AI x Medicine: Identifying CTSH Gene in Diabetic Foot Ulcer Using Bioinformatics and Machine Learning
Summary: This Twitter-highlighted publication from the Journal of Inflammation Research (2024) employs bioinformatics and machine learning to pinpoint CTSH as a critical extracellular matrix (ECM)-related gene in diabetic foot ulcers (DFU), validated in mouse models. Analyzing GEO datasets, the study identified CTSH’s upregulation in DFUs, correlating with inflammation and poor healing. ML models predicted CTSH’s diagnostic value (AUC 0.85), while knockdown in mice enhanced wound closure via reduced fibrosis and improved angiogenesis. This positions CTSH as a biomarker/target for personalized DFU therapies, addressing ECM dysregulation in diabetic wounds.
Key Highlights:
- Bioinformatics: CTSH upregulated in DFU datasets; enriched in ECM-receptor interaction and PI3K-Akt pathways.
- ML Prediction: Random forest model AUC 0.85 for DFU classification; CTSH as top feature.
- Mouse Validation: CTSH knockdown accelerated closure (day 14 vs 21), reduced inflammation (IL-6/TNF-α), enhanced collagen deposition.
- Clinical Relevance: Potential biomarker for refractory DFUs; therapeutic knockdown to improve healing.
- Publication: Journal of Inflammation Research (2024); DOI: 10.2147/JIR.S472345.
Keywords: CTSH gene, DFU biomarker, bioinformatics, machine learning, ECM regulation