Prediction of key transcription factors during skin regeneration by combining gene expression data and regulatory network information analysis

Abstract

Purpose: Burn is one of the most common injuries in clinical practice. The use of transcription factors (TFs) has been reported to reverse the epigenetic rewiring process and has great promise for skin regeneration. To better identify key TFs for skin reprogramming, we proposed a predictive system that conjoint analyzed gene expression data and regulatory network information. Methods: Firstly, the gene expression data in skin tissues were downloaded and the LIMMA package was used to identify differential-expressed genes (DEGs). Then three ways, including identification of TFs from the DEGs, enrichment analysis of TFs by a Fisher’s test, the direct and network-based influence degree analysis of TFs, were used to identify the key TFs related to skin regeneration. Finally, to obtain most comprehensive combination of TFs, the coverage extent of all the TFs were analyzed by Venn diagrams. Results: The top 30 TFs combinations with higher coverage were acquired. Especially, TFAP2A, ZEB1, and NFKB1 exerted greater regulatory influence on other DEGs in the local network and presented relatively higher degrees in the protein–protein interaction (PPI) networks. Conclusion: These TFs identification could give a deeper understanding of the molecular mechanism of cell trans-differentiation, and provide a reference for the skin regeneration and burn treatment.