Very interesting! However, I think the most relevant comparison is not to something like a base or prime editor but to PASSIGE (covered briefly in a footnote here, and it does rely on prime editing hence DNA repair, so okay I guess it's different) or more directly to CASTs (CRISPR-associated transposases) which also make scarless DSB-free gene-size insertions (https://www.science.org/doi/10.1126/science.adt5199). It seems like bridge editing maybe more useful for making mouse models, but it seems like the potential therapeutic advantages over CASTs are pretty minor. Am I missing something?
For mouse models it makes little difference for now, at least not at current efficiencies. This is also somewhat conditional on our ability to make a stable transposase protein, hasn't always been easy in this field. You can get arbitrarily large deletions relatively efficiently with regular Cas9 (~10%) and then just breed mice with the correct genotype.
Once installed, Cre/loxP is typically near 100% efficient and used for more complex studies - I can't think of anyone who would use it these days just to make a deletion mouse line. With Cre/loxP you will typically induce a specific rearrangement in specific cell types or at specific timepoints - can't do that with bridge editors until they get really very efficient. BrEs might be a great tool in therapeutic contexts though, where efficiency is (relatively) less important than safety.
Very interesting! However, I think the most relevant comparison is not to something like a base or prime editor but to PASSIGE (covered briefly in a footnote here, and it does rely on prime editing hence DNA repair, so okay I guess it's different) or more directly to CASTs (CRISPR-associated transposases) which also make scarless DSB-free gene-size insertions (https://www.science.org/doi/10.1126/science.adt5199). It seems like bridge editing maybe more useful for making mouse models, but it seems like the potential therapeutic advantages over CASTs are pretty minor. Am I missing something?
For mouse models it makes little difference for now, at least not at current efficiencies. This is also somewhat conditional on our ability to make a stable transposase protein, hasn't always been easy in this field. You can get arbitrarily large deletions relatively efficiently with regular Cas9 (~10%) and then just breed mice with the correct genotype.
Once installed, Cre/loxP is typically near 100% efficient and used for more complex studies - I can't think of anyone who would use it these days just to make a deletion mouse line. With Cre/loxP you will typically induce a specific rearrangement in specific cell types or at specific timepoints - can't do that with bridge editors until they get really very efficient. BrEs might be a great tool in therapeutic contexts though, where efficiency is (relatively) less important than safety.
Do scars inhibit genetic function?