How is CRISPR KO efficiency so high?

I've been afraid to ask this since forever, but I feel like I don't understand CRISPR at all.

So we cause indels via Cas9...

And 2/3 of the time, the resulting indels cause a frameshift and we get no functional protein from that gene. Fine.

But 1/3 of the time the resulting indels wouldn't cause a frameshift and likely wouldn't do anything. We could easily have a functional protein from this I would think.

And let's take a primary cell. 2 copies of geneX.

So (2/3)*(2/3)= 0.44 -> 44% chance of both copies getting a frameshift. And therefore 100-44=56% chance of at least one copy not getting a frameshift (and therefore probably encoding a functional protein). So less than 50% odds of a successful KO?

And for cancer lines that have lots of copies? Would be super low odds all copies get a frameshift. ~80% odds of no KO if 4 copies of a gene (by my logic).

How do I see data published that looks like 90+% reduction in protein (excluding monoclonal lines)?

How does this work? What am I missing?


Login or Signup to leave a comment
Brown Valkyriealmost 4 years ago

I think what you are missing, is that most of publications are on monoclonal levels, they screen first on polyclonal cell lines and then go clonal to get that less than 50%! This is applicable in the lab!

Violet Gorgonalmost 4 years ago

Respected Young scientists,

                                       I hope you are doing well. I need ' **'** **CRISPR-Cas: A Laboratory Manual** of Cold Spring Harbor Laboratory''. please if anyone here ve that manual in pdf send me. I will be very thankful to him/her.
White Elfalmost 4 years ago

I have asked the same question to Feng Zhang a few years ago at the first Genome engineering keystone conference. He told me 2/3 chance of reading frame shift is just in theory. In reality, the chance is often higher than 2/3.

He did not explain why. The outcome of indel could be predicted based on the sequence on the targeting site. Please check the following papers

https://www.nature.com/articles/nmeth.3015

https://www.nature.com/articles/s41586-018-0686-x

Based on my own experience, it is not uncommon to achieve nearly 100% knockdown at the protein level in CRISPR edited cell pools.

Find your community. Ask questions. Science is better when we troubleshoot together.
Find your community. Ask questions. Science is better when we troubleshoot together.

Have a question?

Contact support@scifind.net or check out our support page.