Thanks AP.

Thanks!

Yep. I've been there.

The second is far more silly. When I finished figure 1, there was an awkward gap between the graphs. So I found something to fill it. But anyway. I am sure they will not be in the final version of the paper... they are very likely the first to the chop when space reatrictions and revisions kick in.

Fair point. I sometimes wonder if sometimes we go too far. The reasons for the pics are two: first, most people will only open the paper. Scan the figures and move on. The more visual info I give them, the better they may remember the paper in the future and come back for a proper read.

Please let us know if you have any comments on our manuscript or in the technique. We are happy to share the things we have produced and/or to collaborate and help with your protein of interest.
#Share #OpenScience

🧵6/6 — The story behind the manuscript
This work has the crucial contribution of many talented masters students, who helped optimizing various aspects of the system. At the end, An-Sofie, my lab manager and Katja, a postdoc pushed to story until the end.
#CreditWereCreditIsDue

🧵5/6 — The Big Picture
EndoNB works on endogenous proteins. No rare antibodies. No overexpression.
It’s scalable and ready to explore the 1000+ surface proteins we’ve barely studied.
We’re done understanding vesicular trafficking by looking at a handful of model surface proteins.

🧵4/6 — Proof of Concept
With EndoNB, we:
✅ Tracked ligand & antibody effects (Integrin β1, TFR, AXL)
✅ Directly compared receptor uptake (integrin β1 vs β5)
✅ Revealed TMEM123’s endocytic behavior
With easy, clean and reproducible results.

🧵3/6 — The Solution
We CRISPR-tag the protein of interest with a tiny Alfa-tag.
Then use a nanobody-SNAPtag probe with a self-destruct button (okay, a 3C protease site)
Mix them, let internalize, chill at 4°C and use 3C to cut surface signal.
What’s left? Only internalized cargo.

🧵2/6 — The Problem
Cell surfaces are packed with proteins constantly moving in and out.
Studying them? Really hard.
Current tools rely on good antibodies (which are rare) and inefficient signal removal protocols.
So, we built something better.
#CellBiology

Nice! Well-deserved. Congratulations, mate.

Hi Rhys. Fyi: I could finally solve the issue. The guys on our HPC cluster found a suitable combination of packages.
Thanks!

Every time I solve one incompatibility, another one appears. I solve it, another appears... and the cycle goes on until the whole thing breaks. I need to use a container to install it ( I cannot use conda), which complicates things. All I need is the set of packages that talk to each other.

I am having many errors. Since the release of the git repo, python, numpy, scipy, tensorflow, biopython, pyrosetta, Jax, cuda, jaxlib, haiku, dm-tree (and others) have newer versions that are either not intercompatible or not compatible with the original scripts.

YES! PIs on the bench. That's how it should be.