Integrated gene expression profiling analysis reveals SERPINA3, FCN3, FREM1, MNS1 as candidate biomarkers in heart failure and their correlation with immune infiltration

Figures

• 

  Figure 1. Two-dimensional PCA cluster plot before and after sample correction and volcano map of DEGs. (A) PCA plot of the sample batching effect between chips was eliminated before SVA correction. (B) PCA plot of the instance batching effect between chips was eliminated after SVA correction. (C) Volcano map of DEGs; genes that were up-regulated differentially are represented by the red dots, whereas genes that were not significantly different are represented by black dots, and down-regulated differential genes are represented by green dots. PCA, Programable Counter Array; DEG, differentially-expressed gene; SVA, system verilog assertion.

• 

  Figure 2. Pathway analysis of differential genes. We use the Metascape database, annotation, visualization, GO, and KEGG analysis of differential genes. (A) Represents the enriched GO and KEGG pathways, and (B) represents the pathways. (C) PPI network diagram of differential genes. The PPI network diagram derived from distinct genes was produced utilizing String data in order to build a gene interaction network and show it using the Cytoscape software program. GO, gene ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein protein interaction.

• 

  Figure 3. Diagnostic factor screening and validation. (A) LASSO model. (B) The LASSO logistic regression approach was used to search for diagnostic criteria with the least amount of shrinkage and selection. Different colors represent different genes. (C) The Boruta algorithm. LASSO, least absolute shrinkage and selection operator; The Boruta algorithm, elimination algorithms were then employed to screen and validate the HF diagnostic variables.

• 

  Figure 4. External data sets verified the core genes.

• 

  Figure 5. ROC diagnostic efficacy of core genes. The ROC curve of diagnostic efficacy verification; the higher the AUC value, the better diagnostic efficacy. ROC, receiver operating characteristic; AUC, area under curve.

• 

  Figure 6. Correlation between SERPINA3, FCN3, FREM1, MNS1 and infiltrating immune cells. (A) Correlation between SERPINA3 and infiltrating immune cells. (B) Correlation between FCN3 and infiltrating immune cells. (C) Correlation between FREM1 and infiltrating immune cells. (D) Correlation between MNS1 and infiltrating immune cells. The size of the points indicates the force of the link between the genes and immune cells; the more significant the issues, the stronger the correlation, and the smaller the issues, the weaker the correlation. The P value, the greener the color, the lower the P value, and the purpler the color, the higher the P value. P<0.05 indicated statistical significance.

• 

  Figure 7. GSVA analysis of key genes. In order to identify the pathway that shows significant variation between samples, the median value of gene expression was used as the tangent point, and gene expression levels were placed into two groups based on their positions: those with high expression in the high-expression group and those with low expression in the low-expression group. This method was used to show the molecular mechanism of the core genes that are linked to disease progression. GSVA, gene set differentiation analysis.

• 

  Figure 8. Correlation diagram of expression regulation of core genes. An interaction map of positive and negative related core genes was drawn; green represents negative correlation and red represents positive correlation.