Related Experiment Videos
Molecular Markers and Diagnostic Model Specific for Barrett's Esophagus
Kexin Shen1, Shujuan Zhang1, Shurong Ma2
1Department of Gastrointestinal Colorectal and Anal Surgery, China-Japan Union Hospital of Jilin University, Changchun, China.
Summary
Researchers identified novel biomarkers for early Barrett's esophagus (BE) diagnosis. Key genes like CREB3L1, HNF1B, and IL35 were found, leading to an accurate XGBoost classifier for BE prediction.
Area of Science:
- Molecular biology
- Genomics
- Bioinformatics
Background:
- Barrett's esophagus (BE) progression biomarkers are understudied.
- Early diagnosis of BE is crucial but challenging.
- Novel molecular markers are needed for BE detection.
Purpose of the Study:
- To identify novel molecular markers for early Barrett's esophagus diagnosis.
- To develop a predictive model for BE using identified key genes.
Main Methods:
- Downloaded gene expression profiles (GSE100843) from Gene Expression Omnibus.
- Identified differentially expressed genes (DEGs) using limma, clustered them using mclust.
- Evaluated pathway/function enrichment and identified key genes (CREB3L1, HNF1B, IL35) via coexpression and genetic algorithms.
- Constructed an XGBoost classifier for BE prediction.
Main Results:
- Identified 2598 DEGs, clustered into nine gene sets.
- Found nine significantly deregulated functional and pathway terms in BE.
- Identified CREB3L1, HNF1B, and IL35 as key genes associated with BE.
- Achieved 93% accuracy on training and 87% on validation datasets with the XGBoost classifier.
Conclusions:
- CREB3L1, HNF1B, and IL35 may serve as novel biomarkers for Barrett's esophagus.
- The developed XGBoost classifier demonstrates high efficiency and robustness for BE diagnosis.
- This approach may aid future clinical diagnosis of Barrett's esophagus.