Identifying potential three key targets gene for septic shock in children using bioinformatics and machine learning...

Area of Science:

• Pediatric critical care medicine
• Bioinformatics
• Computational biology

Background:

• Septic shock in children is a life-threatening infectious disease with high mortality.
• Understanding the mechanisms of death in pediatric sepsis is crucial for improving patient outcomes.
• Early prediction of mortality risk aids in clinical decision-making and treatment strategies.

Purpose of the Study:

• To identify key genes and pathways associated with fatal sepsis in children using bioinformatics and machine learning.
• To provide a theoretical basis for targeted therapies, including Traditional Chinese Medicine (TCM).
• To uncover novel genetic markers for predicting sepsis-related death in pediatric patients.

Main Methods:

• Analysis of gene expression profiles from the GEO database (GSE4607) comparing non-fatal and fatal pediatric sepsis cases.
• Identification of differentially expressed genes (DEGs) and enrichment analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways.
• Construction of protein-protein interaction (PPI) networks and application of machine learning algorithms (LASSO, Random Forest, SVM-RFE) to identify core diagnostic genes.

Main Results:

• 83 differentially expressed genes were identified, with 78 upregulated and 5 downregulated.
• Protein-protein interaction network analysis highlighted 17 candidate genes.
• Three core genes—CD163, MCEMP1, and RETN—were identified as having diagnostic value and potentially regulating pathways like complement and coagulation cascades.

Conclusions:

• CD163, MCEMP1, and RETN were identified as key genes associated with sepsis-related mortality in children.
• These genes may jointly influence critical biological pathways involved in sepsis pathogenesis.
• The findings offer a novel understanding of pediatric sepsis lethality and suggest potential therapeutic targets.