Identification of an 11 immune-related gene signature as the novel biomarker for acute myocardial infarction diagnosis

This study explored the valuable immune gene signature for diagnosis of acute myocardial infarction (AMI). Three training gene expression datasets (GSE48060, GSE34198, and GSE97320) and one validation dataset (GSE109048) were downloaded from GEO database. The differentially infiltrated immune cells were analyzed by CIBERSORT. The immune genes were screened by WGCNA with differential immune cells as phenotype. Differentially expressed genes were analyzed in training datasets, followed by differential immune gene identification. The immune genes with diagnostic value were filtered by univariate regression analysis and LASSO model construction. ROC curve evaluated the diagnostic value of this model. A total of 304 differential immune genes were obtained, which were closely related to immune response-related pathways. An 11-gene signature (ADAMTS1, CNN2, DHRS13, DUSP1, FASLG, GNPTAB, NARF, PHC2, RAB7A, VNN3, and YIPF3) was filtered to be diagnostic biomarker for AMI (AUC = 0.805) and validated in GSE109048 dataset (AUC = 0.608). Based on the diagnostic model, high- and low-risk groups showed ten differential immune cells (such as T cells gamma delta, Macrophages M0, and Neutrophils) and differentially activated immune pathways (such as Antigen_Processing_and_Presentation and Cytokine_Receptors). PHC2 showed the highest positive correlation with Neutrophils and Cytokine_Receptors. The 11-gene signature could be served as a novel biomarker for the presence of AMI.