Identification of Ferroptosis-Related Biomarkers in Alopecia Areata Through Machine Learning

This study explored the role of ferroptosis, a type of cell death associated with oxidative stress, in alopecia areata (AA), a common hair loss condition. By analyzing transcriptome data from AA patients and controls, the researchers identified four ferroptosis-related genes (SLC40A1, LCN2, CREB5, and SLC7A11) as potential biomarkers for AA. They used machine learning techniques, including SVM-RFE and LASSO regression, for feature selection and constructed a nomogram to predict AA with high accuracy (AUC = 0.9052). The expression of these genes was confirmed to be reduced in AA patients through immunofluorescence, with significant differences observed for SLC40A1 and CREB5. These genes were primarily localized to the outer root sheath and near the sebaceous glands. The study's integration of immune cell infiltration analysis and machine learning provides new insights into AA diagnostics and therapeutics, highlighting the connection between AA and ferroptosis.