This paper should clear up a lot more than anyone on here truly can about the CRISPR system.
CRISPR/Cas9 is very unique, even compared to other proteins purified from bacteria like Taq Poly.
The reason is due to how complex it is actually induce this protein. Essentially you are transfecting a plasmid with the Cas9 protein and its guide RNA (gRNA) on the vector. Transfecting itself is not easy, not to mention you are essentially adding a mechanism to a cell trying to knockout genes and you've got a handful. In fact doing that for one cell line, and noting the effects of what you knocked out, would be enough to produce a paper in a pretty impactful journal. (Image from: https://www.systembio.com/)
Also some recent advancements in CRISPR include:
Basically these researchers were trying to deduce a method allowing for more control of the CRISPR technique, since CRISPR has had issues with off-target effects. While it isn't a change in how CRISPR works, or even where, the change is how much you can control the reaction. Simply put they devised a way that allows you to turn CRISPR on/off using light allowing more precise control of its activity. Surely it will be helpful to reduce off target effects.
"This was accomplished by fusing the light-inducible heterodimerizing proteins CRY2 and CIB1 to a transactivation domain and the catalytically inactive dCas9, respectively." (Lauren R Polstein & Charles A Gersbach)
This one is a bit more interesting in that it is a direct change to the CRISPR system itself. Cas9 has been the standard cutting protein used for the CRISPR system although it has had controversy due to its somewhat imprecise cutting. Cpf1 has been shown to be more precise (so their data says) and it is now being implemented by some to see if this is truly the case. Combined with the above "light-induced" system this brings CRISPR down to safer levels when talking about off-target effects.
Here are some good online seminars to get you started:
- CRISPR/Cas9 from start to finish
- Improve CRISPR-Cas9 experiments with rationally designed guide RNAs
These seminars I can't really comment directly on because they explain everything so well on their own. Anything I say really will only detract from what they are teaching since I am by no means an expert with CRISPR but simply someone who is looking into possibly using it for my lab. Right now there is a lot of guide material on how to properly format your CRISPR experiments but my personal opinion is to give the technique maybe a year or two more when it is more standard for research.
That is unless you want to directly research the method and improve it in your own way.
Let me know if this answers your questions sufficiently.
Sorry that I cannot do much more in terms of explaining the links. The studies are not ones I can easily access unless I am at work since I usually access them through the university.
As far as the repair mechanism goes (your #3 question) I can't really answer that for sure but I have recently received a pamphlet that describes how to ensure the cells uses HJ rather than NHEJ.