So you can target a mutated site with a zinc finger. Question: what do you do with the mutated site? Well, you... have to replace it. Now, there is a process by which cells, this is the other natural process, by which cells repair double-stranded breaks. Now, the cells in any organism are constantly being bombarded by ionising radiation, X-rays, things you don't... and the cell, at any one time, there are thousands of breaks in the DNA. The DNA has the most elaborate machinery for repairing this. You know DNA is its most precious content and the way it's done is that when there's a double-stranded cut there's a whole dance of proteins, it's really choreography, an ensemble of proteins, one protein which is well known Chek2 finds the double-stranded cut and switches off... it switches off any of the rest of the cell cycle. Other proteins come along and bind to the cut and there are different outcomes. You can actually do different kinds of repair processes. One pretty dominant repair process is carried by... what is called homologous combination... with the sister chromosome. What happens is that you... quite elaborate protein structure is built, remember I asked you for some reprints on the electron microscopy? A protein scaffold is built and the sister chromosome comes alongside the... place where the mutant is, where the double stranded cut is and there is an exchange by what is called homologous recombination which is a natural process. In genetics it's called gene conversion, it's a well know process in yeast genetics and lo and behold the... a piece of DNA, and from the Sangamo experiments, it's probably something like several hundred bases, are cut out, this is not known, and the... the DNA of the sister chromosome is copied and repaired so that's how you repair. It's... homology driven, or homology recombination driven repair and it's a natural process. Now people had been trying for a long time to do what they call gene targeting. This goes back to [Mario] Capecchi in 1989. And what he was trying to do, he was trying to make mutants and so what he did was introduce large quantities of mutant DNA into a cell and hope that by homologous combination the mutant would replace the DNA but the efficiency is about... one in 10 to the five. That's one in 100,000 cells get repaired. Now that's enough if you're going to do something in a mouse or something like that because you can then cultivate those cells and so on, in fact it's quite good, but generally it doesn't work. So a lady called Maria Jasin... J-A-S-I-N in France thought... demonstrated if you can make a double-stranded cut then the frequency of re-combination went up by a factor of 5,000. And she did this in a very clever experiment. She couldn't target the... what she did was to do a principle experiment she took a... a well-known endonuclease, it's called homo-endonuclease which a well-known for endonuclease which has an 18 base pair target, so she took a chromosome artificially and artificially inserted an 18... this 18-base pair target and then she used this homo-endo nuclease and then she did this... and showed that you can do a homologous recombination and it went very, very fast, 5,000 times. Not fast, efficient, it also went... goes quite fast. So the next thing to do would be to do it, it's fairly obvious from there on, is that you must target the gene, the natural gene. The way of targeting it, you can't build in the site, you must have a zinc finger which targets the natural mutated site, if you can replace the mutation or the other way round you could introduce a mutation in the natural sequence, anyway we're talking about gene correction or gene repair or gene editing. 'Gene editing' is what The Guardian or The Times called it when they wrote this up last... April.