<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div>On Dec 6, 2558 BE, at 3:58 PM, Rafal Smigrodzki <<a href="mailto:rafal.smigrodzki@gmail.com">rafal.smigrodzki@gmail.com</a>> wrote:</div><blockquote type="cite"><div><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Sun, Dec 6, 2015 at 6:36 PM, Dan TheBookMan <span dir="ltr"><<a href="mailto:danust2012@gmail.com" target="_blank">danust2012@gmail.com</a>></span> wrote:<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><br><div>The fear would be accidentally eliminating something essential to the biosphere and human existence, no? For instance, can a gene drive used to get rid of a malarial spreading mosquito be prevented from spreading to other mosquitos? (I'm also wondering about good old Darwinian evolution working around the mechanism in this particular example. But, of course, that might be easier to deal with once you have the tools and knowledge.)</div></div></blockquote><div><br></div><div>### You can design a reverse gene drive to render critters immune to a released version.</div><div><br></div><div>A gene drive generally is transferred only to offspring, so it should not find its way to other species, unless the species are capable of interbreeding and producing viable offspring. This may be occasionally of concern, for example if you try to use a gene drive to eliminate African bees and end up wiping out European bees.</div><div><br></div><div>Evolution would of course find ways of incapacitating gene drives but I think that the arms race between CRISPR designers and critters would be mostly won by designers. You should be able to quickly design a counter to conceivable evolved anti-CRISPR mechanisms. If a critter has a Dicer mutation that directs it towards the RNA guides used by CRISPR, you just disable Dicer. Etc.</div></div></div></div></div></blockquote><br><div>Much of this would depend on time sensitivity too. The designers or the team needs to think about this to avoid something happening so quickly that it can't effectively be undone or that the cost of undoing is extremely high. I'm not urging a precautionary approach here -- just not want that ignores all risks as if the technology is self-fixing because, well, we can out a rival gene drive in place. </div><div><br></div><div>You wouldn't want to do this willy nilly only to find that, say, there's unknown contacts or that the species in question has an unknown ecological relationship that's important. Of course, I don't believe this is insurmountable.</div><div><br></div><div>With the Darwinian stuff, I was thinking more of microbes. Use a gene alteration to make vector X immune to pathogen Y and mutant Y' might find a way around this. I wasn't thinking of vector X' evolving to spread Y. In general, pathogens evolve at a faster rate than their vectors, no?</div><div><br></div><div>That doesn't mean don't do it. Just be aware of the possibility and maybe try to figure out -- maybe via lab or field tests -- how likely this scenario is. (My guess -- and this is pure speculation by my feeble fevered brain -- malaria probably doesn't have a big risk of this, but maybe Ebola does.)</div><div><br></div><div><div style="line-height: normal;"><span style="line-height: 20px; background-color: rgba(255, 255, 255, 0);">Regards,</span></div><div style="line-height: normal;"><span style="line-height: 20px; background-color: rgba(255, 255, 255, 0);"><br></span></div><div><div style="line-height: normal;"><span style="line-height: 20px; background-color: rgba(255, 255, 255, 0);">Dan</span></div><div style="line-height: normal;"><span style="background-color: rgba(255, 255, 255, 0);"> Sample my Kindle books via:</span></div><div style="line-height: normal;"><a href="http://author.to/DanUst" style="background-color: rgba(255, 255, 255, 0);"><font color="#000000">http://author.to/DanUst</font></a></div></div></div></body></html>