Re: tweaking the sgRNA, we did consider making a parallel sgRNA deletion/point mutant library, but combining this with the protein library would yield a combinatorial explosion in the overall library size and thus be difficult to screen- may be worth doing for specific protein variants, though!

Agreed! We had some expectations based on the SpCas9 MISER results that SaCas9 would show a high degree of modularity re: tolerating domain deletions, but it was kind of surprising just *how* deletion-tolerant it is - for example, the big "HNH" deletion region spills over into RuvC-II and III. (...)

Wow... this is my first experience on Bluesky, and I'm just now learning that I can't go back and revise my posts 🤦‍♂️ apologies for the mess!

Nonetheless - hope this was helpful 😄 Thanks for reading our paper!

3 (cont'd): (...) SpRYCas9 (now available from NEB!), a target plasmid, and one of any common bacterial cloning strains. Plasmid recombineering is (modestly) more complex and requiring of special reagents/strains, and moreover in our hands has proven to be a bit finicky and problematic.

3. Two of the main "selling points" of the new SpRY-based strategy (at least in my opinion) are speed and simplicity: the library cloning protocol can be carried out in as few as three days, and there are very few "moving parts" - all you need is the sgRNA library, (...)

2 (cont'd): We initially tried using a blunt-cutting Type IIS enzyme, MlyI, in an approach that was in some ways more similar to the original MISER strategy, but the enzyme turned out to be uncooperative (it's a long story 😅)

2 (cont'd): (...) site of each "deletion" (thus technically making them replacements). Single-codon deletions can sometimes be very interesting/useful (a great example can be found in the linked article: pubmed.ncbi.nlm.nih.gov/24856363/), and thus it would be ideal if they were sampled. (...)

2. One limitation of the MISER approach as it was originally conceived was an inability to generate very small (e.g. one or two codon) deletions; the fixed-sequence overhangs generated by digesting the inserted Type I restriction sites would necessarily result in a two-codon "scar" at the (...)

Hey there! Great question 😃 the answer is severalfold:

1. Part of the motivation was simple novelty - we thought it would be neat to apply SpRYCas9 as a protein engineering tool, especially in light of its PAMless nature making it almost uniquely well-suited for this particular task

(1/3)