Identification of key circadian rhythm genes in skin aging based on bioinformatics and machine learning.

Skin aging is often accompanied by disruption of circadian rhythm and abnormal expression of circadian rhythm-related genes. In this study, we downloaded skin aging expression datasets from the GEO database and utilized bioinformatics and machine learning methods to explore circadian rhythm genes and pathways involved in skin aging, revealing the pathological and molecular mechanisms of skin aging. Results showed that 39 circadian rhythm-related genes (CRGs) were identified in skin aging, and these CRGs were enriched in signaling pathways such as glucagon signaling pathway, insulin resistance, thyroid hormone signaling pathway, and adipocytokine signaling pathway. Three key skin aging-related CRGs, SIRT1, ARNTL, and ATF4, were identified based on machine learning. Additionally, we found that skin aging was associated with infiltration of immune cells including NK cells activated, Macrophages M1, Mast cells resting, T cells CD4 memory activated, and Macrophages M2, and the expression of the three key skin aging-related CRGs was correlated with these immune cells. Finally, SIRT1, ARNTL, and ATF4 were all down-regulated in skin aging and had a good ability to distinguish young skin tissue from aging skin tissue. In conclusion, three key CRGs, including SIRT1, ARNTL, and ATF4, which are closely related to skin aging, were obtained based on bioinformatics and machine learning technology screening. These three key CRGs were potential risk genes for skin aging and also associated with changes in the immune microenvironment in skin aging. The language used in this paragraph follows the guidelines for scientific writing specified by SCI, making it clear and concise.