Screening necroptosis genes influencing osteoarthritis development based on machine learning

Machine learning can be applied to identify key genes associated with osteoarthritis (OA). This study aimed to explore the differential expression of necroptosis-related genes (NRGs) during the progression of OA, identify key gene modules strongly linked to the onset of OA, and assess the role of CASP1 and its correlation with immune cell infiltration in OA. Gene expression profile data were obtained for OA and normal tissues: GSE55235 (10 OA and 10 normal synovial tissues) and GSE46750 (12 OA and 12 normal synovial tissues). Differential expression analysis identified 44 NRGs. Weighted gene co-expression network analysis revealed that the turquoise module, including 2037 genes, showed a strong correlation with OA. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analyses showed that these genes were predominantly involved in regulating the JNK cascade, cellular response to oxidative stress, and Toll-like receptor signalling pathways. The support vector machine model exhibited the highest predictive performance (area under the curve of 0.883). Additionally, CASP1 expression in OA tissues was considerably elevated compared to normal tissues and was strongly associated with immune cell infiltration. These findings deepen our understanding of the pathophysiological foundation of OA and identify possible molecular targets for creating innovative OA therapies.