On this page, we invite comments and questions about the Talbot and Amacher 2014 CRISPR protocol. We'll also post updates to the protocol as it develops.
Monday, October 6, 2014
cas9 synthesis
In this thread, we invite comments and questions about cas9 synthesis.
I was recently asked a couple of questions about CRISPR target design that I thought would be of interest to the community, so I have posted these questions and my answers below:
1) How many CRISPRs should be initially designed for each target gene?
Around two thirds of gRNAs induce acceptable levels of mutagenesis, so in the past I tested one site before moving on to a new CRISPR target. However, the shortened guide oligo templates are cheap, so I’m now picking up two from the getgo. When I pick up “backup” CRISPR targets, I prefer to chose sites that don’t overlap one another.
2) Is it better to design a 21 bp CRISPR which begins with GG, or a 20 bp CRISPR which has a mismatch to cause GG?
This question is a little trickier. It seems likely that 21 bp or 1 bp mismatch CRISPR would both work. I’ve successfully used CRISPR with 1 bp mismatch (see also Woong…Yeh, 2013), but have never tried CRISPR targets >21 bp. (Ran…Zhang 2013) found that CRISPR longer than 20 bp get somehow cleaved down to 20 bp in human cells, and that longer CRISPR have the same activity as shorter CRISPR (small N, strange result). Conversely, (Fu…Joung 2014) indicated that shorter CRISPRs outperform longer ones, but only looked at CRISPR < 20 bp. On the balance, I’d say go with what we’ve successfully tested: 1 bp mismatch. Alex Scheir’s lab had good luck with the SP6 promoter, which doesn't require 5' GG, and we’ll be testing out this promoter in our lab over the next two weeks.
I was recently asked a couple of questions about CRISPR target design that I thought would be of interest to the community, so I have posted these questions and my answers below:
ReplyDelete1) How many CRISPRs should be initially designed for each target gene?
Around two thirds of gRNAs induce acceptable levels of mutagenesis, so in the past I tested one site before moving on to a new CRISPR target. However, the shortened guide oligo templates are cheap, so I’m now picking up two from the getgo. When I pick up “backup” CRISPR targets, I prefer to chose sites that don’t overlap one another.
2) Is it better to design a 21 bp CRISPR which begins with GG, or a 20 bp CRISPR which has a mismatch to cause GG?
This question is a little trickier. It seems likely that 21 bp or 1 bp mismatch CRISPR would both work. I’ve successfully used CRISPR with 1 bp mismatch (see also Woong…Yeh, 2013), but have never tried CRISPR targets >21 bp. (Ran…Zhang 2013) found that CRISPR longer than 20 bp get somehow cleaved down to 20 bp in human cells, and that longer CRISPR have the same activity as shorter CRISPR (small N, strange result). Conversely, (Fu…Joung 2014) indicated that shorter CRISPRs outperform longer ones, but only looked at CRISPR < 20 bp. On the balance, I’d say go with what we’ve successfully tested: 1 bp mismatch. Alex Scheir’s lab had good luck with the SP6 promoter, which doesn't require 5' GG, and we’ll be testing out this promoter in our lab over the next two weeks.