Unfortunately, no. This is limited by the intrinsic fidelity of the Cas9 enzyme. There have been efforts, however, to engineer CRISPR reagents that cut DNA only when two Cas9 molecules bind next to each other on the DNA (http://www.nature.com/nbt/journal/v32/n6/abs/nbt.2908.html), which has helped decrease off target cutting.They say that the accuracy is low and even modify unwanted sites, does that mean a longer guide RNA in cas9 makes for a better search algorithm?
Gene editing by CRISPR makes use of the cell's normal DNA repair mechanisms to insert the new strand of DNA at the site of DNA cleavage. The cell will recognize double-stranded breaks in DNA and, through a process called homology-directed repair, it can recognize another strand of DNA with similar sequence to the sequences surrounding the break (usually this is the second copy of the DNA on the other chromosome, but in the case of gene editing, it is a foreign DNA molecule injected into the cell by researchers). The DNA repair machinery can then copy the sequence from that other DNA strand onto the broken DNA strand in order to repair the double stranded break.So I get how they are able to cut the DNA at the desired location, but how do they go about inserting a new strand of DNA at the gap? Do they also cut off a strand of DNA at the process?