iiinova — where your digital self keeps evolving

Pedro Rocha @pedroprocha.bsky.social · 08/09/2025

🚨🚨🚨 Please repost We are looking for postdocs to join our lab at NIH. Apply: www.nichd.nih.gov/research/atNICHD/Investigators/rocha/apply Learn more about training at NIH : www.training.nih.gov/research-tra...

02524

Pedro Rocha @pedroprocha.bsky.social · 21/07/2025

It had been 18 years since i last "commissioned" my college buddy @pedromics.bsky.social Our paper showing dramatic phenotypes upon loss of a single CTCF motif gave us the best excuse to do it again. We recruited MC hammer to deliver the main message: U Can't Touch This! tinyurl.com/4rz8kusj

2254

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

Other boundaries of the domain? 3' was also lethal...but it's complicated. Check the paper. Biggest surprise was that even though Fgf3 and Fgf4 have opposite chromatin marks in neighboring cells of the blastocyst, the CTCFs between them are not required for this and mice are fine.

121

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

What makes this boundary able to insulate Ano1 enhancers so well, when many other boundaries cannot? To our surprise you can delete the 3 motifs pointing to the silent FGFs but loss of the single motif that points to the active enhancers again causes encephalocele! You really cant touch some CTCFs.

142

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

Because we did another rescue. If we extend the 24kb boundary deletion to 145kb to include those putative enhancers, mouse embryos no longer develop encephalocele and the FGF expression is no longer upregulated.

131

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

With help from @andersshansen.bsky.social we used high-resolution capture micro-c to show that the FGFs interact much more with the brain enhancers of Ano1 when we lose that boundary. How do we know that those elements are really the enhancers?

141

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

The amazing Matt Anderson from @marklewandoski.bsky.social lab then helped us show that without the boundary, the FGF genes are expressed in the brain and recapitulate the pattern of Ano1, a gene found in the upstream domain. All this FGF signaling cant be good for brain development.

131

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

And it did! Deleting the 24kb 5' border caused encephalocele and cleft face in 100% of HETEROZYGOUS embryos! How do we know it's because of CTCF? Because we replaced the 24kb by just 600bp of CTCF motifs and that rescued the phenotype! To this day I am shocked this worked!🤯

291

Pedro Rocha @pedroprocha.bsky.social · 26/02/2025

3 FGF genes reside in 1 TAD separated by CTCF clusters. These FGFs have different expression patterns and functions. We thought that deleting these boundaries could really mess up expression of the FGF genes and developmental roles.

131

Pedro Rocha @pedroprocha.bsky.social · 06/11/2023

Using Micro-C in fly cell lines, we show that some (not all) Polycomb proteins still bind Polycomb Response Elements, and engage in strong focal contacts despite K27me3 loss and gain of K27 acetylation. Check data at: resgen.io/pedrorocha/K...

100