In this episode of The Future of Medicine, we welcome Priscilla Chan, MD, pediatrician and co-founder of Biohub, a first-of-its-kind research initiative combining frontier AI with frontier biology to accelerate progress toward curing or preventing all disease.
Dr. Chan shares how her experience caring for children with rare and undiagnosed conditions shaped her commitment to transforming how science is done. She discusses how patient-led research communities are driving breakthroughs, and how new approaches to data sharing and collaboration are reshaping the pace of discovery.
The conversation explores Biohub’s work to build a “virtual cell”—a computational model designed to simulate human biology—and how advances in artificial intelligence, large-scale datasets, and imaging technologies could allow scientists to better understand disease at its most fundamental level. These tools may one day make it possible to predict disease risk earlier, test interventions virtually, and personalize treatment based on an individual’s biology.
Dr. Chan also reflects on the future of medicine, where the boundaries between research and clinical care continue to blur, and where physicians increasingly engage with data, biology, and technology to guide patient care in real time.
Looking ahead, she shares her vision for a more proactive and precise healthcare system—one that moves beyond treating illness to anticipating and preventing it.
The Intro
"It's not some distant far away land where papers get published and then 20 years down the line we get a drug."
Dr. Priscilla Chan is co-founder of Biohub. 10 years ago, Dr. Chan and her husband Mark Zuckerberg launched their audacious mission to cure or prevent all disease.
"The science has moved incredibly quickly because the patients are bringing assets to the table."
In this fireside chat, Dr. Chan speaks about how her time as a pediatrician shaped her vision today, what Biohub is doing to try to compress decades of scientific discovery into months, and how technologies like the virtual cell and AI might transform both research and clinical practice.
“We are fully committed to building tools faster, more effective, and more efficient.”
Her vision for the future of medicine may sound like science fiction, but with the power of AI, their team believes that soon enough it will be very real daily practice.
Welcome to Stanford Department of Medicine's inside look at the future of medicine.
The Transcript
Dr Euan Ashley: We are very excited to welcome you to Stanford. Thanks for being here.
Dr. Pricilla Chan: Thanks for having me.
Dr. Ashley: When we think about the philanthropy and the work you've done through CZI and Biohub, it seems, I think, to flow from your identity as a physician. I mentioned your pediatrics residency. You've talked in the past about how you remember times when people would show up with really very little knowledge about the rare condition that their kid would have, and wanting to really try and help those kids. Talk a little bit about your inspiration for the work that you're doing with CZI, and we'll get on to some of the newer things relating to Biohub shortly.
Dr. Chan: In pediatrics at UCSF, we saw kids from all over come. We just didn't know what they had. And oftentimes, you get a printed PDF or you're trying to search on PubMed, a name of a gene, or maybe a bunch of symptoms, and there's just not very much information available, and there's definitely not a clear way to translate that to what you're supposed to do when they're hospitalized with you. That was my first taste of, oh my goodness, this is where basic science needs to come in and is the source of hope. Because I will tell you, when I was applying to residency, a beloved mentor of mine was like, you're not going to go down the basic science route. And I was like, okay, whatever it takes to get into residency, right? And so I sort of went down a different route around leadership and service. But it was those moments when I was like, this is all we have, and this is all that this family is clinging to and in search of answers for their child. So when we had the chance at CZI to work with rare patient groups, I remembered exactly what this was like. And in 2019, we started the program where we gave some seed funding to patient advocacy groups. At first I thought — my imagination had not been open to what they were going to do. I was imagining sort of patient support, walking each other down the route of diagnosis and treatment options awareness, but what actually happened was phenomenal. And now we've done multiple cycles of the rarest ones, where we give funding to rare disease groups and they have since taken that funding to convert it into research agendas for their disease. 50 groups working on rare diseases, about 20,000 researchers now plugged into those research agendas, and the science has moved incredibly quickly because the patients are bringing assets to the table — incredibly important assets without which the research would not happen. They've built disease models, they've helped cultivate cell lines — hundreds of cell lines, hundreds of disease models — then clinical registries, biobanks, clinical studies, natural history studies. All of that makes it easy for folks who then have the skills in basic science to contribute and actually move their science forward in a way that will make a difference for their disease. One that I'm particularly excited about this week is around the FOXG1 mutation, which causes a rare neurodevelopmental disorder in children. The group just got FDA approval for their clinical study for gene therapy. And I can't even say this without getting a little teary. The mom who started the group promised her daughter they'd have a clinical study by the time the child was 10, and they got it approved on Amara's 10th birthday. That is incredible. It's also incredibly fast. So that work has been incredible and inspiring to see how you can do science differently.
Dr. Ashley: It's really just incredible to see what these rare disease communities can do. As you know, we've been involved here through our Undiagnosed Diseases Network, which you have supported — and thank you for that — for many years with the rare disease community. But I think until you really look straight in the eye a rare disease parent, a rare disease family, and understand where they're coming from — they talk about being on an island often before they're diagnosed, and then joining a community when they find that diagnosis, and how that's so empowering. When I think of that and compare it to what we often get from reviewers, which is, where's the actionability in your diagnosis? — the parents of these kids never ask where's the actionability. To them, it's incredibly empowering. That's the start of their mission, and then they come together, they bring together groups like the FOXG1 group that you mentioned, and they are focused on a cure. There is no one with more energy and motivation and focus than a rare disease parent.
Dr. Chan: Totally. The parents are so powerful and often selfless. Because oftentimes this is not going to help their child. And it's about making it better for everyone else.
Dr. Ashley: Because it can sometimes be too late by the time their action has led to big changes. But I think the rare disease community — we're so grateful that you in particular, and obviously CZI itself, backed them in this unusual way, which was to combine support for bringing the parents together and the families together with connecting them to the right scientists, because that's the magic. The families on their own can support each other, but the cures are going to come from the connection to science. And science has really been obviously at the center of everything that you've done and the huge ambition that you have. I wanted to jump up to some of the newer things, because you've announced some really exciting new things just this year. I was lucky enough to be at your launch for Biohub and with some of the new focus. You've talked about the virtual cell, which may be the hottest thing in biology right now, and there are many groups focused on it. Give us a sense of the scale of that project and why you think that could be the answer to many of the questions that we're trying to answer in basic science and translational medicine.
Dr. Chan: First of all, the virtual cell is super exciting, and there's not even a true consensus of what we're all talking about. So maybe I'll say what it means for us at CZI. We want to be able to build a virtual model that helps us understand the underlying biology that powers the human cell. We want to have a computational model that allows us to quickly and more cheaply look at cause and effect in the cell — look at various disease states, perturbations, how various genes contribute to changes within the cell that may later cause downstream disease. That work is super exciting. I think for medicine in general — and you work in precision medicine, we call it frontier medicine, I think they're very similar things — to cure, prevent, and manage all disease. That is a big mandate that we don't think we're going to achieve alone. And the reason why we think we have a good chance at it is because at CZI, we are fully committed to building tools that make every other scientist faster, more effective, and more efficient at their work and research. If we can speed everyone up, then science just moves faster and we can have new discoveries, new knowledge that impacts patients' lives.
Dr. Ashley: And to clarify for everyone, maybe some who hadn't realized — that is your mission, to cure and prevent all disease?
Dr. Chan: Well, we dropped "manage." We're just going to cure.
Dr. Ashley: Okay, that seems fine too. That's not a small ambition. We love that ambition in this audience.
Dr. Chan: It was very — I will tell you, most reasonable scientists could not look at us in the face with a straight face and say yeah, that sounds like a good idea, 10 years ago. But we've gotten lucky because right now we're at a moment where I think it actually is feasible and possible. And the virtual cell model is going to play a large role in this. Right now, a lot of basic science is done through someone having a clever idea, happy accidents, years of dedicated trial and error. And what we're hoping to do with the virtual cell model is actually de-risk a lot of that — have a computational model that allows scientists to test a lot of ideas, especially riskier ideas, and then say, okay, these are the few that we think are most important and highest yield. And then we'll test it in the wet lab or in a model or whatnot. But what is the holy grail for me is developing a model that allows us to understand an individual's biology. You talked about the diagnostic journey, which is terrible, and you've walked through the diagnostic journey with patients many times. We have fewer answers than questions when we're walking patients through that journey. And it's actually sometimes a relief and lucky when someone gets a rare disease diagnosis. More often than not, you say, I don't know, we looked at your whole genome and we think you have these three unusual things and we don't really know what that means and we don't know what risk profile to put you into. That is such a tricky part of the patient journey and your journey walking through it with them as the physician. I want us to be able to have virtual models that look at an individual's biology and say, based on these three mutations, we see that it impacts this protein downstream, and that protein is integral in this process. And we can look at the risk profile, and we can also look at treatment, and we can predict natural history for something that we have no word for right now — we just have a constellation of symptoms and a few genes that we don't know how they actually impact. That model will truly get us to frontier biology, precision medicine in a way that allows us to treat the individual that walks into the clinic and really close the gap between basic science, clinical research, and treatment. That is super exciting.
Dr. Ashley: No, I mean, clearly the cell is the individual unit of biology. It's the singular unit, and we have gotten better and better understanding. We have single cell genomics, single cell proteomics. We're increasingly able to measure things at a single cell resolution. But I think the scope and ambition of modeling an entire cell — especially a human cell; I mean, perhaps some recent papers are starting to get close with much simpler cells — but to model an entire cell means multi-dimensional, and then I think part of it as well — and I think you've embraced this already — is also the perturbations. Because it's one thing thinking about a quiescent cell just sitting there and trying to predict its gene expression or something similar, but really what you want, as you've been talking, is to be able to predict what happens when you mutate this particular gene in this particular way. And that's another scale again.
Dr. Chan: And then there's the spatial aspect, right? Things aren't just soup within a cell — you're looking at how things are arranged within the structure of an individual cell. The really exciting thing is that maybe 24 months ago we were asking, are the models powerful enough? Is there enough compute? We spent a lot of time building up a very robust GPU cluster to allow us to do this type of AI research. But right now the key constraint is data and looking at different data formats. There is great data in biology, but most of the data available — whether it's single cell or spatial work or protein work — is often in proprietary data sets, and often configured to answer specific questions that someone had gone into the work looking to answer, which is reasonable. But what we really need in biology right now are foundational data sets that are open and standardized and that everyone has access to in order to do their research and build upon. And we're very focused on that at the Biohub. We've had some lucky history. You know, 10 years ago when people told us we were stamp collecting, we started doing single cell and we built out the Human Cell Atlas Tabula Sapiens right here at Stanford. And then we built a data set called CellxGene, which is one of the largest single cell data sets available. And the incredible thing about those data sets is that they were built for general use and general knowledge. We learned that no one group is going to build all these data sets. A very happy accident happened with CellxGene. We were trying to get a single cell data set off the ground and we realized that annotating the data was really hard. So CellxGene was not originally meant to be a very large data set. It was an annotation tool that allowed researchers to identify the cells that they wanted to look at. This little tool became very useful for scientists, and because they were all using the same tool, they happened to use the same data format, and then we had data that was building upon itself, and it became useful, and people started giving their data back. We seeded the original data — we paid for and seeded the original data — and now about three quarters of the data in CellxGene is not something that we put into the world ourselves. People are just building this data set together. And because it's standardized, it is incredibly useful and is used by folks across academia and industry. And so we're following that playbook. We are looking to work with academic partners and industry partners to build the next data sets in our billion cell project. And we are very careful to look across species, across ancestry and human data, age, and disease states. The key will be to get multiple parties to be excited and willing to put it in the same data format for us to be able to build a foundational asset for everyone to be able to use.
Dr. Ashley: So let me just be clear — you just said a billion cells.
Dr. Chan: I know, it got big real fast.
Dr. Ashley: We were talking about one, and clearly, as you mentioned, over the last decade you and others — with a very big focus for CZI on single cell sequencing, as you mentioned the Human Cell Atlas and all that work, the mouse one as well and multiple organisms — that was in the range of hundreds of millions of cells. But you've raised the bar a little bit more.
Dr. Chan: The technique has been established; it's really about getting the right folks involved, doing the science, and bringing it back together. We're not limited by the number of cells anymore. It's really about whether we have the right views on the data. The world moves so fast — it's so hopeful.
Dr. Ashley: I love also that you mentioned the spatial element, because you've also had this interest in imaging. And I feel like, talking about perturbing the cell, time is a kind of lost dimension. We stop cells or we kill them and we measure them. Whereas in life, time is part of our lives, and certainly we go to see doctors at different points in time. So ultimately, if we go all the way to patients, time is an inevitable dimension. But with cells we haven't traditionally thought about the spatial element. You have unique tools from your imaging Biohub that —
Dr. Chan: I hope no one here ever needs to become a cryo-EM slide. It's much better to be dynamic, warm, and alive. Both have cryo abilities and our electron microscopes, and we're very proud of the work we've done there. It has been pretty high ROI investing in a few things like a laser phase plate in partnership with a group up in Berkeley to really improve our contrast and resolution. So we can see down to the atomic level of what's happening in various cells and preparations. That's really cool. We can actually say, with this mutation, this protein changes in this configuration, and we can see it clearly with this imaging ability. Next is going to be looking at video — we want video, not static slides. So what are the different labeling methods, without labeling, that allow us to look at living, dynamic cells at a resolution that helps us understand how the gene connects to the protein, connects to the living cell.
Dr. Ashley: No, it's amazing. And it's remarkable how often the scientists who do that kind of work aren't really talking to the scientists who do the other work. So being able to bring them together — that convening power becomes really important.
Dr. Chan: It's been awesome, and that's actually the whole premise of our research groups in the Biohubs. We put out a grand challenge — whether it's tissue engineering or cell engineering or single cell work — and we say, this is what we want to do, this is our big pie in the sky, do you want to do this with us? We invite folks from whatever seat they may come from — AI researchers, specialists in the physics of lasers, physicians, engineers — coming together and building together as a team. These multidisciplinary teams are able to do the work towards our grand challenges, answer these questions while building important data sets in partnership. We have nine universities — Stanford is one of them — that partner with us deeply. And at the same time we build something that's open and available for everyone far beyond the nine universities to build upon. That's our big goal.
Dr. Ashley: It's amazing. One of the things that happens when you bring people from very different parts of science together is they have slightly different views on how long things are going to take. As you were relaunching Biohub for the next 10 years, the AI people were on one side saying, biology? We can solve that in two or three years.
Dr. Chan: Maybe 18 months.
Dr. Ashley: And the biologists are on the other side going, what? We've been doing this for decades. How do you parse that? Who do you believe? How do you resolve that? Or do you just throw them together and see what happens?
Dr. Chan: Well, this is not a new problem for us, because we have always had a mix of very different people coming together. Cori Bargmann, our founding scientist leading our work at CZI, joined us 10 years ago. I adore her, love her. And she — we wrote down one day in a team meeting, we have to sequence our priorities. And she writes down "sequence the priorities" and then looks around and goes, "Where will we get the samples from?" And I was like, that's not what we're talking about, we just need to write down what we want to do, Cori. So we've always had the excitement and the additional work of getting people from different trainings and different perspectives to come together. Who do I believe in this? Well, I believe no one. Like a healthy skeptic. And the cool part is having people say, well, this is why I think it's impossible. My favorites are the people who think it's impossible, because the people who think everything is possible just haven't gotten down far enough. But the people who think things are impossible — you ask them to enumerate why they think it's impossible, what are the blockers — and then there are people in the room who can say, oh, I actually from a different experience know how this works. And that's when the problem solving, or the resource allocating, and figuring out how to solve the problems together happens. I would say science already looks different. Therapeutics are being identified and built using AI right now, and right now it's anecdotal — you're like, oh, this cool thing happened. But I think maybe in five years we are systematically working in a new way, and the breakthroughs aren't anecdotal. We're really looking at marching through a deep understanding of human biology and helping patients on a regular basis.
Dr. Ashley: I was wondering a little if Mark was on one side of that and you were on the other, and you were balancing the two.
Dr. Chan: The Mark and Priscilla collaboration — what does that look like? A little bit. Where you see Mark is really through the engineer's perspective — the idea of doing single cell transcriptomics and building a virtual cell model where you understand how the system works. That very much appeals to him, because he wants to be able to understand and instrument the system, understand where the bugs are, look at cause and effect, and tinker and say, here's the system, if we understand something at the fundamental level then we can understand human biology. For me, it's always the question of, are we answering important questions for patients and for clinical use? That's the bit where I'm always asking, is this interesting? Tell me why building a series of mirrors around the cryo-EM is going to help patient care later down the line. And I think the work is so exciting because we have a strategy that really closes the gap between the physics in electron microscopy and the impact in a patient. Bringing together basic science and clinical work side by side — that is the future of medicine.
Dr. Ashley: That's — we couldn't agree more with that. And having computation next to — and by the way, you mentioned your cluster — it's the largest nonprofit GPU cluster basically outside of tech. It's the largest GPU cluster.
Dr. Chan: Yes.
Dr. Ashley: So you're ready to do this. But having that next door to these wet labs, as you're talking about, and the cryo-EM, and then having on top of that your philosophy of what does this mean for a patient — that seems like the north star of what you're trying to do.
Dr. Chan: It's an incredibly exciting time for biology right now.
Dr. Ashley: We talked about cells. You also have a project called the virtual immune system, which is really interesting. I think some of the researchers are thinking about essentially immune cells floating around the body, surveying and looking for disease, extinguishing disease wherever they find it. I do not underestimate immune cells. T cells are exquisite. We all have our favorite cell.
Dr. Chan: What's yours?
Dr. Ashley: Maybe an NK cell. Oh, I have a friend who does a lot of work on that. She's kind of persuaded me over the years. What about you?
Dr. Chan: Oh, what is my favorite? I have to say a heart cell, actually. I thought you were going to say heart cell.
Dr. Ashley: I should have said cardiomyocyte. So that's obviously the answer. Just forget I said that —
Dr. Chan: I thought your favorite cell is the cardiomyocyte.
Dr. Ashley: Let me rephrase my answer.
Dr. Chan: I don't know if I have a favorite cell. Maybe my favorite cell doesn't exist yet.
Dr. Ashley: That's a good answer. That is a very good answer.
Dr. Chan: So you're talking about our work at our New York BioHealth where we're doing cellular engineering. Our goal is to be able to build an underlying platform — and I use "technology platform," not the software kind — where we can help cells be targeted to a specific destination, read out a certain state that we're looking for, encode it back into the cell's own DNA, and lyse itself, so we can pull out the cell-free DNA and read out the data that it has gathered within the human body. And then the next step will be, can you get it to take an action? That would be incredibly cool. The example I like to give is about the heart: can you use this technology to send a cell to read the plaque burden of a coronary artery, encode the data, lyse, and read it out — so that to get that information you're injecting cells into a patient and then drawing blood and sequencing it? Obviously that is still science fiction, hopefully not for long, because we have a lot of unanswered questions about the inflammation that you would trigger, what would happen, can we actually read with fidelity. But that's the type of technology and platform we think about building, because that's a general purpose technique. Then you can do a lot of other things — you can go into very privileged places in the human body and read out data that's useful for us to know, in a minimally invasive way. But that's where partnership with the community comes in. We see ourselves as building the platform and technology, in partnership with others. We need scientists, physician-scientists, to be thinking about the applications of that technology and how they would use it in their daily work.
Dr. Ashley: That creative process — you used the word "science fiction." I was going to use that myself, because it seems so futuristic. But I've been thinking about this, even just over the last few days, about how so much of the world we look around us right now, particularly where AI is concerned, looks like science fiction from just a few years ago. If you do your own version of science fiction and look forward 10 or 15 years — and we were talking about the movie in theaters right now, Project Hail Mary, which sounds like it's really excellent.
Dr. Chan: It's so good.
Dr. Ashley: Yeah.
Dr. Chan: Rocky — that's the name of the alien, right? And they name the planet — I'm not going to say anything. No spoilers.
Dr. Ashley: But it's a good movie. Science fiction is creative people, many of whom understand science, thinking about what the future could be like. In a way, that's kind of what you're doing, and also trying to help build it. If you have succeeded and you recently looked forward to the next 10 years, given the tools we have today as well as the foundation you built over the last 10 years — how do you see medicine having changed? Because we are the ones practicing, and we all love science, and lots of us have labs as well, but we are hoping, as you are I think with your vision, that we will change how we do medicine.
Dr. Chan: I talked a bit about being able to have that virtual model of individuals. I think that is super powerful — to be able to look at an individual's genetics and actually translate those genetics in a meaningful way to clinical practice. I hope very much that we will start looking at individuals that way in 10 to 15 years from now. But also — this is Stanford University, and God bless physician-scientists, especially those who've gotten an MD/PhD, that is so long. I think if I stayed in medicine, I would still be finishing if I were on that route.
Dr. Ashley: And you know that you're always welcome back.
Dr. Chan: That is actually my secret retirement plan. I wanted to be a pediatric cardiologist.
Dr. Ashley: Oh, well now we're talking.
Dr. Chan: And I actually did a rotation here. It was one of those things where maybe Stanford had some advantages over UCSF. So I spent some time here. It was really something I wanted to do and I haven't closed the door on that. But that's a separate conversation — kids have to go to college and I'll be the oldest fellow that's ever been. When you're ready, we know some people. I keep my board status active because I plan on this. But the part that I think will hopefully change — the way physicians can evolve their practice and the way we evolve training — is that everyone actually becomes a physician-scientist. It's not some distant far away land where papers get published and then 20 years down the line we get a drug. I don't want us to operate that way anymore. It is slow and ineffective. We have better tools than that. How can we actually close the gap so that everyone is thinking deeply about the biology, the protein structure, the mutations that are happening within individuals and the factors that put them at risk, and use that in our clinical practice every day? That is how we get to the individualized care that we're all looking for. And that is how we can very quickly translate what's being learned in a lab to impact patients. That has to look different in 10 years.
Dr. Chan: Thank you, Dr. Ashley, as always.
Dr. Ashley: And please join me in thanking Dr. Priscilla Chan. Thank you everyone.