Abstract
OBJECTIVES: The purpose of this study is to leverage bioinformatics techniques to identify differentially expressed genes in Alzheimer's disease (AD), explore potential biomarkers for its early diagnosis, and provide new insights for the early diagnosis and treatment of AD. METHODS: Two Alzheimer's disease-associated datasets, GSE66351 and GSE153712, were obtained from the Gene Expression Omnibus (GEO) database. Differential methylation analysis was conducted on the raw data utilizing the R programming language. Key genes were discerned by integrating LASSO regression, Pearson correlation analysis, and protein-protein interaction network analysis (PPI). Furthermore, the functional roles of these genes were investigated via Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) enrichment analyses. To assess their causal association with AD, a Mendelian randomization analysis was performed. RESULTS: In two AD datasets, we identified a total of 387 overlapping differential methylation sites, which mapped to 297 genes. The GO enrichment analysis indicated that these genes are involved in a range of biological processes, such as signal transduction, cell cycle regulation, as well as the function of neuronal cell bodies and synapses. Furthermore, KEGG pathway analysis uncovered that these genes play crucial roles in the PI3K-Akt and TGF-beta signaling pathways. By utilizing a combination of LASSO, Pearson correlation analysis, and PPI network interaction analysis, we have identified five pivotal genes: EBF1, IGF1, EGR2, PRDM16, and RBL2. Finally, Mendelian randomization analysis demonstrated that the IVW analysis for cg00000029 (RBL2) yielded statistically significant results with an odds ratio (OR) of 1.201, and a 95 % confidence interval (CI) ranging from 1.089 to 1.325, corresponding to a p -value of 0.000249. CONCLUSIONS: This study not only confirmed the known genes linked to AD but, more significantly, revealed the potential connection between the RBL2 gene and AD. Furthermore, it verified a causal link between RBL2 and the risk of AD onset. This finding suggests that the RBL2 gene could serve as a promising biomarker for the early diagnosis of AD, thereby offering novel avenues for future research and clinical interventions.