Identification MNS1, FRZB, OGN, LUM, SERP1NA3 and FCN3 as the potential immune-related key genes involved in ischaemic cardiomyopathy by random forest and nomogram

Figures

• 

  Figure 1. A flow chart for analysis. DEGs, differentially expressed genes; GO, gene ontology annotation; KEGG, kyoto encyclopedia of genes and genomes pathway enrichment analyses; DO, disease ontology analysis; RF, random forest; MNS1, meiosis-specific nuclear structural 1; FRZB, frizzled-related protein; OGN, osteoglycin; LUM: lumican; SERPINA3: serpin family A member 3; FCN3: ficolin-3.

• 

  Figure 2. Differentially expressed genes (DEGs) in ICM and healthy samples. (A) Volcano plot of the 39 DEGs. Red dots represent significantly upregulated genes, and blue dots represent significantly downregulated genes. (B) Heatmap of the 39 lipid-related DEGs in ICM and control samples. Red blocks indicate high-expression genes, and blue blocks indicate low-expression genes.

• 

  Figure 3. Functional enrichment analyses for DEG genes in the training set. (A) Gene Ontology (GO) enrichment analysis. (B) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. (C) Disease Ontology (DO) analysis.

• 

  Figure 4. The results of the top 8 DEGs screened by random forest. (A) The plot of performance in log scale against epoch number. (B) The importance of the top 8 DEGs ranked by the mean decrease in accuracy. (C) Heatmap of the top 8 genes. Red blocks indicate high-expression genes, and blue blocks indicate low-expression genes.

• 

  Figure 5. Establishment and verification of the ICM diagnostic signature and nomogram. (A) Eight predictive nomograms of genetic diagnosis. (B) Calibration curve in the training set. (C) Calibration curve in the testing set. (D) Decision curve analysis (DCA) in the training. (E) DCA in the testing set. (F) Clinical impact curve in the training set. (G) Clinical impact curve in the testing set. (H) ROC curve in the training set. (I) ROC curve in the testing set.

• 

  Figure 6. Pattern of immune cell subtype infiltration in the training set. (A) Heatmap of the 22 immune cell proportions in each sample. (B) Correlation heatmap of the 22 immune cells. The deeper the red, the stronger the positive correlation. The darker the pink, the stronger the negative correlation. (C) Representative violin plot showing the different fractions of infiltrated immune cells.

• 

  Figure 7. Correlation between 8 key genes and infiltrated immune cells. Representative graphs showing the correlation between the infiltrated immune cells and SERP1NA3 (A), SLCO4A1 (B), FCN3 (C), FRZB (D), LUM (E), MNS1 (F), OGN (G) and RNASE2 (H). The correlation strength is shown by the dot size; the p values are expressed as the changes in dot colour, and p < 0.05 was considered statistically significant. SERPINA3: serpin family A member 3; SLCO4A1, solute carrier organic anion transporter family member 4A1; RNASE2, ribonuclease A family member 2; FCN3, ficolin-3; FRZB, frizzled-related protein; LUM, lumican; MNS1, meiosis-specific nuclear structural 1; OGN, osteoglycin.

• 

  Figure 8. External validation of the key genes in the clinical samples. The expression levels of MNS1 (A), FRZB (B), OGN (C), LUM (D), SERPINA3 (E), RNASE2 (F), FCN3 (G) and SLCO4A1 (H) in the clinical samples. MNS1, meiosis-specific nuclear structural 1; FRZB, frizzled-related protein; OGN, osteoglycin; LUM: lumican; SERPINA3: serpin family A member 3; FCN3: ficolin-3; SLCO4A1, solute carrier organic anion transporter family member 4A1; RNASE2, ribonuclease A family member 2.

• 

  Figure 9. Receiver operating characteristic (ROC) curve analysis. ROC curve analysis of MNS1 (A), FRZB (B), OGN (C), LUM (D), SERPINA3 (E), RNASE2 (F), FCN3 (G) and SLCO4A1 (H) in the clinical samples. MNS1, meiosis-specific nuclear structural 1; FRZB, frizzled-related protein; OGN, osteoglycin; LUM: lumican; SERPINA3: serpin family A member 3; FCN3: ficolin-3; SLCO4A1, solute carrier organic anion transporter family member 4A1; RNASE2, ribonuclease A family member 2.