Bioinformatics Analysis of the Cuprotosis Gene in Immune Infiltration of Chronic Kidney Disease

Background： Chronic kidney disease is currently a major public health challenge worldwide, and modeling based on gene expression profiling is essential to guide individualized treatment of the disease. According to recent studies, cuprotosis, one of the forms of death of cells, appears to contribute to the progression of various diseases. Therefore, the present study aims to explore clusters associated with cuprotosis genes in chronic kidney disease, delve into immune infiltration, and construct predictive models. Methods： The GSE37171 (GPL570) dataset was downloaded from the Gene Expression Omnibus for analyzing expression profiling and immune characterization of cuprotosis regulators in CKD. Samples were classified into different clusters based on cuprotosis-related genes (CRGs) of kidney disease. Differential expression pathways and biological functions among clusters were identified through gene set variation analysis. The weighted gene co-expression network analysis algorithm was adopted to identify specific differentially expressed genes of clusters. A machine learning model was built to construct and validate nomogram risk prediction maps. Results： A total of seven cuprotosis-related genes are differential genes between chronic kidney disease and control group, with differences in immune infiltration between the two groups. Two different clusters are identified based on the expression profiles of the cuprotosis-related genes. And according to the differences in immune infiltration, it is hypothesized that the prognosis of Cluster 2 may be worse. Cluster 1 may be associated with cellular lipid anabolism, fibrosis, signal reception, inflammation, and other processes, while Cluster 2 is more closely related to DNA replication and binding, cellular protein synthesis and transport, peroxisome, etc. The predictive performance of the four selected machine learning classifiers is compared and a prediction model is developed, which provides the highest predictive validity in the test cohort (AUC = 0.992), indicating satisfactory performance. The model is verified to exhibit good predictive efficacy. Conclusion： The study systematically illustrates the complex relationship between cuprotosis and chronic kidney disease and develops a promising predictive model to assess cuprotosis subtypes in patients with the disease, revealing the underlying molecular mechanisms that lead to its