Integrated multiple microarray studies by robust rank aggregation to identify immune-associated biomarkers in Crohn's...

Area of Science:

• Genomics
• Immunology
• Bioinformatics

Background:

• Crohn's disease (CD) is a complex autoimmune disorder with poorly understood molecular mechanisms.
• Identifying novel biomarkers is crucial for improving CD diagnosis and treatment.

Purpose of the Study:

• To identify novel diagnostic biomarkers for Crohn's disease using machine learning approaches.
• To investigate the molecular mechanisms and immune cell infiltration in CD pathogenesis.

Main Methods:

• Comprehensive analysis of Gene Expression Omnibus (GEO) datasets for CD.
• Robust Rank Aggregation (RRA) to identify differentially expressed genes (DEGs).
• Machine learning algorithms (SVM-RFE, RF, LASSO) for characteristic gene selection and validation via ROC curves and immunohistochemistry (IHC).

Main Results:

• 203 significant DEGs were identified between CD patients and controls.
• Machine learning identified key biomarkers including AQP9, LCN2, and NAMPT, validated in an external cohort.
• A diagnostic score based on AQP9, LCN2, and NAMPT expression levels was established.
• Elevated levels of dendritic cells, macrophages, and NK cells were observed in CD patients, linked to the IL-17 signaling pathway and humoral immune response.

Conclusions:

• Novel gene biomarkers (AQP9, LCN2, NAMPT) have been identified for CD diagnosis.
• The study provides insights into the molecular mechanisms and immune landscape of CD.
• The developed machine learning score offers a reliable tool for CD diagnosis.