• Nick
    13
    Hello world! I'm happy to start my first discussion here on Thoughtage and thanks to @Phil Tanny for getting this sight built. This discussion grew out of a conversation Phil and I were having on another site about his concerns with the potential dangers of CRISPR as a biotechnology. I won't rehash the entire conversation here, but the main question that arose was, "Does CRISPR in its current or future forms pose an existential threat to humanity on par with nuclear weapons?" I have some expertise on the subject and my view is that it is not an existential threat, but I thought that Phil had a great question about the current ease and power of the technology: "What kind of work could a biology grad student working on their own, with a budget of $10,000, do today?" I'll start off this discussion with a ballpark answer to this question which will hopefully spurs more discussion.

    Ok, here we go!

    Let’s start with the basics. As always, the best place to start when attempting to predict the future is the past. For decades, we have had the ability to take enzymes from bacteria to do genetic engineering in bacteria, but also in many other species including mammals (first genetically modified mouse was published in 1974). The ability to make those mutations has been most useful to humanity in that it’s allowed us to do experiments to understand the way that life works. This helps us understand how the universe works, our place in it, and generates useful tools that can keep us healthy longer. A decade ago, making a single transgenic mouse took an entire PhD (~5-6yr) now you can do it in 3-6months with CRISPR (if you’re an expert). That’s by far the most powerful implication of the tech. Our ability to use genetics to learn how our bodies work is significantly improved and, frankly, the halt of CRISPR for research is a nonstarter. So, to sum up superpower #1 of CRISPR is that it makes genetics in hard-to mutate organisms about 12-24x faster for people who know what they’re doing.

    Superpower #2 of CRISPR is that it requires a very small number of components to be operational and they can be directly injected to make genetic changes in one cell without relying on the cell dividing or mating, or other nonsense that you used to have to do if you wanted to breed a hotter pepper, or a nicer dog etc. This opens up the possibility of making changes in fully-developed adults. In the past, we thought the best you could hope for with gene therapy was to maybe prevent you from passing something on to the next generation.

    With those two superpowers, in mind let’s do some back of the envelope calculations on what a junior grad student (GS) could do. To make a transgenic mouse line for a reasonably talented student coming in with a BS in molecular biology would probably take 1-2yr to learn all the skills involved. A GS in molecular biology at many institutions makes ~$100 per day so the labor cost to learn the skills alone is not trivial. If we’re talking about the raw materials cost assuming you are a pro, you could probably make 2 transgenic mouse lines with 10k.

    With regard to the potential of CRISPR to be a civilization-ending power it’s important to think not only about what is possible today, but what will be possible in the future. It will only get faster and cheaper. What will that mean? It will mean that if the technology is implemented laissez faire (this is not a given, but let’s take the most extreme example to err on the side of caution) eventually it will become trivial to make whatever changes to an organism’s DNA you want. Ok, granted. In this end stage scenario the technology will be limited by the generation time of the target organism (gotta wait for stuff to grow up), the lifespan of the target organism (can only do so much damage in one lifetime), but most of all by the user’s decisions of mutations to make.

    I think there is a very natural analogy to computer programming here. The input devices to write computer code, such as keyboards, (CRISPR in this analogy) have existed unchanged for a long time and, while hardware imposes some time scales (generation time, lifespan), the power of computer programming is primarily limited by the user’s creativity (this is my experience with writing software, but I’ve certainly never done anything that’s pushing the envelope, so maybe this is a lack of experience on my part).

    So, now we’ve arrived at the zenith of genetic engineering, it may be that another technique supplants CRISPR before we get here, but the end stage of any genetic engineering tool is you can trivially make any change you want. I’m going to pause here to give you a chance to ask questions for clarification or bring up things you want more detail on. After I hear back from you, I’ll go through some of the worst possible scenarios I can imagine if we were to get to this point and there was a rogue or bad actor or just an accident.


  • Phil Tanny
    191
    Thanks for lending your expertise here Nick, I'm looking forward to learning more about biology. There's a LOT to discuss here, so bite off whatever interests you most, and I'll follow where you lead.

    To set this stage for readers, Nick is a biology expert, whereas my main accomplishment in the field was to pass a "Biology For English Majors" class about 50 years ago.

    A decade ago, making a single transgenic mouse took an entire PhD (~5-6yr) now you can do it in 3-6months with CRISPR (if you’re an expert). That’s by far the most powerful implication of the tech.Nick

    Ok, I understand this part, CRISPR makes gene editing far more efficient. Does CRISPR also make gene editing easier? Or just faster?

    Superpower #2 of CRISPR is that it requires a very small number of components to be operationalNick

    Can you give us a glimpse in to what such components are?

    This opens up the possibility of making changes in fully-developed adults. In the past, we thought the best you could hope for with gene therapy was to maybe prevent you from passing something on to the next generation.Nick

    So CRISPR opens a door on genetic based medicine? And what is the proper name for that?

    To make a transgenic mouse line for a reasonably talented student coming in with a BS in molecular biology would probably take 1-2yr to learn all the skills involved.Nick

    That's why I framed the question as I did. This suggests that many intelligent educated persons could probably access this technology if they put their mind to it. This matters to me as it suggests the potential for widespread adoption beyond the reach of governance mechanisms.

    If we’re talking about the raw materials cost assuming you are a pro, you could probably make 2 transgenic mouse lines with 10k.Nick

    Ok, so the learning time involved in the main barrier to entry, more so than other costs. Right?

    With regard to the potential of CRISPR to be a civilization-ending power it’s important to think not only about what is possible today, but what will be possible in the future. It will only get faster and cheaper.Nick

    Yes, that is my concern. Faster, cheaper, easier, more powerful, and more accessible to more people.

    This is just my speculation, but I'm guessing what seems to currently be the primary barrier to entry, knowledge, will be lowered over time.

    It will mean that if the technology is implemented laissez faire (this is not a given, but let’s take the most extreme example to err on the side of caution) eventually it will become trivial to make whatever changes to an organism’s DNA you want. Ok, granted.Nick

    We're agreeing here. What do you mean by "implemented laissez faire"?

    the power of computer programming is primarily limited by the user’s creativityNick

    This is a good example.

    but the end stage of any genetic engineering tool is you can trivially make any change you want.Nick

    Yes, we're on the same page. I'm obviously not that knowledgable about the details of how CRISPR currently works. I'm just attempting to use basic common sense to predict where this is going, and I see what you just said. Sooner or later, by some method or another, editing genes becomes trivial, and in the hands of an ever expanding group of users.

    I'll explore this in another thread, but to summarize, my concerns are less about CRISPR in particular than they are about the knowledge explosion as a whole. More on this later...

    Great first post from the forum's first member! I'm stoked!
  • Nick
    13
    Does CRISPR also make gene editing easier? Or just faster?Phil Tanny

    Both. In some sense faster and easier are not effectively different in a world where time, money are limited. But there are fewer steps involved that take less time.

    Can you give us a glimpse in to what such components are?Phil Tanny

    People are tinkering around with lots of different versions of CRISPR so the components can vary (CRISPR evolved in the wild so it comes in many different flavors) but the thing they all have in common is at least one protein that grabs DNA and cuts it and a guide RNA that integrates into the protein to tell the complex where to bind. The Cas9 protein, which is probably the most famous right now, can do all the protein functions itself, so you can have a functional compiles of just one protein and one RNA.

    So CRISPR opens a door on genetic based medicine? And what is the proper name for that?Phil Tanny

    I would say the door was already opened and was opening wider and CRISPR started us opening it faster than we thought we could. Lots of terms for it, people would know what you're talking about with your term but gene therapy is one I see thrown around too. Genetic medicine could also include treating someone taking into account their genetics without making changes so is a bit less specific.

    This suggests that many intelligent educated persons could probably access this technology if they put their mind to it. This matters to me as it suggests the potential for widespread adoption beyond the reach of governance mechanisms.Phil Tanny

    I agree that it puts it in reach, but I don't agree that just because something is widespread means it can't be productively governed (I know you'll have lots of thoughts on this, but I'm going to leave it to that for now to try to get through some of your other questions).

    Ok, so the learning time involved in the main barrier to entry, more so than other costs. Right?Phil Tanny

    Depends on whether your time or money are more limited. I think the point is that either way, the barrier isn't insurmountable.

    We're agreeing here. What do you mean by "implemented laissez faire"?Phil Tanny

    Full steam ahead, no regulation, no moratoria.

    I'm just attempting to use basic common sense to predict where this is goingPhil Tanny

    I think you're coming at this from a good place, but humans have a pretty terrible track record of predicting anything with common sense (or uncommon sense for that matter). Given that there is a definite cost to not researching and using gene therapy I think the path forward should be carefully advancing technology, not stopping because of predictions that may never come to pass.

    We will of course disagree on the word "careful" but as I've mentioned (I think in our previous discussions, but maybe not on Thoughtage) when compared to pioneers of previous technologies, Jennifer Doudna has been an uncommonly strong voice for the regulation and education of a new, powerful technology.


  • Phil Tanny
    191
    Hi again Nick, thanks for keeping this discussion alive.

    Jennifer Doudna has been an uncommonly strong voice for the regulation and education of a new, powerful technology.Nick

    Best I can tell, this is true. I don't doubt her good intentions.

    I just think the concept of regulation is on the edge of becoming outdated thinking. It seems a reasonable plan so long as the technology is limited to the professional class. These folks have a lot to lose, and thus are susceptible to being regulated.

    I think we should assume that bad and stupid people will use this technology too, as I think Doudna does assume. Then the question becomes, how much trouble can bad actors create with this technology? What is the scale of the power involved here? That question can be divided in to the current moment, and reasonable projections of where this is headed.

    Honestly, I'm less concerned with CRISPR specifically than I am with the larger question of the knowledge explosion as a whole.
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment