Lessons in genome engineering: opportunities, tools and pitfalls

CRISPR/Cas technology allows the creation of double strand breaks and hence loss of function mutations at any location in the genome. This technology is now routine for many organisms and cell lines. Here we describe how CRISPR/Cas can be combined with other DNA manipulation techniques (e.g. homology-based repair, site-specific integration and Cre or FLP-mediated recombination) to create sophisticated tools to measure and manipulate gene activity. In one class of applications, a single site-specific insertion generates a transcriptional reporter, a loss-of function allele, and a tagged allele. In a second class of modifications, essential sequences are deleted and replaced with an integrase site, which serves as a platform for the creation of custom reporters, transcriptional drivers, conditional alleles and regulatory mutations. We describe how these tools and protocols can be implemented easily and efficiently. Importantly, we also highlight unanticipated failures, which serve as cautionary tales, and suggest mitigating measures. Our tools are designed for use in Drosophila but the lessons we draw are likely to be widely relevant.\n\nAUTHOR SUMMARYThe genome contains all the information that an organism needs to develop and function throughout its life. One of the goal of genetics is to decipher the role of all the genes (typically several thousands for an animal) present in the genome. One approach is to delete each gene and assay the consequences. Deletion of individual genes is now readily achieved with a technique called CRISPR/Cas9. However, simple genetic deletion provides limited information. Here we describe strains and DNA vectors that streamline the generation of more sophisticated genetic tools. We describe general means of creating alleles (genetic variants) that enable gene activity to be measured and experimentally modulated in space and time. Although the tools we describe are universally applicable, each gene requires special consideration. Based on our experience of successes and failures, we suggest measures to maximise the chances that engineered alleles serve their intended purpose. Although our methods are designed for use in Drosophila, they could be adapted to any organism that is amenable to CRISPR/Cas9 genome modification.