Our method involves stable, random integration of Cas9 followed by scarless excision. It may be desirable in future iterations of this protocol to target a piggyBac-exciseable Cas9 transgene to a safe harbour locus, such as AAVS. This would eliminate problems with transgene copy number and reduce the chance of deleterious integration sites. Although our data suggest that piggyBac excision is robust and that the frequency of leaving a scar at the excision sites was below our detection limits (0/75 excision sites evaluated; Suppl. Data), a targeted integration strategy would nevertheless reduce the risk and facilitation validation of scar-free excision.
