Exploring the molecular characteristics of inflammatory bowel disease from the perspective of hypoxia-related genes

Figures

• 

  FIGURE 1. Technology roadmap.

• 

  FIGURE 2. Differential Gene Expression Analysis. (A) Volcano plot illustrating the differential expression analysis of IBD samples versus Control samples in the Combined GEO Datasets. (B) Venn diagram displaying the intersection between DEGs and HRGs in the integrated GEO Datasets (Combined Datasets). (C) Heatmap depicting the expression of HRDEGs across sample cohorts in the unified GEO Datasets. (D) Group comparison plot of HRDEGs between different sample cohorts within the unified GEO Datasets (Combined Datasets). (E) Correlation heatmap of HRDEGs, showing gene-gene relationships. IBD samples are shown in red, and Control (Control) samples are shown in blue. In the heat map, red denotes high expression and blue denotes low expression. In the group comparison map, the symbol ** is equivalent to P < 0.01, signifying high statistical significance. *** denotes a p-value <0.001, signifying high statistical significance.

• 

  FIGURE 3. Combined Datasets analysis between IBD and Control cohorts using GSEA. (A) Mountain plot illustrating six key biological functions identified through GSEA in the IBD cohort relative to the Control cohort within the Combined GEO Datasets. (B–G) GSEA results showing marked enrichment of all genes in the following pathways: WP_TGFBETA_RECEPTOR_SIGNALING (B), WP_MAPK_SIGNALING_PATHWAY (C), WP_IL6_SIGNALING_PATHWAY (D), KEGG_JAK_STAT_SIGNALING_PATHWAY (E), WP_CYTOKINES_AND_INFLAMMATORY_RESPONSE (F), and WP_INFLAMMATORY_BOWEL_DISEASE_SIGNALING (G). GSEA. The analysis applied criteria of adj.p < 0.05 and FDR (q value) < 0.25 for significance.

• 

  FIGURE 4. DM for IBD. (A) Forest plot illustrating the LRA of 23 HRDEGs. B-C. Visualization of the gene count yields the lowest error rate (B) and the highest accuracy (C) in the SVM algorithm. (D,E) DM plot (D) and variable trajectory plot (E) of the LASSO regression model. (F) Venn diagram showcasing the overlap between HRDEGs included in the LASSO regression model and those encompassed in the SVM model. (G) Boxplot showing the functional similarity analysis of model genes.

• 

  FIGURE 5. Diagnostic and Validation Analyses of IBD. (A) Nomogram depicting the impact of Model Genes on the DM for IBD in the Combined GEO Datasets. (B) DCA plot for assessing the clinical applicability of the IBD DM. (C) ROC curve for the RiskScore in the Combined GEO Datasets, evaluating diagnostic accuracy. (D) Diagnostic ROC curve of the RiskScore for the GSE48958 dataset. (E) Diagnostic ROC curve for the RiskScore in the GSE75214 dataset. (F) Diagnostic ROC curve for the RiskScore in the GSE179285 dataset. (G) Diagnostic ROC curves for BHLHE40, EDN1, and EGR1 in the Merged GEO Datasets. (H) Diagnostic ROC curves for ENG, HIF1A, and HK2 in the Merged GEO Datasets. (I) Diagnostic ROC curves for IL1B, IL6, and MMP2 in the Merged GEO Datasets. (J) Diagnostic ROC curves for MMP9, NOS2, and PPARG in the Merged GEO Datasets. (K) Diagnostic ROC curve for SPP1 in the Merged GEO Datasets. An AUC >0.5 indicates a molecule’s diagnostic potential, with AUC values closer to 1 representing superior diagnostic efficacy. AUC values between 0.5 and 0.7 suggest low accuracy, 0.7 to 0.9 indicate moderate accuracy, and AUC values above 0.9 indicate high accuracy.

• 

  FIGURE 6. Risk Cohort Immune Infiltration Analysis using the CIBERSORT Algorithm. (A) Group comparison of ICI between high-risk (High) and low-risk (Low) cohorts of IBD specimen. (B,C) Correlation analysis outcomes of ICI abundance in low-risk (Low) (B) and high-risk (High) (C) IBD samples. (D,E) Bubble plots depicting the correlation between ICI abundance and model genes in low-risk (D) and high-risk (E) IBD samples. The examination utilized single-sample Gene-Set Enrichment Analysis (ssGSEA). In the plots, “ns” indicates a p-value ≥0.05 (not statistically significant), while “***” indicates p-values <0.001, indicating high statistical significance. Red indicates the high-risk (High) cohort, and blue indicates the low-risk (Low) cohort of IBD samples.

• 

  FIGURE 7. Regulatory Network of Model Genes. (A) mRNA-miRNA Regulatory Network of model genes, where red nodes indicate mRNAs and blue nodes indicate miRNAs. (B) mRNA-TF Regulatory Network of model genes, where red nodes indicate mRNAs, blue nodes denote TFs, and purple nodes indicate genes that function as both mRNA and TF.