This is possibly the most reasonable piece I've seen linked on the topic. CRISPR/Cas9 techniques do not give us anything we couldn't do before--- after all ther

This is possibly the most reasonable piece I've seen linked on the topic. CRISPR/Cas9 techniques do not give us anything we couldn't do before--- after all there have been TALENs and zinc finger nucleases for quite a while. It just made the process of designing a new nuclease system much, much cheaper. However, no endonuclease driven system is going to be efficient at introducing mutations consistently. The basic problem is that these systems require the cooperation of the genomic repair systems of the cell. There are several ways which these repairs are mediated, and only one (homology directed repair) yields the desired outcome. That this happens at a low rate, and requires the introduction of DNA templates only diminishes the overall efficacy and multiplexity of the process. Even when flooding cells with preassembled Cas9/gRNA complexes that are immediately active, the overall rate of homozygous introduction of the target mutation is in a few %. Knowing this, I find it hard to believe that CRISPR will be how we ultimately engage in high efficiency genome modification. It depends too much on the biology of the target system. That said, the hype around it has initiated a great and timely public discussion about the use of these tools in medicine.