Identification of Oxidative Stress-Related Markers for Chronic Atrophic Gastritis Using WGCNA and Machine Learning

Western medical treatment and surgery are less effective in treating chronic atrophic gastritis, and there is no definitive treatment for chronic atrophic gastritis. Some studies suggest that oxidative stress regulation may play a role in the treatment of chronic atrophic gastritis, but the specific mechanism and related genes are not clear.

This study utilizes differential expression analysis, WGCNA, and machine learning to identify critical oxidative stress-associated biomarkers for CAG.

This study performed differential expression analysis and weighted gene co-expression network analysis (WGCNA) on datasets related to chronic atrophic gastritis (CAG). Intersection analysis of differentially expressed genes (DEGs), module genes, and oxidative stress-related genes was conducted, followed by functional enrichment analysis of overlapping genes. Core genes were screened using Support Vector Machine Recursive Feature Elimination (SVM-RFE), Least Absolute Shrinkage and Selection Operator (LASSO), and Random Forest algorithms. Perform immune infiltration analysis of core genes and construct a ceRNA network.

As a key gene, CASP3, highly expressed in CAG, demonstrated robust diagnostic potential. Immune infiltration analysis via the CIBERSORT method showed macrophages and T-cells as predominant cell types in CAG tissues, with CASP3 potentially regulating these through macrophage apoptosis and inflammatory cytokine release. A ceRNA network identified 15 miRNAs and 35 lncRNAs closely correlated with CASP3.

CASP3 could serve as a novel biomarker for CAG and provide insights into its pathogenesis.