Analyzing the Molecular Mechanism of Hydrocortisone Treatment for Scar For-mation Based on Transcriptomic Data and Machine Learning Algorithms
This study aimed to elucidate the molecular mechanism of hydrocortisone(HC)in the treatment of keloid scars through bioinformatics analysis,machine learning algorithms,and network pharmacology.Four keloid fibroblast(KFs)datasets were obtained from the GEO database.The two-dimensional structure of HC was retrieved from PubChem,and the pharmacophore model of HC was predicted using Pharmmapper.The intersection of HC target genes and differentially expressed genes(DEGs)derived a Venn diagram,yielding HC-related differentially expressed genes(HcDEGs).HcDEGs were subjected to GO and KEGG enrichment analysis using Metascape,and a protein-protein interaction(PPI)network was constructed.Machine learning algorithms were employed to identify the core HC-related differentially expressed genes(Core-HcDEGs).Enrichment analysis of co-expressed genes of Core-HcDEGs and their correlation with immune cell infiltration were performed.Changes in gene expression levels before and after HC treatment were compared using external data,and molecular docking was conducted using AutoDock.GO and KEGG analysis revealed that HcDEGs were enriched in multiple tumor and inflammation-related pathways.Five Core-HcDEGs were identified through machine learning algo-rithms.Enrichment analysis of co-expressed genes showed a high enrichment in inflammation and tumor-related pathways.Correlation a-nalysis of immune cell infiltration demonstrated a strong association between Core-HcDEGs and various immune cells.The expression of MMP2 and TEK was upregulated in KFs,while ALDH2 expression was downregulated.After HC stimulation,the expression of MMP2 and TEK decreased,while ALDH2 expression increased.Molecular docking results indicated stable binding effects between HC and MMP2,TEK,and ALDH2.HC exerts its therapeutic effects on keloid scars by modulating the expression of MMP2,TEK,and ALDH2.
NETL
NSTL
万方数据
