Characterizing Duodenal Immune Microenvironment in Functional Dyspepsia: An AutoML-Driven Diagnostic Framework

Area of Science:

• Gastroenterology
• Immunology
• Computational Biology

Background:

• Functional dyspepsia (FD) is a prevalent gastroduodenal disorder with an unclear pathogenesis.
• Emerging evidence implicates duodenal immune activation in FD development.

Purpose of the Study:

• To identify immune-related diagnostic biomarkers for functional dyspepsia.
• To develop a high-performance diagnostic model for FD using automated machine learning.
• To elucidate the biological mechanisms underlying FD pathogenesis.

Main Methods:

• Mendelian randomization and genome-wide association studies (GWAS) identified FD-associated genes.
• Differential gene expression analysis and immune infiltration assessment were performed on patient data.
• Weighted gene co-expression network analysis (WGCNA) and maximal clique centrality (MCC) identified key genes.
• Automated machine learning (AutoGluon) and logistic regression developed a diagnostic model.
• Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Set Enrichment Analysis (GSEA) explored biological pathways.
• FD rat models were used for validation.

Main Results:

• GWAS revealed 259 genes linked to FD, enriched in immune and inflammatory pathways.
• Expression analysis showed altered immune cell populations, including increased plasma cells and decreased regulatory T cells.
• Nine biomarkers demonstrated high diagnostic performance (AUC=0.94).
• Automated machine learning models (LightGBMLarge, XGBoost) were identified as optimal for clinical use.
• GSEA and GO/KEGG analyses linked high-risk FD to immune-inflammatory pathways, cell adhesion, oxidative stress, and nutrient metabolism.

Conclusions:

• This study proposes novel immune-related diagnostic biomarkers for functional dyspepsia.
• The findings provide insights into FD pathogenesis and potential therapeutic targets.
• An automated machine learning model offers accurate identification of high-risk FD patients and characterizes biological changes.