Clinical and experimental proof suggests that hypertension nephropathy (HN) is a chronic inflammatory disease. Our study aims to disclose the role of immune-related genes in the progression of HN. Using the Gene Expression Omnibus (GEO) database, two human HN gene expression datasets (GSE37455 and GSE37460; n = 35) along with the relevant controls (n = 43) could be as the discovery metadata to analyze for differentially expressed genes (DEGs) in HN. Three different machine-learning algorithms were integrated to screen immune-related genes in HN. Receiver-operating characteristic (ROC) curves were generated to estimate diagnostic efficacy. The diagnostic value and expression levels of these candidate genes were validated in the GSE104954 dataset (20 HN patients and 3 controls). Single sample gene set enrichment analysis (ssGSEA) was used to evaluate immune cell infiltrations, and immune checkpoints were quantified. The expression of the potential genes was confirmed in vivo. As a result, a total of 220 DEGs were identified between HN and control samples in these datasets, of which 52 were immune differential genes. The magenta module in WGCNA was the highest correlation. Two immune-associated genes GBP2 (guanylate binding protein 2) and TMSB10 (Thymosin β10) for HN were obtained after the intersection of genes screened by machine learning. The expression levels of GBP2 and TMSB10 were validated using discovery and validation cohort data sets. Following the ssGSEA analysis, we identified potential immune cell types in HN patients, as well as revealed the correlation between immune-related genes (GBP2 and TMSB10) and immune cells. Furthermore, the mRNA and protein levels of GBP2 and TMSB10 in vivo were consistent with the bioinformatics analysis which confirms the accuracy of our analysis. Our results demonstrated that GBP2 and TMSB10 are promising immune-related genes for the diagnosis of HN, which may help in the development of more precisely tailored HN immunotherapy.