Integrative Bioinformatics Analysis and Machine Learning Reveal TSPAN7 Regulates the Involution of Infantile...

The involution of infantile hemangioma (IH) involves complex lipid metabolism reprogramming and immune modulation, but the underlying molecular mechanisms remain incompletely understood. This study aimed to identify key lipid metabolism-related genes that may drive IH involution using a bioinformatic perspective.

The GSE127487 dataset of IH was analyzed. Time-series gene expression profiles linked to involution were identified using the R package Mfuzz. Differentially expressed genes (DEGs) between proliferation and involution phases were overlapped with involution-associated temporal genes to identify key involution-related genes. Lipid-associated genes with relevance scores > 2 were retrieved from the GeneCards database. The overlap of these genes with key involution-related genes yielded lipid-related involution genes. Machine learning algorithms were applied to further screen hub genes. WGCNA was conducted to identify co-expression modules of the target hub genes during the involution phase, followed by GO and KEGG enrichment analyses. Immune cell infiltration was quantified using MCP-counter. Mediation analysis was performed to explore potential relationships between hub genes and immune cells.

Three hub genes, including CFH, TSPAN7 and MIR324, were finally identified. WGCNA revealed four modules co-expressed with TSPAN7 specifically during the involution phase. Functional enrichment analysis showed that these modules were mainly enriched in "natural killer cell-mediated cytotoxicity" and "leukocyte-mediated cell killing". MCP-counter analysis showed that NK cells were enriched in the proliferation phase. Mediation analysis revealed the direct effect of TSPAN7 on IH involution was non-significant (P = 0.474), while its indirect effect mediated by NK cells was significant (P = 0.012).

Three key lipid metabolism-related genes associated with IH involution (CFH, TSPAN7, and MIR324) were identified. Among them, TSPAN7 may be associated with IH involution through NK cells. These findings provide clues for further investigations into the roles of lipid-related genes and immune cells in IH involution.