biomarkers were validated by quantitative reverse transcription polymerase chain reaction (qRT-PCR) and immunohistochemical (IHC). Additionally, a competing endogenous RNA (ceRNA) network was constructed to explore potential regulatory interactions.

RESULTS

Through bioinformatics and machine learning approaches, three glycolysis-related biomarkers-HMGB2, SLC7A5, and ADM-were identified. The diagnostic model based on these GRGs demonstrated high predictive accuracy, with an AUC of 0.92 in the training set and 0.85 in the validation set. Subsequently, qRT-PCR and IHC confirmed the differential expression of hub genes in human cartilage samples. Furthermore, immunocyte infiltration analysis revealed distinct immune cell infiltration profiles between OA and HC groups. Notably, lncRNA XIST was found to regulate all three biomarkers, indicating its potential as a therapeutic target for OA.

CONCLUSION

This study provides novel insights into the role of glycolysis in OA pathogenesis and highlights its potential as a target for diagnosis, prevention, and treatment strategies.