Abstract
Atherosclerosis (AS) is a systemic disease closely related to inflammatory cell infiltration and immune cell activation, often leading to myocardial infarction and stroke and is the leading cause of death worldwide. AS is asymptomatic in its early stages, which leads to a low rate of early diagnosis of the disease and often delays treatment. Therefore, it is extremely important to explore potential biomarkers and molecular mechanisms for the diagnosis and treatment of AS, not only to improve early diagnosis and early treatment of patients but also to reduce the risk of death. The datasets GSE43292 and GSE100927 containing atherosclerotic plaques and normal arterial tissues (including 101 cases of atherosclerotic samples and 66 cases of normal tissue samples) were downloaded from the Gene Expression Omnibus database. The relationship between gene expression and immune cells was analyzed by the CIBERSORT package. Then the differentially expressed genes, weighted gene co-expression network analysis, and immune-related genes (IRGs) set were used to screen out the differentially expressed IRGs. These differentially expressed IRGs were then analyzed by constructing random forest model, support vector machine model, and generalized linear model. Next, a nomogram was established to assess disease risk, the calibration curve, decision curve analysis curve, and clinical impact curve were used to assess the validity of these models. The molecular mechanisms of these biomarkers were analyzed using single-gene gene set enrichment analysis. Potential target drugs for these molecules were identified in the Drug-Gene Interaction database. We screened 5 potential immune-relevant biomarkers (SYK, PTPRC, ITGAL, FGR, and IL10RA) associated with AS, constructed diagnostic models, and predicted potential therapeutic agents. Our findings, derived from integrated bioinformatics analysis, provide novel candidate genes and insights for the future diagnosis and treatment of AS, which warrant further experimental validation.