Abstract
Diabetic nephropathy (DN) is a kidney disease. Mitochondrial and endoplasmic reticulum stress (ERS) significantly contribute to diabetic nephropathy (DN), although the precise mechanisms involved have not yet been fully understood. The objective of this research was to explore the potential of mitochondrial and ERS genes as pivotal genetic elements in individuals with DN and to elucidate their fundamental molecular mechanisms. The datasets GSE30528 and GSE30122 were obtained from the Gene Expression Omnibus (GEO) database. Firstly, differentially expressed genes (DEGs) (DN and control samples) were identified by differential expression analysis. Candidate genes were obtained by intersecting the DEGs with mitochondria and endoplasmic reticulum stress-related genes. The key genes were identified through three machine learning methods, the receiver operating characteristic (ROC) curve analysis and expression validation. Subsequently, a nomogram model for DN was constructed. Moreover, gene set enrichment analysis (GSEA), immune infiltration, molecular regulatory networks of key genes were explored, Later, predicted their drugs. Finally, three key genes (GPX1, PPIF and VDAC1) were identified by expression validation and ROC validation and three key genes were all down-regulated in DN. Meanwhile, RT-qPCR analysis yielded the same results. In addition, the nomogram model of key genes was constructed, and the model had a good prediction effect. GSEA showed that the top 3 most prominent pathways shared by the 3 key genes included oxidative phosphorylation, glutathione metabolism, and ribosome. Immune cells, including gamma-delta T cells, activated mast cells, and M2 macrophages, exhibited differential infiltration between the DN group and the control group. A total of 23 lncRNAs targeting intersecting miRNAs of three key genes. There were 4 drugs associated with the three key genes. In this research, three key genes (GPX1, PPIF and VDAC1) mitochondrial and endoplasmic reticulum stress-related gene in DN were identified, providing a potential theoretical basis for DN treatment. However, this study still has certain limitations. This study only used a single dataset for analysis and validation, so the results of the study may not fully reflect the diversity of DN patients.