Bioinformatics analysis and validation of novel biomarkers and competitive endogenous RNA networks involved in pyroptosis in diabetic nephropathy

Figures

• 

  Fig. 1. Flowchart of this study program.

• 

  Fig. 2. Identification of DEPRGs. (A) Volcano plot of 1429 DEGs in the GSE30529 dataset. It contains 909 significantly upregulated genes, represented by red triangles, and 520 significantly downregulated genes, represented by green triangles, whereas gray triangles represented stably expressed genes. (B) A veen plot of intersecting genes from PRGs and DEGs. The blue circle represents the number of PRGs, the red circle represents the number of DEGs, the intersecting area between blue circle and red circle was DEPRGs (C) Heatmap of 22 DEPRGs in the GSE30529 dataset which contains 2 significantly downregulated genes (VEGFA, MST1) and other 20 genes with significantly upregulated. The red bars represent control specimens from human kidneys and the green bars represent specimens from DN patient kidneys. The squares with the color from blue to red represent the correlation coefficient from negative to positive. (D) The correlation analysis of 22 DEPRGs by Heatmap.The squares with the color from blue to red represent the correlation coefficient from negative to positive.

• 

  Fig. 3. GO and KEGG enrichment analysis of 22 DEPRGs. (A) Bubble plot of enriched GO (BP, CC, MF) terms. The dots from small to big represent the gene radio from low to high. The dot with color from blue to red represents p adjust form high to low. (B) Bar plot of enriched KEGG terms. The area of the bars represents the gene counts. The bars with color from blue to red represent q adjust form high to low. (C) Chordal graph of enriched GO terms. The left side displays genes, while the right side depicts significant enriched GO terms. The connection between genes and GO terms refers to that genes were enriched in related GO terms. (D) Eleven diagrams of enriched GO terms shows the specific genes involved in each GO terms. Blue dots on the left indicate genes, while the right side table was the description of the GO terms. Z-score from light red to dark red represents the genes expression in GO terms from decreasing to increasing. The GO and KEGG imagery were constructed as per the citation guidelines: www.kegg.jp/kegg/kegg1.html.

• 

  Fig. 4. Construction of PPI Networks and Screening of Hub DEPRGs.The node represents the genes; the edge represents the relationship between them. (A) PPI network of 22 DEPRGs. (B) Top 15 candidate Hub DEPRGs were screened by “MCC” algorithms of CytoHubba plug-in from Cytoscape, Hub DEPRGs were arranged in descending order from red to yellow based on degree value. (C) Cross-validation plots of LASSO regression analysis. The top X-axis represents the number of variables in model decreasing from left to right, the bottom X-axis represents log(λ) value. The Y-axis of binomial deviance represents the degree of difference between the predicted value of the model and the true value. The smaller the error rate, the better the model. The left dotted line (lambda.min) represents the value of X-axis corresponding to the minimum error rate. Six candidates of Hub DEPRGs were predicted by the minimum cross-validation error with the optimal log(λ) value. (D) The LASSO coefficient distribution plot of the six Hub DEPRGs was applied to identify the eigenvalues of the constructed diagnostic signal. Each curve corresponds to a candidate Hub DEPRGs. (E) The SVM-RFE algorithm identified the latent Hub…

• 

  Fig. 5. Validation of Hub DEPRGs in DN. (A) Expression box plot of the 5 Hub DEPRGs in the validation dataset GSE142025. (B) Expression box plot of the 5 Hub DEPRGs in the validation dataset GSE96804. The blue bars represent control specimens from normal individuals, and the red bars represent specimens from DN patients. (C–F) ROC curves of CASP1 (C), TXNIP (D), IRF9 (E) and TRAF3 (F) in the training dataset GSE30529. ROC curves the represents receiver operator characteristic curves; AUC represents area under the curve; CI represents credibility interval.

• 

  Fig. 6. The landscape of immune infiltration in DN. (A) Bar plot shows the composition of 22 types of immune cells across samples. The different colors of the bars represent different immune cells. The area of the bar represents the percentage content of different immune cells in each samples (B) Correlation heatmap of 22 types immune cells. Red indicates positive correlation, and blue indicates negative correlation. (C) The violin plot of the immune cell proportions in both DN and normal groups.The comparisons of the immune cell infiltration level between DN and normal groups. The blue bars represent control specimens from normal individuals, and the red bars represent specimens from DN patients.

• 

  Fig. 7. Correlation between immune cells and Hub DEPRGs. Correlation between CASP1 (A), TXNIP (B), IRF9 (C), TRAF3 (D) and infiltrating immune cells in DN and normal sample groups.The size of the dots from small to big represents the degree of correlation coefficient from low to high and the colour of the dots from yellow to green represents the p-value from high to low. The right side of the numbers depict p-value, with red color indicating a p-value less than 0.05.

• 

  Fig. 8. Screening of DEmiRNAs, DElncRNAs and construction of ceRNA regulatory network. (A) Volcano plot of 674 DEmiRNAs in the GSE51674 dataset. (B) Volcano plot of 308 DElncRNAs in the GSE218344 dataset. The green triangle indicates down-regulation of gene expression, the red triangle indicates up-regulation of gene expression. (C) CeRNA networks. The orange diamond represents mRNAs, the red elliptical represents miRNAs and the green rectangle represents lncRNAs. The gray lines indicate the interactive relationship between the three and the validation of ceRNA networks was constructed. (D) Experimentally validated ceRNA network. The lines with different colors indicate their interactive relationship.

• 

  Fig. 9. Establishment of DN model rats and validation of the ceRNA networks. (A) UTP, SCr and BUN levels in rats. (B) Pathologic morphology of rat kidney tissues observed by HE, PAS and TUNEL staining. (C) Protein levels of NLRP3, pro-casp1, cleaved-casp1, GSDMD, GSDMD-N, IRF9, TXNIP and TRAF3 by Western blotting, and quantitative analysis in rat kidney tissues (The eight target gene bands originated from different gels, and the gels/blots were cropped. Samples derive from the same experiment and that gels/blots were processed in parallel. Original blots/gels are presented in Supplementary Fig. S2). (D) The relative expression of mRNAs and miRNAs in rat kidney tissues was detected by qRT-PCR. P value was shown as: * p < 0.05; ** p < 0.01.

• 

  Fig. 10. Validation of the ceRNA networks in clinical samples. The relative expression of mRNAs (A), miRNAs (B) and lncRNAs (C) in human blood samples were detected by qRT-PCR. P value was shown as: * p < 0.05; ** p < 0.01.