I'm helping a grad student validate a CRISPR insertion with long-read sequencing and the target site is messed up with SVs, probable exogenous DNA and multiple insertions. This isn't my area, but a quick review of recent papers about this suggests most people don't look for this stuff.

When I first agreed to help out I was like. "oh yeah, this will be easy. flank-insert-flank. should be trivial. we should see it with mapped reads. at worst we do a quick assembly". right. never tell anyone that.