CRISPR StAR: facilitating high resolution genetic screening in complex in vivo models

Technology Description

CRISPR StAR is based on CRISPR switch (see Partnering Opportunity for 44-00004), and takes advantage of sgRNA expression vectors, in which the sgRNA gene is interrupted by a disruption cassette flanked by a pair of recombinase recognition sites, which upon recombination facilitate the formation of an active sgRNA. However, the CRISPR StAR vector also carries a second pair of recombinase recognition sites which upon recombination facilitates the formation of an inactive sgRNA that serves as an internal control. Both recombination events are mutually exclusive and strictly stochastic, and accordingly yield a specific ratio of active and inactive sgRNA. In negative selection screens, the innovative sgRNA constructs are activated only after their representation has recovered. True essential genes are determined not only by depletion of the corresponding sgRNA (which is per se difficult to distinguish from a lack of transduction or recovery), but rather by a stochastic drift in the ratio of active and inactive sgRNA.

CRISPR-StAR's effectiveness was evaluated e.g. in vivo using a mouse melanoma cell line, with genome-wide screens conducted across 143 animals. On average, there were 2.3 cells per sgRNA or 8.7 cells per gene. CRISPR-StAR successfully distinguished between essential and non-essential genes in sparse and heterogeneous datasets, outperforming conventional CRISPR screening. It demonstrated superior performance in discriminating between in vitro depleting genes and controls. Reproducibility of two independently cloned datasets, generated two years apart, improved significantly with an R coefficient of 0.54. Overall, CRISPR StAR effectively reduces noise from data sparseness and heterogeneity, enhancing genetic screening in vivo.