<![CDATA[Acne (Acne vulgaris) is a chronic skin disease affected by Cutibacterium acnes infection and inflammatory pathways that trigger innate immune responses, such as inflammasome activation. The expression of inflammation-related genes plays a critical role in acne pathogenesis and immune modulation. This study aims to identify compounds from tea leaves (Camellia sinensis var. assamica) that can treat acne by influencing the expression of inflammatory-related genes through in silico analysis. The GSE6475 dataset was utilized to identify differentially expressed genes (DEGs) between acne-affected and normal skin samples (each group n=6). A total of 573 DEGs were identified and mapped to the KEGG inflammatory pathway. The hub gene analysis results showed six genes, including CXCL1, STAT1, and PIK3 (adj. P-value < 0.05). These key genes were then used to cross-validate skin grouping with acne lesions and normal skin. The structure of compounds (natural products) in tea leaves (C. sinensis var. assamica) was obtained from the PubChem database, and their activity against target proteins associated with the identified key genes was predicted using the SkelSpheres descriptor and Support Vector Regression method. This quantitative structure–activity relationship (QSAR)-based machine learning approach was selected because it enables high-throughput prediction of inhibitory potential using chemical descriptors and experimentally derived bioactivity data, providing broader predictive power than conventional molecular docking or molecular dynamics, which rely mainly on structural and energetic estimations. The in-silico prediction results showed that compounds such as theobromine, assamsaponin, procyanidin, and caffeine have exhibited good predicted activity (IC₅₀ ranging from 1.125 to 1.320 μM) as potential inhibitors of the PI3K/Akt pathway, which is known to play a role in the pathogenesis of acne. ]]>
Copyrights © 2025
