What if we could locate, sequence, and identify every protein in a cell? This would unlock a transformative way to study biology and disease.

That question shaped my PhD, and I’m excited to share our new preprint on the Principles of In Situ Protein Sequencing, now live on bioRxiv.

Thanks to my mentor @eboyden3.bsky.social , and collaborators Adam, Yasu, Liyam and Anu!

We are excited to release this method to the community, as we work on the first in vivo screening campaign using it!

The method uses the new safe harbor pIGLET lines made by the Mosimann lab, and we include in the discussion some extra ideas about potential adaptations to worms, flies, mice and pigs as well.

For that, we need methods for pooled library transgenesis in vivo. We invented a way to do that, in zebrafish (my favorite model for biosensor applications). Our trick is based on exploiting delayed site-specific integration to get libraries of >1,500 variants in each animal.

To make useful tools that work well in vivo, we need to screen them in vivo. Engineering a whole animal to test each variant is slow and unscalable, but if we can express libraries of proteins in vivo, each animal can turn into hundreds of parallel experiments.

When you screen genes in a dish you learn about how they function in a dish. When you optimize your protein in fibroblasts, you might end up with a biosensor, enzyme, or tool that works great in fibroblasts, but doesn’t work in the living brain.

New preprint out!
www.biorxiv.org/content/10.6...
One of the most salient lessons I learned in all my adventures with screening and protein engineering is the importance of context. The context in which you test a gene or protein determines what function it has.