Identification and Validation of Calcium-Related Diagnostic Markers for Acute Myocardial Infarction via Bioinformatics Analysis and Machine Learning.

The incidence of acute myocardial infarction (AMI) is increasing, and existing diagnostic techniques exhibit limited capacity for early AMI diagnosis. Given the robust association between Ca levels and the AMI initiation, calcium-related genes represent promising biomarkers for the early diagnosis of AMI.The expression data of patients with AMI and normal samples were obtained from the gene expression omnibus database. Weighted correlation network analysis (WGCNA) was applied to identify genes associated with AMI. Signature genes were screened using the least absolute shrinkage and selection operator, support vector machine-recursive feature elimination (SVM-RFE), and random forest algorithm. A diagnostic model based on the signature gene was established and evaluated. The CIBERSORT algorithm was used to determine the levels of immune cell infiltration, and the single-sample gene set enrichment analysis (ssGSEA) scores of the immune cells were calculated. The regulatory network of competing endogenous RNA (ceRNA) based on the signature genes was constructed using cytoscape. The DGIdb database was used to identify potential drugs for AMI that may interact with the signature genes.A high-performance diagnostic model based on four signature genes was established. The CIBERSORT algorithm and ssGSEA analysis revealed differences in immune cells between the patients with AMI and normal groups. The ceRNA regulatory network revealed multiple lncRNA and miRNA targeting signature genes. Niacin, nitroglycerin, arsenic disulfide, and quercetin are potential drugs that interact with the signature genes.Four signature genes were selected as calcium-related biomarkers of AMI that could serve as diagnostic markers for the disease. Additionally, the predicted ceRNA network and drug interaction network associated with these genes offer new perspectives for the treatment of AMI.