Bioinformatics-driven identification and validation of diagnostic biomarkers for cerebral ischemia reperfusion injury

Objective: This article aims to identify genetic features associated with immune cell infiltration in cerebral ischemia-reperfusion injury (CIRI) development through bioinformatics, with the goal of discovering diagnostic biomarkers and potential therapeutic targets. Methods: We obtained two datasets from the Gene Expression Omnibus (GEO) database to identify immune-related differentially expressed genes (IRDEGs). These genes' functions were analyzed via Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG). Tools such as CIBERSORT and ssGSEA assessed immune cell infiltration. The Starbase and miRDB databases predicted miRNAs interacting with hub genes, and Cytoscape software mapped mRNA-miRNA interaction networks. The ENCORI database was employed to predict RNA binding proteins interacting with hub genes. Key genes were identified using a random forest algorithm and constructing a Support Vector Machine (SVM) model. LASSO regression analysis constructed a diagnostic model for hub genes to determine their diagnostic value, and PCR analysis validated their expression in cerebral ischemia-reperfusion. Results: We identified 10 IRDEGs (C1qa, Ccl4, Cd74, Cd8a, Cxcl10, Gmfg, Grp, Lgals3bp, Timp1, Vim). The random forest algorithm, and SVM model intersection revealed three key genes (Ccl4, Gmfg, C1qa) as diagnostic biomarkers for CIRI. LASSO regression analysis, further refined this to two key genes (Ccl4 and C1qa), With ROC curve, analysis confirming their diagnostic efficacy (C1qa AUC = 0.75, Ccl4 AUC = 0.939). PCR analysis corroborated these findings. Conclusions: Our study elucidates immune and metabolic response mechanisms in CIRI, identifying two immune-related genes as key biomarkers and potential therapeutic targets in response to cerebral ischemia-reperfusion injury.