Bioinformatics combined with machine learning analysis of diagnostic biomarkers related to immune infiltration in the pathogenesis of keloids

Methods

Keloid fibroblast microarray dataset was obtained from GEO database, weighted gene co-expression network(WGCNA) and differential gene analysis were constructed to screen differentially expressed genes (DEGs), Functional enrichment analysis was used to identify the main pathways, protein-interaction network combined three machine learning algorithms to further detect key genes, draw ROC feature curves to diagnose the value of key genes, and analyze the distribution of immune cells and the correlation between key genes and immune infiltrating cells.Results A total of 80 DEGs were identified by WGCNA combined with differential gene analysis, which were mainly enriched in the pathways of regionalization, skeletal system morphogenesis and embryonic skeletal system development.