The nice thing about CRISPR is that you can use pretty long sequences for matching, so in theory you can just pick a sequence that doesn't occur anywhere else a

The nice thing about CRISPR is that you can use pretty long sequences for matching, so in theory you can just pick a sequence that doesn't occur anywhere else and cut there. (Before CRISPR, you would use restriction enzymes with a short, fixed recognition sequence, typically 4-6 base pairs [1]. In that case it was a bit tricky to find the right enzyme that cuts the DNA in the right place, and not in an unwanted place. And you couldn't just cut the DNA in any place you want.) [1]: https://en.wikipedia.org/wiki/List_of_restriction_enzyme_cutting_sites:_A https://en.wikipedia.org/wiki/List_of_restriction_enzyme_cut...