Identification and validation of biomarkers associated with cellular senescence and demethylation in acute myocardial infarction

Figures

• 

  Fig. 1. Screening of differentially expressed genes (DEGs). (a) Volcano plot of DEGs between the acute myocardial infarction (AMI) group and the normal group; (b) Heatmap of DEGs between the AMI group and the normal group; (c) Venn diagram of candidate genes, showing the overlap among DEGs, cellular senescence-related genes, and demethylation-related genes; (d) Circular plot of Gene Ontology (GO) enrichment analysis for candidate genes. The outer ring displayed different GO terms, with colored blocks representing different categories (yellow for biological process (BP) pink for cellular component (CC), blue for molecular function (MF)). The middle ring indicated the number of background genes under specific functional terms. The inner ring’s colored blocks represented the “Rich Factor,” a measure of gene enrichment in specific functional terms; (e) Circular plot of Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment pathways for candidate genes. The outer ring showed different KEGG metabolic pathways, the middle ring indicated the number of background genes under specific pathways, and the inner ring’s colored blocks represented the “Rich Factor” for measuring gene enrichment in…

• 

  Fig. 2. Screening of key genes. (a) Least absolute shrinkage and selection operator (LASSO) regression analysis. The left panel showed the Lasso coefficient spectrum, with the horizontal axis representing the logarithm of the regularization parameter λ (logλ) and the vertical axis representing variable coefficients; each curve depicted the coefficient trajectory of a variable as λ changed. The right panel showed the LASSO regression regularization path plot, with the horizontal axis representing logλ and the vertical axis representing model prediction error; the curve illustrated how prediction error changed as λ increased; (b) Boruta algorithm analysis, with the horizontal axis indicating gene names and the vertical axis showing feature importance scores; (c) Feature selection results using the support vector machine-recursive feature elimination (SVM-RFE) algorithm. The horizontal axis represented the number of selected variables, and the vertical axis represented cross-validation accuracy; the curve showed fluctuations in model accuracy as the number of variables changed; (d) Venn diagram of characteristic genes, showing the overlap of genes screened by three machine learning methods;…

• 

  Fig. 3. Gene set enrichment analysis (GSEA) of key genes (a) BCL3; (b) MMP9; (c) NAMPT; (d) TLR4.

• 

  Fig. 4. Immune microenvironment analysis. (a) Immune infiltration analysis, displaying stacked bar charts of different immune cell abundance proportions colored by cell type. Left: Normal group; Right: AMI group; (b) Box plots of immune cell infiltration differences between the normal and AMI groups. Red represented control samples, and blue represented disease samples. *** indicated p < 0.001, ** indicated p < 0.01, and ns indicated no significance; (c) Correlation analysis between differentially infiltrating immune cells, with red indicating positive correlation and green indicating negative correlation; (d) Correlations between key genes and differentially infiltrating immune cells, with red indicating positive correlation and blue indicating negative correlation. *** indicated p < 0.001, ** indicated p < 0.01, * indicated p < 0.05, and ns indicated no significance.

• 

  Fig. 5. Prediction network diagrams of key gene-targeted drugs. (a) lncRNA-miRNA-mRNA regulatory network, with left nodes representing mRNAs, middle nodes representing miRNAs, and right nodes representing lncRNAs; (b) Transcription factor (TF) regulatory network of key genes, with red nodes representing key genes and yellow nodes representing TFs targeted by key genes; (c) Network diagram of drugs and key genes, with left nodes representing key genes and right nodes representing predicted drugs for key genes; (d) GeneMANIA network diagram, where different colors represented distinct functional interaction relationships; (e) Friends analysis of key genes, with the vertical axis representing key genes and the horizontal axis representing functional similarity coefficients.

• 

  Fig. 6. Expression validation of key genes in clinical tissue samples. The ratio of AMI samples to control samples was 5:5. Statistical analysis was conducted using the t-test, and *indicated p < 0.05.