Machine learning identifies molecular targets of Di (2-ethylhexyl) phthalate in pulmonary arterial hypertension

Figures

• 

  FIGURE 1. Flow-chart of datasets analysis in this paper.

• 

  FIGURE 2. Identification of DEHP target proteins. (A) Chemical structure of DEHP. (B) Target prediction using CHEMBI, PharmMapper, and SwissTargetPrediction.

• 

  FIGURE 3. Identification of PAH-related target genes. (A) Volcano plot shows DEGs based on logFC and significance. Red dots are upregulated, green dots are downregulated, and grey dots are non-significant. (B) Module-trait relationships heatmap shows correlations between WGCNA-identified modules and sample traits (Control vs. Treatment). Values in boxes indicate correlation coefficients and p-values. (C) Venn diagram shows DEGs (purple) and WGCNA modules (yellow), with brown indicating common genes from both methods. (D) Heatmap shows DEG expression patterns across samples. Red indicates upregulation, blue indicates downregulation. (E) Gene dendrogram from WGCNA shows hierarchical clustering based on co-expression. Module colors in the lower panel represent different gene modules.

• 

  FIGURE 4. Identification of DEHP-associated disease targets in PAH. (A) Venn diagram compares genes linked to DEHP exposure (purple) and PAH (yellow), with 60 overlapping genes (2.8%). (B) PPI network visualizes interactions among overlapping genes. Red nodes = upregulated, blue nodes = downregulated, edges = predicted interactions. (C) KEGG analysis shows enriched pathways for overlapping genes. X-axis = gene ratio, dot size = gene count, color gradient = adjusted p-value (red = higher significance).

• 

  FIGURE 5. Identification of core genes in DEHP-induced PAH. (A) Model performance comparison: heatmap shows AUC values for various models across cohorts. Left column = models, right column = AUC (higher = better). Colors indicate cohort sources. (B) ROC Curves: ROC curves for key genes (ALKBH2,AOC2,BCL2L10,CTBP2, DNM2, ERLIN2, HPS6,PON2,RABGGTA,SLC4A7,SORT1,PDE4D). X-axis = false positive rate, Y-axis = sensitivity. AUC indicates predictive performance. (C) Volcano plot: volcano plot shows DEGs. X-axis = logFC, Y-axis = -log10(P-value). Red = upregulated, green = downregulated, with key genes labeled. (D) Feature importance ranking: a bar graph ranks top genes by feature importance. Larger bars = greater contribution to the model. (E) Violin plot: violin plot shows gene expression distributions across conditions. Width = data density, colors = expression levels. (F) SHAP Summary Plot: SHAP summary plot shows gene contributions to predictions. Negative SHAP = lowering effect, positive = increasing effect.