TFCP2L1 as a potential diagnostic gene biomarker of Keloid given its association with immune cells-a study based on machine learning and RNA sequence

Abstract

Keloids represent a specific type of dermal tumor, where delayed treatment greatly heighten risks of recurrence. This study thus seeks to identify key genes for early keloid diagnosis and compare how immune cell infiltration between keloid lesion skin (KLS) and keloid adjacent non-lesional skin (KNL) differ. Utilizing two GEO datasets, which encompassed KLS (n=8) and KNL (n=6) samples, 564 differentially expressed genes (DEGs) were identified, with a primary emphasis on inflammation- and fibrosis-related pathways. The gene TFCP2L1 was pinpointed from the intersection of LASSO and SVM-RFE algorithms, exhibiting downregulation in KLS when juxtaposed with KNL. To aid prognosis and identification of potential immunotherapy targets, the CIBERSORT algorithm was employed to assess prelative proportions of immune cells. In the KNL samples, there was an observed increase in follicular helper T cells, activated NK cells, and activated dendritic cells. Conversely, M1 macrophages were elevated in KLS samples. TFCP2L1 exhibited a positive correlation with activated NK cells and activated dendritic cells, while it showed a negative correlation with M1macrophages. These findings highlight the intertwined roles of TFCP2L1 and immune cell infiltration, lending to TFCP2L1’s potential as a diagnostic biomarker for keloids and highlights how TFCP2L1 might influence keloid development through immune cell populations.

Keywords

• Big data analytics
• Machine learning
• Diagnostic gene biomarker
• Keloid