Fig 1.
Study flow chart.
Fig 2.
Identification of differentially expressed genes (DEGs).
(A) The heatmap of DEGs between control samples and diabetic nephropathy (DN) samples. The abscissa represents the clustered samples, and the ordinate represents DEGs. Red indicates up-regulated expression, while blue indicates down-regulated expression. (B) Volcano of DEGs. (C) Gene Ontology (GO) enrichment analysis of up-regulated genes in DN samples. (D) Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of up-regulated genes in DN samples. (E) Gene Ontology (GO) enrichment analysis of down-regulated genes in DN samples. (F) KEGG enrichment analysis of down-regulated genes in DN samples.
Fig 3.
Immune microenvironment of diabetic nephropathy (DN) samples.
(A) Differences of immune cell subtypes between control samples and DN samples. (B) Differences of stromal score, immune score and ESTIMATE score between control samples and DN samples. (C) The correlation between immune score and immune cell subtypes. Different colors indicate different correlation coefficients. (D) The correlation between immune score and cuproptosis genes. Different colors indicate different correlation coefficients. (E) The correlation between immune cell subtypes and cuproptosis genes. * represents P < 0.05, ** represents P < 0.01, *** represents P < 0.001.
Fig 4.
Consensus clustering base on expression of cuproptosis genes.
(A) Differences of cuproptosis genes between control samples and diabetic nephropathy (DN) samples. (B) The cumulative distribution function (CDF) curves for different values of k. Different colors represent different K values. (C) Consensus clustering matrix with K = 2. (D) Principal component analysis of DN samples based on gene expression. (E) Heatmap of cuproptosis genes between cluster C1 and cluster C2. (F) Differences of cuproptosis genes between cluster C1 and cluster C2. (G) Gene set variation analysis of cluster C1 and cluster C2. (H) Differences of immune cell subtypes between cluster C1 and cluster C2. B Differences of stromal score, immune score and ESTIMATE score between cluster C1 and cluster C2.
Fig 5.
Identification of key genes of different phenotypes.
(A) Dynamic tree cut to get modules related to cluster C1 and cluster C2. (B) The correlation between modules and clusters. Different colors indicate different correlation coefficients. The value in parentheses represents P-value. (C) Dynamic tree cut to get modules related to control and DN samples. (D) The correlation between modules and control and DN sample sets. Different colors indicate different correlation coefficients. The value in parentheses represents P-value. (E) Intersection of differentially expressed genes and phenotypic key genes.
Fig 6.
Construction and evaluation of RF, SVM, GLM, and XGB machine models.
(A) Boxplots showed the residuals of each machine learning model. Red dot represented the root mean square of residuals (RMSE). (B) ROC analysis of four machine learning models based on 5-fold cross-validation in the testing cohort.
Fig 7.
Nomogram construction and evaluation.
(A) Nomogram for risk of disease. (B) Calibration curve for nomogram. (C) Decision-making tree curve for nomogram. (D) Receiver operating characteristic (ROC) curve for nomogram of training cohort. (E) ROC curve for nomogram of test cohort.
Fig 8.
The relationship between key genes and immune microenvironment.
(A and B) DCXR and HRSP12 expression between control samples and DN samples. (C and D) DCXR and HRSP12 expression between cluster C1 and cluster C2. (E and F) The correlation between DCXR and HRSP12 and immune score. (G) The correlation between DCXR and HRSP12 and immune cell subtypes.
Fig 9.
(A) The expression of DCXR and HRSP12 was higher in the DN model; (B) The expression of DCXR and HRSP12 was decreased compared with the normal control after knockout; (C) CCK8 experiment.