A diagnostic model of atherosclerosis based on the oxidative stress-glycolysis co-regulatory network.

Atherosclerosis (AS) is a major cardiovascular disorder, with challenges in early diagnosis and a lack of individualized treatment that require urgent attention. This study employed bioinformatics approaches to identify critical genetic markers linked to AS pathogenesis and explored their underlying molecular mechanisms to facilitate advancements in diagnostic accuracy and therapeutic interventions. We successfully identified genes exhibiting significant differential expression in AS, i.e., oxidative stress and glycolysis-related differentially expressed genes (OSGRDEGs). Through weighted gene co-expression network analysis, three modules (MEturquoise, MEred, and MEgreen) significantly associated with AS were screened, and 72 module genes were found to be identical to OSGRDEGs. A protein-protein interaction network was designed through comprehensive integration of data from the STRING database, followed by visualization and topological analysis employing Cytoscape software. Candidate genes were further evaluated using five distinct algorithms within the CytoHubba plugin, resulting in 12 high-confidence hub genes associated with AS pathogenesis. The 12 hub genes screened by machine algorithm were further screened by modeling to obtain 7 key genes. Finally, statistical analysis revealed marked variations in the infiltration levels of eight immune cell populations across the comparative groups. Monocytes and M0 macrophages showed significant negative correlations in subtypes A and B. Notably, APOE and CXCL1 demonstrated strong positive associations with M0 macrophages and monocytes, respectively, as evidenced by our correlation analysis. This study highlights the use of a bioinformatics approach to identify molecular markers of AS, with future work focused on validating their potential clinical applications.