Bravo Roberto! what a great way to start your new lab back in Italy.

Looks super interesting. Congrats Nikos and team!

The exciting next question is what these regulatory sequences can tell us about the evolution of cell types across species. We’ll be working on this in the coming years.

This is similar to what @tschopplab.bsky.social recently reported for skeletogenic cells in vertebrates. These approaches can be particularly useful to study cell type ontogenetic relationships in non-genetically tractable species and without access to dev stages.
www.nature.com/articles/s41...

We observed that while gene activity groups cell type functionally, whereas CREs and regulatory seqs group them ontogenetically. E.g. mesenteric and tentacle retractor muscles, a dev convergence described by @alisongcole.bsky.social and @ulrichtechnau.bsky.social.

www.nature.com/articles/s41...

With this TF-motif collection, we defined important TF in every cell type (expressed and with high motif accessibility) and we predicted TF-CRE regulatory links. In the example, the predicted cnidocyte GRN.

A major challenge was inferring TF binding motifs. We first derived multispecies motif archetypes from known and de novo motifs, then assigned archetypes to TFs by combining phylogenetic analysis (@zolotarg.bsky.social) with correlations between TF expression and motif accessibility.

Then, we used different sequence models to predict CRE activty. With the crucial help of @steinaerts.bsky.social and @lukasmahieu.bsky.social, who introduced us to the world of DL sequence models. In this example, top motif grammars defining cnidocyte CREs:

We map cell type-specific accessible regulatory landscapes in adult and gastrula stages, identifying over 112,000 candidate cis-regulatory elements (CREs), representing different types of promoters and non-promoter elements.

Bravo @dudinlab.bsky.social!