SLC6A14 as a Key Diagnostic Biomarker for Ulcerative Colitis: An Integrative Bioinformatics and Machine Learning Approach.

Abstract

Ulcerative colitis (UC) is a chronic inflammatory bowel disease characterized by intestinal inflammation and autoimmune responses. This study aimed to identify diagnostic biomarkers for UC through bioinformatics analysis and machine learning, and to validate these findings through immunofluorescence staining of clinical samples. Differential expression analysis was conducted on expression profile datasets from 4 UC samples. Key biomarkers were selected using LASSO logistic regression, SVM-RFE, and Random Forest algorithms. The diagnostic performance of these biomarkers was evaluated using receiver operating characteristic (ROC) curves. Functional enrichment analysis assessed the biological functions of these biomarkers. The CIBERSORT algorithm was used to analyze immune cell infiltration. Regulatory networks for diagnostic markers were constructed. Additionally, immunofluorescence staining was performed on clinical samples to validate the expression levels of key biomarkers. Differential analysis identified 199 significantly differentially expressed genes. SLC6A14 was selected as a key diagnostic biomarker, demonstrating excellent diagnostic performance in training and validation sets (AUC values: 0.973, 0.984, and 0.970). Immune cell infiltration analysis revealed significant increases in Neutrophils and activated Mast cells in UC samples, whereas resting Mast cells were relatively downregulated. Furthermore, SLC6A14 showed strong correlations with various immune cells. The ceRNA network identified 22 lncRNAs and 10 miRNAs associated with SLC6A14. Immunofluorescence staining of clinical samples confirmed that SLC6A14 expression is significantly higher in UC patients compared to normal intestinal mucosa, and its expression increases with UC activity. SLC6A14 has been confirmed as a key diagnostic marker for UC, validated both through bioinformatics analysis and immunofluorescence staining of clinical samples. It maintains regulatory relationships with various non-coding RNAs and plays a significant role in the pathogenesis of UC through its interactions with immune cells.