Comprehensive analysis of diagnostic biomarkers related to histone acetylation in acute myocardial infarction

Figures

• 

  Fig. 1. Identification of histone acetylation regulatory genes (ARGs)-related genes. (a) Gene set variation analysis (GSVA) analysis displayed the scores of ARGs of AMI and control samples in the GSE60993 dataset. (b) Weighted Gene Co-Expression Network to identify key modules related to scores of ARGs, including sample clustering dendrogram and trait heatmap to remove outliers, Analysis of the scale-free fit index (left) and the mean connectivity (right) for various soft-thresholding powers, cluster dendrogram of all DEGs based on similar expression patterns, and correlations heatmap between modules and traits were showed. (c) Venn diagrams for 2,739 ARGs-related genes

• 

  Fig. 2. Collection and functional enrichment analysis of key intersecting genes. (a) Heatmap and (b) Volcano plot of 996 differentially expressed genes (DEGs) between the AMI and control samples in GSE60993. (c) Heatmap and (d) Volcano plot of 66 DEGs between the AMI and control samples in GSE48060. The screening criteria are set to |Log2FC| > 0.5 and P < 0.05. (e) Venn diagrams for 25 intersecting DEGs in two AMI-related cohorts. (f) Venn diagrams for 18 intersecting ARGs-related DEGs in AMI. (g) The Gene Ontology (GO) analysis for intersecting ARGs-related DEGs. (h) The most enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) terms of intersecting ARGs-related DEGs. (i) The protein-protein interaction (PPI) network for intersecting ARGs-related DEGs

• 

  Fig. 3. Screening of five biomarkers in AMI. (a) Cross-validation for tuning parameter selection in the least absolute shrinkage and selection operator (LASSO) Cox model was shown and five characteristic genes were selected. (b) Receiver operating characteristic (ROC) curves of the LASSO model-based genes in the support vector machine recursive feature elimination (SVM-RFE) and the extreme gradient boosting (XGBoost) algorithms. (c) ROC analysis of five characteristic genes in the GSE60993 and GSE48060 datasets. (d) Boxplots for the expression levels of five characteristic genes in the GSE60993 and GSE48060 datasets. (e) Results of quantitative PCR (qPCR) for the expression levels of five characteristic genes in the clinical AMI and normal blood samples. * P < 0.05, ** P < 0.01, *** P < 0.001

• 

  Fig. 4. Construction of the nomogram and analysis of five biomarkers. (a) Nomogram was constructed based on five biomarkers. (b) Calibration curve of nomogram (C-index = 1). c-d) Bar chart of enriched canonical pathways of DEGs between the AMI and control samples in GSE60993 was exhibited using the Ingenuity Pathway Analysis (IPA), where orange represents activation, blue represents inhibition, in which the effect of biomarkers for the signalling pathway of calcium-induced T lymphocyte apoptosis was further mapped. e) Gene set enrichment analysis (GSEA) results of five biomarkers

• 

  Fig. 5. Immune related analyses and drug prediction targeted biomarkers. (a) Pearson correlation heatmap between five biomarkers and four differential expressed immune cells. (b) The structure of the highest scoring drug ATOGEPANT, and molecular docking results of targets and ATOGEPANT using AutoDock. (c) The lncRNA-miRNA-mRNA regulatory network diagram. Blue represents lncRNAs, green represents miRNAs, and red represent biomarkers