Identification of immune-related mitochondrial metabolic disorder genes in septic shock using bioinformatics and machine learning

Abstract

Purpose: Mitochondria are involved in septic shock and inflammatory response syndrome, which severely affects the life security of patients. It is necessary to recognize and explore the immune-mitochondrial genes in septic shock.

Methods: The GSE57065 dataset was acquired from the Gene Expression Omnibus (GEO) database and filtered by limma and the weighted correlation network analysis (WGCNA) to identify mitochondrial-related differentially expressed genes (MitoDEGs) in septic shock. The function of MitoDEGs was analyzed using the Gene Ontology (GO) analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA), respectively. The Protein-Protein Interaction (PPI) network composed of MitoDEGs was established using Cytoscape. Support Vector Machine Recursive Feature Elimination (SVM-RFE), Random Forest (RF), and Least Absolute Shrinkage and Selection Operator (LASSO) were used to identify diagnostic MitoDEGs, which were validated using receiver operating characteristic (ROC) analysis and Quantitative Real-time Reverse Transcription Polymerase Chain Reaction (qRT-PCR). Furthermore, the infiltration of immunocytes was analyzed using CIBERSORT, and the correlation between diagnostic MitoDEGs and immunocytes was explored using Spearman.

Results: A total of 44 MitoDEGs were filtered, and functional enrichment analysis showed they were associated with mitochondrial function, and the PPI network had 457 nodes and 547 edges. Four diagnostic genes, MitoDEGs, PGS1, C6orf136, THEM4, and EPHX2, were identified by three machine learning algorithms, and qRT-PCR results obtained similar expression levels as bioinformatics analysis. Furthermore, the diagnostic model constructed by the diagnostic genes had fine diagnostic efficacy. Immunocyte infiltration analysis showed that activated immunocytes were abundant and correlated with hub genes, with neutrophils accounting for the largest proportion in septic shock.

Conclusions: In this study, we recognized four immune-mitochondrial key genes (PGS1, C6orf136, THEM4, and EPHX2) in septic shock and designed a novel gene diagnosis model that provided a new and meaningful way for the diagnosis of septic shock.

Keywords: Bioinformatics; Differentially expressed genes; Machine learning; Mitochondria; Septic shock.

Fig. 1

Recognition of DEGs and WGCNA analysis results. (A) Differential gene volcano map; (B) Differential gene heatmap; (C) Sample clustering tree diagram; (D) Soft threshold and scale-free topology fitting index; (E) Combination module under cluster tree; (F) Module trait correlation heatmap. Red represents positive correlations, while blue represents negative correlations

Fig. 2

The functional enrichment analysis results and the PPI network composed of MitoDEGs. (A) Venn diagram of MitoDEGs; (B) GO analysis results; (C) KEGG analysis results; (D) Down-regulated pathway in septic shock (GSEA); (E) Up-regulated pathway in septic shock (GSEA); (F) PPI network constructed by MitoDEGs

Fig. 3

Identification of d-MitoDEGs using three machine learning algorithms. (A, B) Path maps of regression coefficients and cross-validation curves in the LASSO; (C, D) Change curves for prediction accuracy and error values for every gene in the SVM-RFE; (E) Significance of characteristic genes in the random forest algorithm; (F) Wayne map of the characteristic genes acquired by the three algorithms

Fig. 4

Differential expression of d-MitoDEGs and ROC curves. (A) The differences in d-MitoDEGs expression between the SS group and the control group in the training set; (B) Validation of the differences of d-MitoDEGs expression between the SS group and the control group; (C) Validation of the expression levels of d-MitoDEGs between the SS group and the control group using qRT-PCR experiment, *p < 0.05;**p < 0.01; ****p < 0.0001; (D) ROC curve of the diagnostic model in the training set; (E) ROC curve of the diagnostic model in the validation set

Fig. 5

Results of immune cell infiltration analysis; (A) Box plot of the proportions of immunocytes between the SS group and the control group, *p < 0.05;**p < 0.01; ***p < 0.001; ****p < 0.0001; (B) Heat map of immunocyte ratios between the SS group and the control group; (C) Stacked histogram of immunocytes; (D) Heat map of the correlation between immunocytes, with blue demonstrating the positive correlation and red demonstrating the negative correlation

Fig. 6

Correlation between immunocytes and d-MitoDEGs. (A) Heatmap of the correlation between d-MitoDEGs and immunocytes, with red demonstrating positive correlations and blue demonstrating negative correlations. *P < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; (B, C) Correlation scatter plots of PGS1 with CD8 ⁺ T cells and M0 macrophages; (D, E) Correlation scatter plots of C6orf136 with M0 macrophages and CD8 ⁺ T cells; (F, G) Correlation scatter plots of THEM4 with M0 macrophages and naive CD4 ⁺ T cells; (H, I) Correlation scatter plots of EPHX2 with M0 macrophages and naive CD4 ⁺ T cells