Back

Computational identification of key transcription factors for embryonic and postnatal Sox2+ dental epithelial stem cell

Shao, F.; Van Otterloo, E.; Cao, H.

2023-12-23 developmental biology
10.1101/2023.12.22.573158 bioRxiv
Show abstract

While many reptiles can replace their tooth throughout life, human loss the tooth replacement capability after formation of the permanent teeth. It was thought that the difference in tooth regeneration capability depends on the persistence of a specialized dental epithelial structure, the dental lamina that contains dental epithelial stem cells (DESC). Currently, we know very little about DESC such as what genes are expressed or its chromatin accessibility profile. Multiple markers of DESC have been proposed such as Sox2 and Lgr5. Few single cell RNA-seq experiments have been performed previously, but no obvious DESC cluster was identified in these scRNA-seq datasets, possible due to that the expression level of DESC markers such as Sox2 and Lgr5 is too low or the percentage of DESC is too low in whole tooth. We utilize a mouse line Sox2-GFP to enrich Sox2+ DESC and use Smart-Seq2 protocol and ATAC-seq protocol to generate transcriptome profile and chromatin accessibility profile of P2 Sox2+ DESC. Additionally, we generate transcriptome profile and chromatin accessibility profile of E11.5 Sox2+ dental lamina cells. With transcriptome profile and chromatin accessibility profile, we systematically identify potential key transcription factors for E11.5 Sox2+ cells and P2 Sox2+ cells. We identified transcription factors including Pitx2, Id3, Pitx1, Tbx1, Trp63, Nkx2-3, Grhl3, Dlx2, Runx1, Nfix, Zfp536, etc potentially formed the core transcriptional regulatory networks of Sox2+ DESC in both embryonic and postnatal stages.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.