off-target score: same scale as guide sequence score?

[Sergio PV]

Dear community,

I have a basic question that I have not been able to find the answer for.

The interpretation of scores for guide sequence goes easy (optimal is 100%, decreasing by their offtarget hits)

However, for the off target score, it is not clear what is large and what is small and the help page indicates:

"Guides having an aggregate score of greater than 50% are colored green and should be considered candidate targeting sequences if no high-scoring offtargets fall in marked gene regions"

My questions are:

1/ Is the off-target score in the same scale (maximun score = 100) as the guide sequence score? 

2/ what is the score of a high-scoring off-target?

Many  thanks

[Sergio PV]

By the way I am using:

http://crispr.mit.edu/about

I got here directly from the forum page of that website, but it appears to be general for CRISPR/Cas systems

[Victor Pontis]

Hi Sergio, 

1/ Is the off-target score in the same scale (maximun score = 100) as the guide sequence score? 

The on and off target scores are both on the 0-100 scale where a higher number indicates better performance. 

2/ what is the score of a high-scoring off-target?

There is not consensus on what is a "high" off-target score. Higher is better in general but there is not a magic threshold at this point. CRISPR guide scoring is still in its early stages, so the scores are a good heuristic but should only be taken as a starting point for empirical testing. 

I'm an engineer at Benchling where we are making a research platform that includes a CRISPR tool. 

Our CRISPR tool supports over 50 genomes, return off-target scores in seconds, and support the new CRISPR Scoring from Microsoft Research. 

You can check it out here: benchling.com/crispr

Let me know if you have any questions. :)

Victor Pontis

vic...@benchling.com

[Sergio PV]

Hello Victor,

thanks for the reply. If I understand correctly:

If, for example, both the off-target score and a guide score are 80, that would mean that that sgRNA has in theory equal chances to act on and off target?

Thanks

Sergio

[Victor Pontis]

Hi Sergio, 

If, for example, both the off-target score and a guide score are 80, that would mean that that sgRNA has in theory equal chances to act on and off target?

No, that is not what I was trying to explain. 

The on-target score is a heuristic for how likely the sgRNA with a specific series of bases is to successfully bind and cut an exact matching target site. The original on-target score paper was looking at how likely the sgRNA was to cause the null allele. 

I like to think of the off-target score as the likelihood that the sgRNA is going to mess up other parts of the cell's genome. So a higher off-target score indicates that the specific sgRNA is less likely to cut other parts of the genome. 

These two heuristics are not at all designed to be compared against each other. They are two different dimensions that help you measure and think about sgRNA activity. 

Let me know if that's helpful. :)

Victor Pontis

vic...@benchling.com