Fig. 3

Functional analysis of immune-related DEGs and construction of the random forest model. (A) Venn diagram shows 27 genes that were common to DEGs and the InnateDB datasets. (B) Bar chart represents the enriched terms of immune-related DEGs (The parameter is set to P-value < 0.01, minimum = 3, and enrichment factor > 1.5). (C) Network diagram shows the enriched terms of immune-related DEGs. Each enrichment term is a node, nodes with the same color share the same cluster-ID. (D) The plot displays the importance ranking of 25 immune-related DEGs in the random forest model. The abscissa represents the importance value, and the ordinate represents the genes name. (E) AUC distribution plot of the random forest model. The random forest model achieved a better AUC of 0.844 by incorporating the top 9 ranked immune-related DEGs. (F) PPI network constructed based on immune-related DEGs. Red dots represent upregulated DEGs, while green dots represent downregulated genes. The larger size of the dots indicates more interactions with other genes in the network. (G) The heatmap shows the correlation between immune-related DEGs and clinical indices. Red indicates negative correlations, whereas blue indicates positive correlations. The squares with “*” represent | correlation coefficient | > 0.3 and P-value < 0.05. “*”: P-value < 0.05, “**”: P-value < 0.01, “***”: P-value < 0.001. (H) The network diagram shows the top 20 significant correlations between clinical variables and immune-related DEGs