SP6/T7 gRNA database
We attempted to design gRNAs for all Ensembl zv9 genes using a publically available tool1. We then developed an algorithm to sort these gRNAs based on predicted off-targets, sequence data analysis2, and position within the gene. The algorithm first characterizes gRNAs into 3 groups based on the number of predicted off-targets (0, 1, 2+). It then scores using the following formula:
60*[proportion GC content] + 10*[proportion transcripts targeted] – 30*[relative position in gene] + 2[if position 20 is G] – 3[if position 20 is A]
Design restraints included designing only to exons and the gRNA must start with a GG (compatible with T7 promotion) or GA (compatible with SP6 promotion). We report our 4 best gRNAs. We successfully designed 4 or more gRNAs for 87.6% of genes, 3 or more for 91.3%, 2 or more for 94.6%, and designed at least one gRNA for 97.3% of genes. There are some errors in designing gRNAs when the genes overlap, so our database is not definitive.
1. Heigwer, F. , Kerr, G. & Boutros, M. E-CRISP: fast CRISPR target site identification. Nat. Methods 11, 122-123 (2014).
2. Gagnon JA, Valen E, Thyme SB, Huang P, Akhmetova L, et al. (2014) Efficient Mutagenesis by Cas9 Protein-Mediated Oligonucleotide Insertion and Large-Scale Assessment of Single-Guide RNAs. PLoS ONE 9(5)