CRISPRscan

CRISPRscan is a novel algorithm to predict sgRNA efficiency.

Based on a large scale analysis of sgRNA mutagenesis activity in zebrafish, we established rules to predict sgRNA activity in vivo and build the CRISPRscan model integrating these rules. We independently validated with success our predictions using sgRNAs different from the large scale analysis.

Off-targets

CRISPRscan searches for potential genomic off-targets with the following rules:

1. With the method published by Cong et al. Science, 2013, potential off-targets must match perfectly in their seed (12 nt 3' of the PAM sequence) and a maximum of 2 mismatches in the rest of the sgRNA. These rules are more stringent than the method 2. and therefore less off-targets are found.

2. According to Hsu et al. Nat Biotechnol., 2013, potential off-targets can have a maximum of 2 mismatches with the sgRNA.

sgRNA generation

T7 promoterTarget
sequence
Tail annealing sequence
sgRNA primer5’ TAATACGACTCACTATAGG(N=18-17)GTTTTAGAGCTAGAA

Tail primer: 5’ AAAAGCACCGACTCGGTGCCACTTTTTCAAGTTGATAACGGACTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC

Gene annotation

Gene annotation from Ensembl 80 was used. For predictions on zebrafish Zv9, we used Ensembl 78.

Plasmid: Cas9 with nanos 3’UTR

Plasmid for targeting Cas9 expression into the germ line is available at Addgene.

Citing

CRISPRscan: Designing highly efficient sgRNAs for CRISPR/Cas9 targeting in vivo

Miguel A. Moreno-Mateos*, Charles E. Vejnar*, Jean-Denis Beaudoin, Juan P. Fernandez, Emily K. Mis, Mustafa K. Khokha and Antonio J. Giraldez
*Equal contribution

Nature Methods 2015
doi:10.1038/nmeth.3543