The Present And Future Of Gene Editing - EP 54 Jennifer Doudna

One word that comes to mind when you ask what gives me confidence, and the word is viruses. And the reason is that this is exactly what viruses do. They target specific types of cells in the body and they don't mess with any others, and they're very good at honing in on those cells. And so when we think of, for example, everybody's probably familiar with COVID-19 of course, and that is a virus that targets the respiratory system. It doesn't target other types of cells for the most part, right? But it's targeting those respiratory cells. And that's because it is looking at a molecule that is uniquely produced on the surface of that type of cell. And that's really how viruses work.

Great to be here, Ashley.

10. Turned 10.

Chirp chirp.

Yes. Yeah. It's such a small school that you... I don't know. I've only ever run into like a handful of people who went to Pomona in my professional career.

Well if it's any comfort my son did not get into Pomona either.

Oh, well that's very kind thought.

End of 2023.

Correct.

It's really interesting Ashley because CRISPR in the format that you just mentioned where it's being used to treat sickle cell patients, it brings together fundamental science in a fascinating way. It's not only CRISPR technology that's being used to manipulate the genome in patients, but it's also the knowledge of what gene to target. And a casual listener might say, well, wait a minute, we've known about the sickle cell gene for decades, and that's correct. However, the way Casgevy works is a little bit interesting in the sense that the target is not the sickle cell gene itself. It's actually another place in the genome that controls the production of a separate protein called fetal hemoglobin.

Normally, fetal hemoglobin, which helps carry oxygen in our blood when we're fetuses, when we're developing, gets turned off when we're born. And at the same time there's activation of adult hemoglobin. So what happens in sickle cell patients where they have a mutation in the adult hemoglobin protein, they start making a deficient form of that protein which means that they become susceptible to sickle cell crises and all of the horrible phenotype that goes along with that mutation. So what happens with Casgevy and that form of CRISPR is that it's used to turn on the production of fetal hemoglobin and keep it on.

And that means that instead of producing only this mutated form of adult hemoglobin, patients now also make fetal hemoglobin which is perfectly functional and it allows them to live a normal life after treatment.

This is what's so incredible about it. It literally is a one-and-done therapy. And we call it... it appears to be what we call a functional cure. That means that it's not actually correcting the disease-causing mutation, but what it's doing is overriding that mutation. It's making a second protein that makes those patients now not susceptible to their disease. And I'm kind of building somewhere because we're going to talk about baby KJ Muldoon in a second as well. But just for some context, I mean, so it's amazing. It's this one-time cure and but it is expensive and it's quite a difficult therapy to administer. I mean...

Yeah. You're alluding to the fact that it's administered by taking blood stem cells from patients into the lab, doing the manipulation with CRISPR in the lab and then transplanting the edited cells back into the patient using a bone marrow transplant. So it's an arduous procedure. It takes many weeks. It's very expensive. Not ideal.

Oh, 100%. But you know, Ashley, I think this is the way that technologies often roll out, isn't it? What happens is there's an initial breakthrough with a new technology. People get very excited. They start thinking about how it can be used to solve real problems. That's what happened with CRISPR in the case of sickle cell. I mean sickle cell anemia has been known for so long and so many people suffer from it without any effective treatments. And so with CRISPR we now have a tool, a technology that we now know can actually provide this kind of a therapy to patients. It's just that we're now at the point where okay we know it works but we got to get better at how we deliver it and we got to get better at how we manufacture it. So that we can reduce cost. I think those two things are the biggest ticket aspects of the technology, the cost of making the molecule and the cost and time of delivering it into the patient. And those are two areas where we're, as we'll probably get into, we're very actively working on both of those things at the Innovative Genomics Institute.

Yeah. Oh absolutely. And can I just say a couple more things about it because I think one thing to keep in mind is that I think it was really a watershed moment when that drug was approved by the FDA and by other agencies in the UK because it showed everybody that this is now a real thing. It's not a fantasy. It's not a sometime in the future maybe we'll have a therapy. We have one and it works and it's approved. So that's important.

And I think it also was a moment for me at least when, as you have pointed out, we're not done and there's a lot more work to do but it makes it very concrete. You can see the result. I've met patients that have received this therapy namely the most prominent one probably is Victoria Gray.

Well let's see. So, Victoria Gray received the therapy in 2019. She was 33 at that time.

And the story she tells is so moving. She's a mother. She mentioned that one of the motivations for her to get the therapy, even though it was high risk at the time, she was the first US patient to receive the CRISPR therapeutic, she was motivated by her son who wanted to be able to spend more time with her, and she wanted to not have to go through these sickle cell crises that required hospitalization every few weeks. And it's really exciting to talk to her because she explains very well how her life changed completely with this therapy and it makes me and I think anyone that thinks about it or interacts with somebody like that you think gosh I want to make that possible for everybody who could benefit from it. So it's very motivating. It really gave her a new take on her life. It allowed her to go back to work. It allowed her to play with her kids and be engaged in their lives in a way that wasn't possible before. And she's also become quite a spokesperson for the therapy because she, like me, she wants other people to know about it and to be able to get access to it.

You are. And so the thing to understand about... and I get this question a lot, and it's a really good one, is that, does this mean we're editing every cell in the bloodstream? No. The answer is no. The way the therapy works is that the editing occurs initially in blood stem cells. What are those? Well, they're cells that are immature, but they give rise to adult blood cells and fully developed cells that can carry oxygen in the blood. And so, when the editing happens in the lab, those cells are now capable of going on and developing into their mature form, but they now are producing fetal hemoglobin. And that's a permanent change to the DNA, permanent. And so, they get transplanted into the body. They start dividing and replicating and developing. And it's interesting because they are healthier than the cells that are only making the sickle cell form of hemoglobin. They ultimately take over the blood supply and so it's an interesting case where there's actually sort of competition going on between blood cells in the body and the healthier ones prevail.

Have not seen side effects? No. So it's really interesting and now dozens of patients have, if you count all the folks that have been through the various trials that have happened, have received this therapy and it looks like it's a functional cure.

Yeah. 2025. Yeah. Absolutely.

Well, you know, Ashley, I'll just start by pointing out that I was flipping through Nature magazine recently and they have their Nature 10 in 2025. So, every year they point out the 10 scientists who they think are making the biggest impact in science. Number three on the list was Baby KJ.

And it was so exciting to see that for me because I think it again it really captures the enthusiasm, the excitement, the passion that his story was met with by so many people that I've talked to across the spectrum not just scientists of course but many others. Why is that? Well, you know, as you said, KJ was born with a rare metabolic disease so severe that he could not digest protein. And if you can't digest protein, especially as a baby, then you can't grow very well. And so, he was extremely sick. He was in the ICU. He had a terrible prognosis. Not much that could be done. Fortunately, his clinical team at Children's Hospital of Philadelphia recognized that he had a mutation in a gene that was potentially correctable using CRISPR.

Yeah. A single mutation. And so they worked with a large team of people spread out across the US that included the Innovative Genomics Institute, also included the Broad Institute, also included a company called Aldevron that makes molecules, included the FDA of course in terms of getting the right approvals to use this clinically and that led to very rapid development of a CRISPR modality, a form of CRISPR that could be delivered in very rapid order into this patient after appropriate laboratory and animal testing showing that they could correct this mutation. And they're using a format of CRISPR that was developed in David Liu's lab at the Broad as you reported and it was an off-the-shelf technology these days, it's well published, well vetted and tested in many labs. So that form of CRISPR called a base editor was used together with another well-established therapy for delivery to the liver which is the organ that was appropriate for this particular baby's situation and together those, the delivery and the CRISPR technologies could be deployed very rapidly in this case in a matter of months into this patient. And again it apparently, this appears to be a functional cure in the sense that this boy received three injections of the CRISPR therapy to ensure that he had sufficient numbers of cells in his liver that were corrected and then he was sent home and he turned one and he looks very healthy.

You know I think that's right and it was very exciting again it was sort of like the Casgevy moment to me but in the sense that it was another watershed example of how CRISPR can be leveraged to cure a rare genetic disease in real time in a time frame that was meaningful for that patient. It wasn't done much later than would have actually helped him. It was actually done in a time frame that was useful clinically. And again like the Casgevy case, it's incredibly motivating because on the one hand you can see how possible it is now to do this type of thing for rare disease to diagnose it, to create the CRISPR therapy quickly, to test it in appropriate settings quickly and then to work with regulators to make sure that it can be brought to the clinic in a time frame that made sense for the patient. But it was expensive to do it and it was a very hodgepodge effort. I mean it required pulling together many different players and there was a lot of serendipity involved and it would be great to be able to make that process cheaper, faster, easier to deploy. I mean that's really I think where again we hope we can play a role at the IGI is to make it easier, streamline that process in the future and that was a big motivator for setting up this center for rare disease at UC San Francisco that is just opening and we hope to treat our first patient in April.

Well, I think there's both honestly. Now, we have some great editors and as these examples we're discussing just demonstrated, but yeah, the big challenge right now is... I would say the biggest bottleneck in my mind is the delivery. How do we get these tools into the cells, tissues, organs where they can be effective in the body? Or frankly in any organism. I mean, we may talk a little bit about plants, I'm assuming, and delivery is a challenge there as well. So, this is a kind of a universal challenge. How do we get editors into cells where they can do their work? And but on the... let me just before we leave the tool part of it, the CRISPR part of it, I think there's still a lot of creative work happening in that space too because as you mentioned, we had sort of CRISPR v1.0 maybe where you could cut DNA, you could induce cells to repair it with the introduction of a small change that could disrupt a gene. That's how the Casgevy form of CRISPR is working for sickle cell patients. Then there are base editors that make a chemical change to a single letter of the DNA which is maybe v2.0, right? And then prime editors maybe going... maybe we would call that v3.0, where we're actually rewriting little specific snippets of sequence and making targeted changes. And these are all great. They're not... I don't in my view I don't think that particularly that prime editing is a technology that's yet at the point where it's clinically useful because it's a big molecule. It's a bit unwieldy. Talk about delivery challenges. That's a big one right there. And controlling it the way people might like and improving its efficiency I think is still to be done. So that's... it's still very much a developing area of the field and of course there are other approaches that are coming along as well for being able to introduce new snippets of DNA into genes to make corrections and people are also thinking creatively about how you could design CRISPR entities that would be able to make wholesale changes to genes that would correct whole families of mutations because in many cases you have diseases that involve a gene, but if we look deeply at what particular mutations are causing that disease, there's a whole collection of them. So we're going to need either many forms of CRISPR depending on the patient's mutation or we're going to need a form of CRISPR that's able to cover them all. And so, that's still very much being worked on. And so there's a lot of research going on and lots and lots of publications and preprints that are coming out all the time with new clever iterations on those themes. But that being said, and I'm very confident that that whole toolbox for manipulating genes is developing very rapidly and it's a very exciting area. But even if we have the best tools in the world for manipulating DNA, if we can't deliver them, then they're not very useful.

So that's the challenge.

Well, there's one word that comes to mind when you ask what gives me confidence and the word is viruses. And the reason is that this is exactly what viruses do. They target specific types of cells in the body and they don't mess with any others and they're very good at honing in on those cells. And so when we think of, for example, everybody's probably familiar with COVID-19 of course and that is a virus that targets the respiratory system. It doesn't target other types of cells for the most part, right? But it's targeting those respiratory cells and that's because it is looking at a molecule that is uniquely produced on the surface of that type of cell and that's really how viruses work. If we think about another virus that probably most people are aware of, have heard of, is HIV. And so HIV is a virus that infects immune cells. And again it doesn't affect random immune cells. It's infecting human T cells in the immune system. And that's because again it's looking at a very particular molecule or collection of molecules that are made on the surface of those cells. And so knowing that, I think one alternative to lipid nanoparticles that's also developing rapidly and multiple parties are exploring this is using the strategies of viruses but to deliver molecules that we want, not viral genomes for infection but in fact molecules for editing for example and that's an approach that is targetable and programmable. And so the exciting thing there is that once you understand how viruses take over and infect certain cells, then you can take that same mechanism but use it for the purpose of delivery of other kinds of molecules.

No, it's because delivery is a really hard problem and it's a problem that affects all aspects of pharmaceutical development. It doesn't matter if we're talking about editors or small molecule drugs or antibody drugs, right? We could be talking about any kind... or peptides, RNAi for people that know about that. Any of those kinds of entities face the same challenge. And so it's an old problem and it's a super hard problem. And so lots of companies have of course been interested in this and interested in solving it. The thing is that I think that for many companies they need to focus on their disease areas of interest. They need to get things done in a time frame that's meaningful to investors. And so in a way they really are not usually in a position to do long-term investigation of new delivery approaches and methods and understanding the chemistry and biology underscoring these types of methods. And so as a result there's been a bit of a hodgepodge... I said that earlier, that same word, but it's sort of a collection of different kinds of delivery strategies that are for the most part either targeting the liver which is a tissue that's particularly adept at taking up molecules which makes it wonderful in that regard but if you want to deliver somewhere else it's actually problematic because you don't really want all of your molecules trafficking to that tissue and not to the tissue that you're trying to target. So I believe strongly, I've come to feel very strongly that this is actually an area where nonprofits like the Innovative Genomics Institute really should be working because what we can do in an interesting way uniquely that I think companies for the most part can't is that we have a very long timeline. We can harness the creativity of student trainees who come in who are very motivated by these kinds of problems. They want to apply their smarts and their intuition and their creativity to solving these kinds of real world challenges and they're not afraid to take a risk and they're not afraid to try things that really might not pan out but boy if they do they would be huge. And I love doing that kind of science. It's the kind of science I've always tried to do in myself and in my own lab. And so I'm now excited to be enabling others that are coming to our institute to look at problems like delivery and be really creative about it and recognize that they could have a huge impact if they can solve this challenge and that we really need their intuition and their willingness to take a big risk to do it.

I would... in the early days of CRISPR I wondered about it, right? Because I really didn't know how long it might take to take something that was published in an academic journal, was clearly exciting to a few academic scientists maybe more than a few, but it really wasn't clear what the timeline would be to going from that point to having something that was safe and effective in a human for curing a disease. And we're there now. And we're not there now with just one thing, right? We're there with multiple things. And so, the fact that it's still expensive and hard to deliver, etc. No. Am I worried about it now? No. I'm not. I think it's going to be solved. And the reason is that there's so many brilliant people of all ranks and from globally now working in this area of science that if it can be done it will be done. It's always hard to predict the timelines for these things, but I certainly think that nobody would have imagined, I certainly wouldn't have imagined that within 10 years of that original publication about CRISPR that we published in 2012, this is with our collaborator Emmanuelle Charpentier, that within just over 10 years from that point that we would have an approved therapy for a rare disease with CRISPR. Stunning. I don't think there's another example, at least in the biotech world, of a technology that's moved that fast from fundamental discovery to actual practical approved therapeutic. So now it's not can it be done, it's just how to do it faster, better, cheaper.

Well I often think of my life BC, before CRISPR, and after. You know it really did change. It was almost like a step function for me. And it happened really almost immediately when we published the paper in the summer of 2012 because it was clear immediately... I knew when we hit submit on that paper that it would mean that a lot of people would read it and recognize that it would be very easy to test this technology for making edits in their cell type or gene of interest. And that's exactly what happened, right? Lots of people then quickly started to deploy it and test it. And even before the end of 2012, even though it wasn't published yet at the time, I was getting emails from people who were starting to use it, not just in cells, but in whole organisms like zebrafish...

You know, for editing, right? And so it was clear that it was going to really change the way we did biology, but the way we did research. And of course, you know, we were all imagining at the time that, you know, there would be opportunities for practical applications, although as we just discussed, it wasn't known and nobody could really predict how fast that would move forward. So, yeah, I just got—I had to really grapple with it. And it might sound a little bit strange, but I had a lot of anxieties in the beginning about CRISPR, partly because of the clear implications of the technology for things like germline editing, which is different than what we've been discussing because it means making heritable changes in the genome that are passed on to future generations. So that's, in my book, kind of a different—I'd put that kind of in a different bucket. And then there were of course all of the opportunities that we haven't yet discussed in other kinds of organisms, not just in humans, but in plants, in microbes, in organisms that affect our health, including fungal systems and of course all kinds of bacteria. So it seemed to me that there was just so much to do. And so I had to really ask myself first of all, how do I want to be involved in that? And it was never a question of am I going to be involved. It was just really how does a biochemist who's always done very fundamental research, how do I do this? And so I really had to make a decision about how to focus my own efforts and energies. And honestly, that was one of the motivations for starting the Innovative Genomics Institute was that I realized it was much, much bigger than me or my lab or even all of the other researchers that I knew at the time that were getting involved in it. It was just bigger than that, you know, and it was going to be bigger. And if we wanted to really have a lasting impact that was global, we had to figure out how to build an organization that would invite in scientists. We're a big public university, University of California, Berkeley. And so we have lots of undergraduates that are wanting to learn how to do research. We have lots of graduate students and postdoctoral trainees that come through our labs. And so the question was how to help all of them get engaged with this field and figure out how they could contribute to it. That's what I really wanted to do. And I wanted to make sure that University of California was at the forefront of guiding kind of what was going to come with this technology. And I was also a big believer that companies had to be involved obviously in terms of scaling what would happen, but that there had to be a partnership between those companies and their investors and their teams and a nonprofit or nonprofits like ours that are focused on innovation, that are focused on very early stage discovery. That's high risk but when it works, really high payoff. And that's where I always like to be myself, is kind of at that bleeding edge.

I guess that's fair. I really want to, you know, I want to make sure that people understand the technology. I like to think that they can come to me or to our organization with a sense of trust that they can trust what we're doing and what we're saying and that we're scholars first and foremost and we're educators, right? And that's what we do. And so we're really all about understanding the science, communicating the science, thinking clearly we hope about where it's headed, but not overpromising, not getting ahead of ourselves, not getting over our skis too far. Always thinking about what's possible and what could happen, both in terms of the positive but also thinking about the potential negatives and risks and moving forward in a thoughtful fashion. So that's really what I try to do, that's the kind of leadership I want to provide at the IGI and that's the way I encourage our students and our trainees to approach their work.

No, I definitely look for lessons and I've been fortunate that I've had a lot of colleagues that have been helpful in my thinking. So, I'm thinking here of a few people that pop to my mind. Janet Rossant who has worked for a long time on understanding very early human development and in vitro fertilization and was very supportive of my thinking as we were navigating those very early days of CRISPR and thinking about where it could go in terms of understanding human biology but also impacting and altering it. So she was very helpful. George Daley at Harvard Medical School, same thing, very helpful. And there were many others but, you know, I think that I'm fascinated by science. I love learning new things. I've been so, I guess, honored that I've found myself, you know, these are the cards I've been dealt, right? I mean, in a way. And so I'm playing the hand I was dealt. And part of that involves just meeting and interacting with so many people that I'm sure I would have never had opportunities to meet and talk to and work with if it hadn't been for CRISPR. Yeah. So in a way I feel like I'm just trying to embrace those opportunities and you're right sometimes that comes along with some stress and some difficulty. There's no question but it's part of the journey.

That's right. We're about 70% funded by donors right now.

Yeah, on some level, you know, a lot of what we do is work at the intersection of technologies. So CRISPR is there but, you know, we're also increasingly using AI as so many others are. We're also using imaging technologies, robotic technologies. I mean, these are all kind of really exciting areas of development that when you look at the intersection, you can do things that just recently weren't possible.

Well, let's start with the cow. I mean, I think that's one that, you know, boy, ask me if something that I never thought I'd be involved in working on. That's it. But it's so exciting because when I learned that not only is methane one of the worst greenhouse gases, but it's also one that a lot of it is produced annually by human activities. A big one is agriculture. And so it's cattle farming, it's also rice farming. Both of those things release a lot of methane around the world every year. And so, you know, when I first started mentioning to people that the IGI had a big effort in agriculture, I got a lot of, you know, strange looks like, what does CRISPR have to do with that? What does CRISPR have to do with climate change? And the answer is, well, it's a technology that is crosscutting. It allows us to manipulate DNA of any kind of cell or organism. And that means that when we think about the challenge of methane emissions and we look at cattle, for example, well, where's that methane coming from? It's coming from microbes in the cow rumen and it's coming from microbes that have particular metabolic pathways that produce methane and so it gets burped out. It gets farted out. But mostly burped I'm told.

Yeah, we're doing it all here today. And so it means that CRISPR can come into play because we can use it not only as a research tool to figure out which genes could be manipulated to reduce methane emissions, but then we can actually do it in a, we hope, permanent way in cattle. And so the idea would be to be able to modify the cow microbiome essentially at birth in calves, once, kind of again a one-and-done manipulation of their microbiome that would be maintained through diet and or the way that they're raised by farmers such that they now are methane-free. And by the way, that also means that instead of that carbon getting released as methane, it's actually deployed in other metabolic pathways that lead to increased milk and meat production by about between two and 5%. Has been demonstrated by our team up at UC Davis.

No, we have cows doing this.

Yeah, they're up at UC Davis.

So we—the IGI now has 24 cows that have been born where their microbiomes are being manipulated and we are monitoring their growth, their, you know, how much weight they're gaining, etc., and also how much methane they're emitting. And the purpose of this is to ultimately come up with a strategy that could be deployed inexpensively and sustainably to farmers and farms around the world. And we hope that farmers will be excited about it because if they understand that not only are their cattle methane-free, which, you know, they may or may not care about that at the end of the day, but they probably really care about having cattle that are 2 to 5% more productive. Yeah. So, it's great.

That's it. That's right.

Yeah, so we're administering it using—we can administer it using viruses that only infect those kinds of bacteria and they take the CRISPR molecules into the bugs.

We want it to be a one-and-done thing.

No, you were not. But this is one of those fascinating aspects of CRISPR that's evolving in real time because I think regulators around the world are grappling with this technology. How do they think about it? How do they regulate it? How do we make sure that it's used safely? But in my opinion, I don't think we want to say well we're not going to use this ever. I think we need to figure out how to use it and how to use it appropriately. And so you may have read that here in the United States, our US Department of Agriculture, the USDA, which regulates our agricultural products, they have decided that they've ruled that technologies like CRISPR that manipulate genes in plants that could be manipulated by traditional plant breeding if we waited long enough—but we now have a technology that doesn't require us to wait maybe 15 or 20 years of crossbreeding, we can actually just go in and manipulate a gene directly. If that can be done with a technology instead of traditional plant breeding but it's ultimately otherwise the same kind of change to a gene, it's not regulated and it's not called a genetically modified plant. Whereas in Europe that exact same manipulation is considered genetically modified because a technology was used to do it. And so I think that this is an area where there's still of course very active discussion, groups in different countries trying to figure out what's the right way to proceed. But I'm pretty confident that in the long run, I'm hopeful anyway that these regulators will recognize that if you're making a change in a plant that is a natural variation that could be achieved, quote, naturally with plant breeding—although we could argue about whether that's natural either—but whether you do it with that approach or you use CRISPR, if you get to the same end product, then it's not a genetic modification in the way that people have called GMOs.

Well, I guess it depends on how you define dramatic. I think if you define dramatic by more people directly impacted, oh yeah, agriculture for sure, right? For sure. It will be much more impactful in that regard. If you're defining dramatic as attention-grabbing, headline-creating, that kind of thing like the baby KJ story, then probably it'll be in healthcare if I had to place a bet because I do think we're going to continue to see really exciting advances with CRISPR for treating diseases of various types. There's also, as you may be aware, companies that are excited and academic groups too about preventive care with CRISPR, right? Could you prevent somebody from getting a cardiovascular disease or a neurological, especially a neurodegenerative disorder? Could you prevent that before it ever starts by making an appropriate tweak to their genome? And so that's something that I do think is also potentially coming in the world of gene and cell therapy with CRISPR that we'll see in the next, yeah, maybe 10 to 20 years that also could have very big impact. It's just that it's harder to predict the rollout of something like that.

No, I think there's a continual catch-up game going on. It's similar to what you described in terms of rocketry or certainly AI. And I think, you know, that's probably true for a lot of technologies where there's a rapid development and then a struggle for agencies and regulators certainly to catch up and stay abreast of what's happening in the world of CRISPR. I think in terms of regulating it and ensuring that it's used safely and effectively, but also that it's used in a timeline that, you know, is sensible regarding especially rare diseases that are quite severe and where there's no other option right now for patients. I think that's where there's a very interesting opportunity right now for regulators to be a little bit bold, which is probably not something they typically want to do for obvious reasons perhaps, but I think there's an opportunity to think differently about the way that a therapy for a rare disease like that will be created, tested, and deployed. If we can come up with a streamlined strategy that probably increasingly does take advantage of AI in the sense that I think as we have more and more data that point to the effectiveness of particular CRISPR modalities, let's say for how they work in different kinds of cells and targeting different genes—we didn't get into the weeds of how the CRISPR technology works but it's a programmable molecule that means that a small piece of it can be changed and it can be directed to different genes. And so it may eventually be possible to predict very accurately how those programmed changes to CRISPR affect its accuracy and its effectiveness. And when we can do that, then you could imagine at some point at least being willing to forego a lot of the experimental work that's currently required because you would say, well, I trust this prediction and I know it's correct, so I don't actually have to go in and test it. Instead, I can move on to, you know, the final steps perhaps of deploying it into a patient. And so, I think that's coming. I really do. And again, it's always hard to predict timelines. Currently, one of the things that's on my mind a lot is how do we get the kind of data accumulated that we can use for training models like that because if you're going to have accurate models that predict things accurately, you've got to have good data to train on. And so that's something that I'm in very deep conversations with both companies and academic groups right now about how to do that, how to organize that better than we typically have in our field and how to make those kinds of data available to many people who are developing interesting creative models and approaches. I would like to see deployment of those data widely so that we get the best chance of getting really good predictive models in the future.

Well, my take on that situation—and I feel like I'm not the most knowledgeable in a way because many others have actually been to China, been to these hospitals and companies that are working together to do rapid clinical testing. I'm just hearing about it from those folks that have traveled there, but I haven't been there myself. I want to just put out that disclaimer. But my sense from many, many conversations and of course lots of people are discussing this right now is what's going to happen to the US biotech industry and to investments in that industry as China continues its rapid advance in the space. And the bottom line for me is that I really care about patients. You know, I really want to see these therapies developed safely and quickly to help people. And so I want to see that done however and wherever it can happen. And China is clearly a really big and really important player in that space. So I think I'd like to see better coordination because I think there are things that happen—again, this is just my opinion—I think that there are things that happen better currently in the US. I think the US is still really good, maybe number one, at innovation in the bio and other areas for a variety of reasons, but I think we're just—it's our culture partly. We're just really, really good at innovation. China is really, really good at implementation. And so, you know, right? And so those are—and you could view those as competitive but you could also view them as complementary. And so I'd like to see increased coordination there. Of course that's not the direction that our politics are going right now but I can tell you that on the ground among scientists, among even companies, you know, there's a lot of really interesting effort to improve that coordination. So I think if that is successful then maybe it's going to turn out to be win-win. This is my naive—this is a spot where I really hope China and the US could come together. It feels like we could in this particular moment in time go so fast with such important stuff.

I suspect it'll be a combination. I think we will continue to be inspired by nature for sure because let's face it, nature has been experimenting for a lot longer than any of us. And the data that AI are trained on is coming from our own experimentation and the collections of data that we're feeding to these models. So ultimately if we want to tap into nature, tap into the largest set of data in the biological world that we could imagine or that we have access to on this planet anyway, it's evolutionary data, right? So it's really coming from nature and all of the experimentation that's happened over time. Of course we're only aware of some of it because we don't see all the failed experiments, right? A lot of those things never panned out and we don't see evidence of them now. But to the extent that we can look at—and this is where frankly I'm fascinated by viruses and by the relationships between viruses and the cells they infect because I think that more than anything else in a way is a very deep, very ancient record of the kinds of experimentation that these systems have gone through as they try to reproduce and they get challenged and they have to come up with other strategies or they have to fight off an invader. This is a really powerful driver of evolution on basically every level as far as I can tell. And so that's a really interesting area. That's a really interesting type of data to be mining.

No, I do. I do. I mean, I'll give you an example from our own work currently. So you mentioned antibodies and that's one area but, you know, these kinds of models are being used to design all kinds of molecules now. And so, you know, we've been working recently on trying to identify and isolate molecules from bacteria that have new kinds of activities, new functions. And the way we're doing it is we're using AI models that can design from scratch. Of course, it's based on what they're trained on, but they can design molecules that will interact with something that we tell it to look for. And it works, right? It's quite astounding. And so that is really enabling. And so it means that increasingly we don't have to do a lot of, you know, bushwhacking in the lab. We can just go directly to molecules that we're pretty confident from the get-go are going to be useful for the thing that we want them to do and just get going on kind of the next step. And so what I tell my students is that I feel like as AI continues to advance and gets trained on better and better data sets for various kinds of tasks, that increasingly it's going to streamline what we do in the lab. We're not going to have to do nearly as many experiments because we will be able to predict which ones to do and which ones are most likely to work and which ones are frankly a waste of time.

Again, I guess I do think that this idea of preventive medicine is very interesting. I do think that the opportunity to protect ourselves from disease before we become sick is really exciting. It's a different way, you know, of course, I'm not sure our healthcare system actually wants that, but it would be transformative if it could be done.

Probably they don't, if we're honest. Right. And so this is a big challenge in the field right now is, you know, I don't know how it'll get resolved. Nonetheless, imagine that, you know, imagine that you could perhaps in the not too distant future, you could have access to your genetic information, which many people do already, and you could interpret that information, which increasingly we're confident in the way that's being done. And you could deploy CRISPR safely to tweak a few genes that would be protective against Alzheimer's or heart attacks or even infectious disease. This would be absolutely transformative and if it could be done quickly, safely, accessibly, again, transformative for health. So that's something I'm very excited about. I don't think a lot of people are—that's not maybe top of mind when they think of cell and gene therapies, but I do think it's an interesting direction to pay attention to in the future.

They are. But, you know, I'm a realist. I mean, I've been around long enough to know that, you know, biology is hard. Biology is complicated and it's hard and it's harder than people think. And you layer on top of that drugs that can actually be effective for things. And that's incredibly hard. And so, I think that, look, I welcome creativity. I love the fact that young people are excited in this area. People who are just starting out in their careers want to make a difference, that's great. I want to encourage that. At the same time, I suspect that, you know, reality being what it is, a lot of those efforts will encounter challenges that are going to be hard to overcome. I certainly hope that people continue to explore. I think that's where new breakthroughs and innovations come from. Frankly, this is something I have loved about being a scientist working and living here in the Bay Area because I do feel that we have a unique culture here. You know, I've worked in other parts of the US and I've traveled all over the world meeting scientists at various institutions and I think there's something quite unique and special about the Bay Area. And I love the fact that we have increasing opportunities to bring different kinds of science and technology together. So, I love the fact that AI, which is, you know, in many ways kind of centered here in San Francisco, that we're able to bring that together with things like CRISPR so that we can do things that have never been possible in the past. So, that's where I get really excited.

Thanks so much, Ashley. Really fun conversation.