This is a good advancement on CRISPR, it gets us one step closer to having treatments for disease that is genetically linked. It also gets us one step closer to having a 'genome debugger' which would be a tool for analyzing what every gene does in isolation so that the effect of changes can be more accurately modeled (and predicted).
The scary part of that for me is that we get closer to having genomic 'zero days' where an underground lab has figured out how to create a pathogen that has a specific (and negative) targeted effect on a particular gene or gene group. Such a tool in the hands of someone who is looking to do some "ethnic cleansing" is a concern for me.
I have to wonder about this fear. Essentially we are claiming that a highly motivated and technically proficient person or group of persons will seek to eliminate an ethnic group. There are many prerequisites to carry this out, including obtaining comprehensive genomic profiles of the group, as well as genomic profiles of closely related groups which you don't want to target, obtaining tissue samples from the group, designing a vector... I'll stop there, but to actually create an ethnicity specific disease is very difficult, and probably limited to large groups of scientists with significant funding. This assumes you want to be more specific than say killing all Asians, or Northern Europeans, which would be easier.
I think drones with facial recognition would be dramatically easier and more feasible... I mean, someone could actually do this next week.
We already have nukes, and you can still do ethnic cleansing most cheaply with conventional weapons and impotent nation states, no need for Super CRISPR.
Nukes have signatures. This does not. Just a little aerosol in your general vicinity would likely do the job and you'd never be the wiser. Besides that it is orders of magnitude easier and cheaper to do this than build a nuke once you can mail order for genetic sequences delivered to your door.
The Problems with Bioweapons (and you need bio here, either a bacteria or virus that delivers the crispr genedrive, a single spray won't be enough), is that they evolve.
Your super virus that you developed to only kill people living in southern florida? Yeah, that now developed to also kill people in new york. And now it's the whole world.
It would work like this: you'd design a bioweapon to hit the Jewish people exclusively (they're the punchball of history, forgive me using them again as an example but I think it is a likely occurrence if this sort of thing is to happen). Then it will may evolve to hit others.
Sorry but I don’t understand how your point differs? I think the gist was that it’s weaponized for a certain group and then evolves to target a much wider group. Isn’t that the same thing as you mentioned?
I purposely didn't select any specific ethnicity or other genetic factor to avoid tapping into stereotypes or causing offence (except to people living in Florida, sorry but not sorry).
But if one wanted to, bioengineering ethnic cleansing (or whatever section of genome you're targeting) can be done with a couple thousand dollars worth of supplies in a garage, no nuclear or conventional arsenal required. The point of importance is the barrier to entry is so much lower.
I hear you, a nuke is the ultimate tool for implementing widespread death. But I don't worry about getting nuked because to build something requires a lot of money, expertise, and parts that are closely monitored. Then you have to figure out how to transport and deliver the damn thing. Not a huge problem if you're a nation-state but bigger for individual actors and their followers.
A CRISPR based biological attack however is something you can pull off with lab equipment you can buy anywhere in the world. For an amount of money that can be acquired much more easily. The risk to the perpetrators has always been that they would screw up and die of their own plague.
The ability to create more precisely targeted biological attacks means that a person who is working on such an attack and doesn't have the necessary genetic makeup for what ever virus vector is being used to transmit it, can do so with much less fear of their own demise. While creating something that can be transported in disguise without being easily detected.
Perhaps an analogy would work.
My kids never had to worry about European teenagers bullying them, but today's kids have contacts all over the world on social networks some of whom can be quite hurtful. There is bullying that happens because the perpetrator "knows" that the target is so far away they are unlikely to be able to cause repercussions. Thus the 'risk' of bullying is lower, and in my estimation happens more frequently.
>I hear you, a nuke is the ultimate tool for implementing widespread death.
That's not my point. I agree that the barriers to entry for biotech are lower.
My point is that whereas the only 'defence' against nuclear weapons is deterrence, i.e more nukes, it is possible to defend against biotech weapons without starting an arms race. This would consist of ongoing, costly research of antidotes by government and university labs (with a high barrier to entry).
For the better or worse, this is the standard language for modern academic writing. Do not blame the authors if this social convention is not to your liking.
I understand that CRISPR can change the genome in a specific chromosome, but how do you change the genome in all cells of a body? Or you don't need to change all cells in order to cure genetic disease?
It's difficult. The standard way in frogs is to just keep injecting them until the altered cells outnumber the regular ones. You wouldn't have to change them all though, often genetic diseases are confined to a particular tissue.
There are also theories around using a live virus to carry the genetic material, theoretically a virus like that could leave behind the desired mutation as it replicates through the body's cells. Promising mechanism, though I don't know of anyone who's pulled it off.
Yes, you can't change the genome in all the cells of a body. Whether this is necessary or not depends on the disease. For example, for certain conditions, it would be enough to edit some bone marrow stem cells, then put them back into the patient, perhaps at the same time as getting rid of most of their existing bone marrow with chemotherapy or radiotherapy. Another example is editing a specific organ only, such as the retina or the liver.
It does not work on all cells at once. For a cell's genes to be edited, the cell must receive its own copy of CRISPR-Cas9 or whatever other gene editing system.
Delivery of gene editing systems to the right cells is a major challenge. At present, most gene editing therapies involve:
* extracting blood cells from a patient, editing them ex vivo, optionally selecting for successfully edited cells, and readministering to patients
* delivering gene therapy with viral vectors and treating diseases where you don't need to edit 100% of relevant cells for the treatment to be effective, and targeting cells that are easier for the viral vectors to infect (typically motor neurons, the eye, blood cells or the liver)
It isn't really possible to edit every or most cells in the body with current technology. It is possible to edit gametes but it is not currently ethical to do so as we don't understand the risks and the potential for abuse is huge
The key advance in this work is showing a way to edit without any external inputs to the cell (in the form of DNA templates). This suggests the possibility of in cell feedback systems. I don't get the best vibe from all the rush to market and focus on correcting generic disorders. Is that really what gene editing is for?
I'm not an expert but I believe that one of the key advances here is that this technology can introduce new DNA via more precise mechanisms, breaking just one DNA strand at a time.
Current CRISPR tech involves "double stranded breaks": cutting both strands of DNA. The body has a few ways of dealing with and repairing these breaks, one of which scientists can use to insert new DNA of their choosing [0]. However in other cases the cell uses a different repair pathway that can introduce mutations (including insertions or deletions of DNA). These mutations can be harmful especially if for example they mutate a tumor suppressor gene
The new approach here apparently has less of a risk of introducing these mutations bc it does not cause double strand breaks, so does not trigger potentially erroneous repairs that cause harmful mutations
But I'm not an expert so if this is incorrect or if I'm missing something please chime in
I don't recall a clear causal link being drawn beyond some favoring off cells with inhibitions in apoptotic pathways being more likely to successfully undergo crispr knockout- certainly a risk but nothing like what you're suggesting with that strong statement. Can you elaborate on your point? I'd love to learn if there was any more research into that, it's an important question for sure.
This discusses a few papers showing that CRISRP-Cas9 can cause "off target" mutations in a tumor suppressor gene called p53. If these mutations render p53 non functional it is possible that these cells can become cancerous
That is one of the specific risks identified although there are in theory other risks
I dont think the cas9 is causing the off target mutations according to those papers, it's more potentially selecting for cells with already dysfunctional p53 in those few cell lines by inducing a p53 mediated dna damage response. It's in cells that arent going to go near humans and are pretty different in form and function than those used in trials, and while it's a risk to consider (not trying to downplay potential risks here- all new medicine and biology must be thoroughly evaluated for safety and risk), it's nowhere near a show-stopping demonstration that CAS9 mediated editing causes cancer and hasn't stopped clinical trials from moving forward
Thanks for the correction, I was thinking of potential for off target edits that lead to cancer but you are right, that's the wrong article for discussion of off target
Several people elsewhere in the thread are discussing the possibility of bioweapons engineered to target specific ethnicities. Wouldn't it be just as feasible to create something similar that would specifically attack cells with a mutated p53?
There may be two different interpretations of "risks" in use here. If your mechanism both causes swathes of unexpected damage and also (because of, perhaps) favors cells with defective p53, certainly there is a reasonable until-it-improves risk of causing cancer.
I think it's worth pointing out that this headline pops up every time there's a minor improvement to CRISPR/casX regardless of whether it's a substantial improvement or not.
Reposts are fine if a story hasn't had significant attention yet: https://news.ycombinator.com/newsfaq.html. On HN we try to give good stories multiple cracks at the bat.
Eventually we want to implement a karma-sharing system so earlier submissions benefit from reposts that catch on. In the meantime, it's a bit of a lottery—but if you submit enough good stories it evens out over time.
Why do people even share this sort of news anymore. I don't think I've ever read an article about novel medical science that is supposed to treat [bad disease] that eventuated in any actual treatment or cure.
Because it's interesting science. Isn't that a good enough reason? It may work out to something amazing or it may not, but thinking about it is cool either way.
Are articles about black holes pointless because in our lifetimes it's extremely unlikely that any human being will ever observe a black hole directly or do any engineering involving black holes?
You didn't answer my question. Yes, of course clinical trials take forever. Why share this article here or anywhere? There is no cure or treatment for anything mentioned in the article. There might be one, but if there is it's a decade away. How is that news?
It's like someone writing an article about a few people potentially being President of the USA a decade from now. Except nobody is actually named in the article. Would you share such an article?
Is there a lot of money invested in CRISPR? Is this all about building hype for CRISPR? Because I can't think of another motive.
It's also a lot like an announcement of any other tool of great importance: a new programming language, a new rocket engine, a new alloy, a new psychological principle - or any other discovery of import. It's profoundly interesting to watch, understand, and organize those discoveries (as news) - precisely so the more pragmatic downstream affects can be achieved.
Those who can put those various discoveries together are those that will be the ones that use those tools in conjunction in order to benefit everyone. But how are they to know about them if the discoveries themselves 'are not news'?
Even if it doesn't lead to a medical treatment, the impact of CRISPR on everyday biomedical research has already been enormous - it's probably the biggest advance in molecular biology lab techniques in the past decade. Getting it to work in humans is of course a huge project, and we can expect many failures along the way, but genetic modification is vastly easier than it was 10 years ago, when we barely had any idea how to make these kinds of changes. Every improvement to the technology makes it that much more likely that we'll see a human application in the near future.
You can read about Sofosbuvir, an actual _cure_ for Hepatitis C, and Zolgensma a novel treatment for spinal muscular atrophy. CRISPR is still very new, this field will be unrecognizable by the end of the next decade.
The tech has already been licensed to a new company backed by top biotech VCs ARCH, GV, F-Prime and Newpath