Identification of immunogenic cell death-related gene classification patterns and immune infiltration characterization in ischemic stroke based on machine learning.

Ischemic stroke (IS) accounts for more than 80% of strokes and is one of the leading causes of death and disability in the world. Due to the narrow time window for treatment and the frequent occurrence of severe bleeding, patients benefit less from early intravenous thrombolytic drug therapy. Therefore, there is an urgent need to explore the molecular mechanisms poststroke to drive the development of new therapeutic approaches. Immunogenic cell death (ICD) is a type of regulatory cell death (RCD) that is sufficient to activate the adaptive immune response of immunocompetent hosts. Although there is growing evidence that ICD regulation of immune responses and immune responses plays an important role in the development of IS, the role of ICD in the pathogenesis of IS has rarely been explored. In this study, we systematically evaluated ICD-related genes in IS. The expression profiles of ICD-related genes in IS and normal control samples were systematically explored. We conducted consensus clustering, immune infiltration analysis, and functional enrichment analysis of IS samples using ICD differentially expressed genes. The results showed that IS patients could be classified into two clusters and that the immune infiltration profile was altered in different clusters. In addition, we performed machine learning to screen nine signature genes that can be used to predict the occurrence of disease. We also constructed nomogram models based on the nine risk genes (CASP1, CASP8, ENTPD1, FOXP3, HSP90AA1, IFNA1, IL1R1, MYD88, and NT5E) and explored the immune infiltration correlation, gene-miRNA, and gene-TF regulatory network of the nine risk genes. Our study may provide a valuable reference for further elucidation of the pathogenesis of IS and provide directions for drug screening, personalized therapy, and immunotherapy for IS.