14 comments

[ 2.9 ms ] story [ 38.9 ms ] thread
This is somewhat old news in the field, but it's nice to see some mainstream coverage of science. I'm glad that they mentioned to off target effects. I think that editing with Crisprs is supposed to be somewhat noisy, so any uses outside of the lab (eg clinical) will probably be limited. But, I haven't worked with them, so who knows where this will be in 5 years.
> Crisprs is supposed to be somewhat noisy

Since you seem knowledgeable: Do you know how they compare to other kinds of methods for cutting/splicing new genes into DNA, such as Zinc fingers or homing endonucleases? Specificity (non-noisy) seems like it would be the most important feature of something like this. Although I guess we don't really know, I could imagine "noise" turning into other horrible diseases, loss of viability, etc.

Edit: And now I notice that they discuss this further down in the article. Nevermind.

will probably be really good for engineering organs (for clinical use) or organisms where you can afford to throw away mistakes.
The noise factor can be remedied with clever tweaking of the Cas9 protein, as mentioned in the article. This research is currently underway.

The key is, optimizing the Cas9 protein is much easier than optimizing an infinite number of designer proteins targeting individual sequences, which is what previous methods required.

Reminds me of: "The original virus rips, or "unglues" - as the name suggests - DNA strands. The strand attempts to repair itself, but its proteins are scrambled in the attempt." — http://ansible.wikia.com/wiki/Descolada
tl;dr: like restriction enzymes, but easily retargetable because they recognise sites using RNA.
It is interesting how profession changes perception of the language. I work as a software engineer and "modify DNA" sounds as a more natural description of the process to me. I am speaking in context of taking existing DNA and using a technique to identify specific locations in order to change them. In other words there is no "redo" if you get something wrong. In news world editing is much more natural and encompassing term.
I think "edit" is appropriate, since you aren't just randomly modifying the DNA, but modifying the DNA at a specific location in order to introduce a specific modification. Furthermore, in biology we usually use "modify" to mean more exotic changes, such as adding specific chemical groups to the DNA (e.g. methylation) that change how cells interpret the sequence.
Software engineers often use the word "update" for operations of that kind on databases, as part of the so called "crud" operations (Create Read Update Delete), which represent the fundamental operations one could apply. I agree with you that "edit" sounds good (and "cred" perhaps better than "crud"), though it sounds a bit more like an ongoing action rather than an atomic operation to me (personal bias I suppose).
"The development of the new tool is an example of the unanticipated benefits of basic research."
Worth noting that the initial discovery - identifying the common sequence motif - is a purely computational problem.
> edit DNA

Isn't there an emacs mode for this yet? I guess Stallman is really slacking these days...