Abstract
Alzheimer's disease (AD) is a major cause of dementia, and this paper explores the unclear roles of mitochondrial dysfunction and programmed cell death in AD. Differentially expressed genes (DEGs) were identified using AD datasets GSE63061 and GSE63060 from the Gene Expression Omnibus. DEGs were intersected with mitochondria-related genes and programmed cell death-related genes to obtain DEGs (in AD) intersected with mitochondrial-related genes and DEGs (in AD) intersected with programmed cell death-related genes. Correlation analysis of these DEGs was used to identify candidate genes. Machine learning algorithms were applied to key genes, followed by functional enrichment, network construction, immune infiltration analysis, drug prediction, and expression validation. Two key genes, superoxide dismutase 1 (SOD1) and translocase of the outer mitochondrial membrane 7 (TOMM7), were identified and linked to pathways like ribosome and chemokine signaling. A strong positive correlation (0.76, P < .001) was found between them. Immune analysis showed differences in 11 immune cells between AD and controls, with TOMM7 positively linked to activated CD8 T cells and negatively to myeloid-derived suppressor cells. SOD1 and TOMM7 are regulated by 4 miRNAs and 71 long noncoding RNAs (lncRNAs). Seventeen potential AD drugs, including urea and nitric oxide, were predicted. Two key genes, SOD1 and TOMM7, related to mitochondria and PCD, were identified as potential targets for understanding AD's etiology, detection, and therapeutic approaches.