Incorporating machine learning, PPI networks to identify mitochondrial fission-related immune markers in abdominal aortic aneurysms

Purpose: This study was to investigate abdominal aortic aneurysm (AAA), a condition characterized by inflammation and progressive dilation of the blood vessels. Methods: To do this, we used weighted co-expression network analysis (WGCNA) and differential gene analysis on samples from the GEO database. Additionally, we carried out enrichment analysis and determined that the blue module was of interest. Additionally, we performed an investigation of immune infiltration and discovered genes linked to immune evasion and mitochondrial fission. In order to screen for feature genes, we used two PPI network gene selection methods and five machine learning methods. This allowed us to identify the most distinctive genes (MCGs). The expression of the MCGs in various cell subgroups was then evaluated by analysis of single cell samples from AAA. Additionally, we looked at the expression levels of the MCGs as well as the levels of inflammatory immune-related markers in cellular and animal models of AAA. Finally, we predicted potential drugs that could be targeted for the treatment of AAA. Results: Through differential gene analysis, our research identified 1249 up-regulated differential genes and 3653 down-regulated differential genes. Through WGCNA, we also discovered 44 genes in the blue module. By taking the point where several strategies for gene selection overlap, the MCG (ITGAL and SELL) was produced. We discovered through single cell research that the MCG were specifically expressed in T regulatory cells, NK cells, B lineage, and lymphocytes. In both animal and cellular models of AAA, the MCGs' mRNA levels rose. Conclusion: We searched for the AAA hallmark chemicals ITGAL and SELL, which most likely function through lymphocytes of the B lineage, NK cells, T regulatory cells, and B lineage. This analysis gave AAA a brand-new goal.