Scientist Stories: Euan Ashley, Outstanding Problems in Bioengineering

Good evening, everyone, and welcome. Thank you so much for tuning in tonight. My name is Kate Bruns, and on behalf of Harvard Book Store, the Harvard University Division of Science and the Harvard Library, I am so pleased to welcome you to this evening's event with Euan Angus Ashley discussing his new book, "The Genome Odyssey: Medical Mysteries and the Incredible Quest to Solve Them", in conversation tonight with Siddhartha Mukherjee. Tonight's lecture is part of our Harvard Science Book Talk series, which brings the authors of recently published science-related literature to our Cambridge community and now beyond. Coming up this month, we will be hosting events with Nobel Prize-winning biologist Paul Nurse and Nobel Prize-winning physicist Steven Weinberg. To learn more about the Science Book Talks, you can visit the webpage harvard.com/science or sign up for the Book Store's email newsletter at harvard.com. We also have a YouTube page where you can view previous talks you might have missed. I will post links in the Zoom chat for those in just a few minutes. This evening's talk is going to conclude with some time for your questions. If you would like to ask our speakers anything, please go to the Q&A chat at the bottom of the screen where you can submit a question. We're going to get through as many as time allows for this evening. I would also just like to say a tremendous thank you for your patronage during these strange virtual times. Your support makes this author series possible and really does ensure the future of a landmark independent bookstore. So thank you to our partners at Harvard University, and thank you to all of you for tuning in and showing up for our authors, indie bookselling, and especially for science. And finally, as you may have experienced in virtual gatherings, technical issues may arise, and if they do, I'm going to do my best to resolve them quickly. So thank you for your patience and understanding. And now I am honored to introduce tonight's speakers. Professor of Medicine and Genetics and Associate Dean at Stanford University, Ewan Angus Ashley is renowned for his contributions to genomic and precision medicine. On the Stanford faculty, Dr. Ashley led the team that carried out the first clinical interpretation of a human genome. Today, he directs the Clinical Genome Program and the Center for Inherited Cardiovascular Disease at Stanford Hospital, where genome sequencing is routinely used in patient diagnoses now. His award-winning work has been recognized by the Obama White House and the National Institutes of Health and has earned him the Medal of Honor from the American Heart Association. Joining him in conversation tonight is celebrated writer, oncologist, and researcher Siddhartha Mukherjee. He is the author most recently of "The Gene: An Intimate History" recognized as one of the most influential books of 2016, and he is the author of "The Emperor of All Maladies: A Biography of Cancer," winner of the 2011 Pulitzer Prize for General Nonfiction. Tonight they'll be discussing Dr. Ashley's brand-new book, "The Genome Odyssey," celebrated as, quote, "destined to be a landmark narrative in the canon of modern science." Of the book, Dr. Mukherjee writes, Dr. Ashley, one of the pioneers of gene sequencing technologies, writes with authority, elegance, and simplicity to enable an in-depth understanding of the most exciting scientific developments of our times. Every curious reader must read this book. We are so honored to host them for this conversation tonight, so without further ado, Dr. Ashley and Dr. Mukherjee, the digital podium is yours.

Uh, well, hello and welcome. Um, welcome, Ewan, and congratulations on your book.

Uh, thanks so much.

So, uh, let's start with the, um, you know, just a little bit of introduction. Um, 1990 was the official date, uh, when the Human Genome Project was announced. And that brings us now to about 30 years, give or take, and that's been— those have been landmark 30 years. So the first human genome was sequenced at the cost of about $3 or $4 billion. Now you can sequence human genomes, not— I would say the active parts of the human genome, the exome, at for about, I don't know, $1,000, $600, $700, something like that. So it's been a, you know, a kind of Moore's Law kind of shift in that moment. So we are also standing, so 30 years from the origin of the Human Genome Project, and of course I can't help but remark that we're also standing in the midst of a pandemic. Where genomics, sequencing of viral genomes, sequencing of viral strains has led to a completely new understanding of infectious disease. So somehow, you know, both these are kind of landmark moments and have really changed the way we think about the world, the way we think about ourselves, and the way we think about our futures. But let me begin with you personally. Ewan, tell us what prompted you to write this book. What was the— what was the inspiration and why did you choose to write this book?

Well, thanks, first of all, for being here and for being in conversation with me today. Yeah, I, you know, I came to science initially because of books, actually, not so much because of teachers. In the end, I had great science teachers, but at high school, I was reading books by people like Richard Dawkins and Steve Pinker and Jared Diamond and those sorts of people really were the ones that brought me into science, and so I'd always admired that. Plus, as I was in medical school, I really enjoyed Oliver Sacks and his work— just tremendous writing. And so at some level, I knew that I might want to try and write one day, and I certainly spent my life as a doctor inspired by these stories of these incredible patients who go through just amazing things, often with rare diseases, where they're on a long medical journey, an odyssey, for sometimes years going from doctor to doctor to doctor, and that's just— was a— is a very inspiring thing to watch. And so part of me really just wanted to tell that story, um, but also inspired by books like yours, you know. I also understand how much of a narrative science is, and that science is really a team sport. We work together and we spend days and days going after a problem, weeks turn to months, eventually the problem cracks and moves forward, and it moves forward in interesting ways. And so my thought was that it might be interesting for others to read about these scientific narratives, the excitement of the chase, as it were, but to try and bring those to human life with the real stories of my patients and other patients that we came to know. So all of those things, I think, together were what eventually led me to start. But I'd say your own writing and Atul Gawande were also both very prominent in the last kind of decade, of course, for all of us interested in really lucid medical and scientific narratives, and so very inspirational to be able to pick those up and then just to try in my own modest way to contribute something small to the canon.

Well, it's not small, it's a big book, and tell us about the writing process for a doctor, and, you know, I'll certainly share my observations, but, you know, you play 3 roles really in this book, right? So on one hand you're the author, on the other hand you're a doctor, and on the third you're a scientist or a geneticist. So you play 3 roles. How— you know, just to contrast this with, for instance, Oliver Sacks, you know, Oliver, who I knew quite well, didn't have that third role because, you know, genetics was just being born. He would ask people, he would ask his patients questions about their family history, but the idea of the genome being a tool, almost a medical tool, a kind of I would call it— I would describe it as a kind of stethoscope into your— into what lies deeply inside you, didn't exist, you know, and so many cases one wonders if those cases would have been solved using that stethoscope which was missing in medicine. So tell us about how you resolve these three —roles, you know, doctor, scientist, and author, and tell us a little bit about writing— the writing as a process. So, you know, the book, you know, goes— is a compendium of stories, they're kind of whodunits, mysteries, and in every case you use the genome as a kind of, as I said, as a kind of medical stethoscope, which would not have been available, you know, 10 or 15 years ago, but is available now. How did you— how did you go about writing this? Did you write at night? Do you write when the stories came up? Did you remember them? What was the way— how did you choose these in particular?

How did you do it? Yeah, well, no, so, so many things there. I mean, yeah, definitely think of the genome as a molecular stethoscope. Years ago, first of all, and I think it would have been amazing if Oliver had access to that. I would only imagine the stories he might have told with this tool at his fingertips as well. Yeah, for me, I think it started very practically, like for many of us, when you have downtime and just a little bit of space, which in a former world that we can barely remember now was when we got on planes. And so I distinctly remember the first time when I started writing, the stories had been in my head for some time. Some patients are, they're so memorable. That you carry these stories around with you for years in many cases. And, you know, I told them to a variety of people at times, obviously in an anonymous fashion. So the idea of potentially going back to revisit with those patients and really trying to dig into their memories and mine, did it with my longtime colleague and nurse partner in the clinic. And we sort of dug through our own memories to try and bring back the details. And then, yeah, I think with a young family, a lot of writing happens either on a plane or early in the morning or late at night, or sometimes you send the family off on a hike or something else and get a few hours to just try to find time. You know, I think that the inspiration of the story and the drive to write it down is where it begins. Of course, as I've learned, it's a long way between just getting the words down on the page and trying to put them into a form that hopefully will will be ingestible by people and enjoyable by people as well. But yeah, juggling those three things is not easy. I think that my current work, I'm a cardiologist at Stanford, and so I have a clinic every week, and then 3 or 4 times a year, I also work in the critical care unit at Stanford. So, you know, there's plenty of, if you like, there's plenty of material for patient stories. So it's almost like one day in the life on a coronary care unit produces 3 or 4 human stories that would be worth telling. And as you mentioned, in the middle of the pandemic, it's almost like those are more acute. Those human stories are even closer up and made all the more extreme perhaps, because often the families are not there until the very last minute in many cases, as has been the case over this last year.

And let's talk a little bit about— let's shift gears and talk a little bit about the human genome and its odyssey. And tell us about sort of what have we learned in the, you know, these 30 years that have passed, or, you know, when the human genome was sequenced, as you know, it was a big event announced in the White House jointly by Francis Collins and by Craig Venter. The president was there and it was, you you know, it was considered a kind of a public-private victory, although as you know very well, that victory was tinged by a public-private competition that preceded it. But in the end, a kind of compromise was reached in which the, you know, the public and private attempts to sequence the human genome were announced jointly and out came a string of 3 billion letters in A, C, T, and G, and if you were a layperson, you would look at that string of letters and say, gosh, you know, 3 billion letters of a human being's genome, what do I make of it? I mean, you know, it's like, it's like a book in which you— a language you don't understand. What has happened? Highlight for us what has happened in the— and take your time to answer this— what have we learned about the human genome since the publication of that ACTG list, or the ACTG code, of the first human being?

Well, I think that you lay it out very nicely as a 30-year odyssey of its own, like the human genome, in fact, and there are— and I dip into that at various points in the book. So There's one story where I dip into the origins of where the human reference sequence came from, like the people who volunteered anonymously to give their DNA to be mixed into a little pot that would then be spat out, as you said, with the— as these letters. It was originally going to be 10 people, and it turned out in the end to be basically 2. So they— and they came from Buffalo in New York. And so I dig a little bit into that history, which is, as you said, it's like the early '90s. And there's some interesting little elements that I certainly wasn't aware of, but I uncovered as a result of that. And then the pivotal moment, as you said, about 10 years later, $3 billion, 10 years, 10 countries, and we have a draft sequence, which was just an incredible moment. But I think then everybody thought, well, suddenly now there's going to be a big change. And in fact, much of the next 10 years was really trying to work out do a little bit more finishing, work out what it meant, start to— you know, the number of genes that were estimated at the time the genome draft sequence came out was something like 40,000. And if you watch the literature, every few years it drops to where it currently has sat pretty much stable at 20,000, just under. So we spent about 10 years just really learning about the anatomy of the genome. And I think that the major part of this book, though, is really these last 10 years, which is I think when the genome finally became relevant, not just for science and discovery the fundamental nature of just knowing the genome and understanding the letters and how much was made up of genes and how much was the parts in between the genes, but making that really relevant to human medical care. And so I talk about at the very beginning of the book walking into my colleague Steve Quake's office, and he'd been the fifth person in the world— he sequenced his own genome with a technology he invented in his lab for about $40,000, which Now, as you said, now you can do it for 500, so 40,000 sounds like a lot, but the one before it was 200,000 and had taken months and months and involved many different teams. So it really was a leap forward, and the beginning of the book is really when I met him, and I was in his office, and he started showing me his genome on the screen, which to me was just mind-blowing because we were at that time used to sending genetic tests that had maybe 3 genes or maybe 8 genes on them, and it took 3 months to come back and cost $5,000, and here it was somebody you know, on my— on campus, just down from my office, like scrolling through his whole genome. And that was just, you know, an incredible moment for me. And then when we realized that he essentially had a pretty significant history of heart disease in his family, including his cousin's son who died suddenly with no known cause, that was the moment things switched from this is a scientific, you know, incredible scientific moment to suddenly a doctor-patient thing, because it was clear that even without his genome, he needed to come to a clinic and be screened given all this disease in his family. But as I was essentially inviting him to my clinic, it became clear that he was also about to be the first person to walk into a clinic basically with his genome, and that we had no idea what we were going to do with it. And so that was really the revelation that led us— this is 2009— to start to work on some computer methods to really understand the genome in the context of medicine, thinking like if you're a GP with 12 minutes and you're now in a world where patients have their genomes, How do you integrate that data in a way that's useful, in a way that can be ingested, in a way that really helps personalize the care of that patient? And so that's the world we've— in this last decade, the world we've kind of stepped along. And I think, you know, at the end of the book, I end up saying that I think after 2009, we were all like naive thinking, well, in a couple of years, everyone will have their genome sequenced. It really, for a moment, felt a bit like that, because in the weeks and months after we published the first paper, lots of people with access to funds were able to get their genome sequenced, and they were calling us up and saying, "Can you help us? Can you help us?" But really what happened over the decade it took to get to this point was a maturing of that technology. We started to get encyclopedias of genes together, connections between genes and diseases. We began to understand the harder parts of the genome, knew what we were getting and what we were missing, where the errors were. And so that's been a really big process, a lot of it computational, but also the technology for sequencing has improved. And that's taken us to the point where really it's almost routine now for rare disease to be sequencing the genome, at least in the US, not everywhere yet, but it's essentially every day we're ordering genome sequencing for rare disease. And then the other two parts of it are when— on rare disease, we just had Rare Disease Day at the end of February. February 28th or 29th every year is Rare Disease Day. And we always get a chance to remind people that 1 in 10 people has a rare disease, which seems mind-blowing every time I think about it, which makes rare disease as a category as common as diabetes. And this genome can be applied to everyone with rare disease. But the question then that comes up is, what about the other 9 out of 10 people? When is the genome relevant for much more broadly for cancer, for cardiovascular disease, the really— the big killers of our society? And I think, interestingly, we're just about at that moment now. We can start to sequence at much lower depth instead of covering the genome 40 times, we can cover it just once or twice and use that information to start to integrate a score that would help us— allow us to make a much more precise prediction of whether someone will have a heart attack, or to put that together with certain well-known genes that can predispose towards cancer if they're mutated, to give an overall sense of risk so that we know what kind of screening to apply. And that day is just coming. So I think places like Stanford and in Boston, those kind of technologies are just becoming available really this year, but will start to spread, I think, to the rest of the world coming on now. So it's been a pretty amazing 20 or 30 years.

So as we— I mean, just to take a very broad look at it, we started moving from, you know, the great pioneer Victor McCusick and others pioneered among the first rare disease clinics in, in the world, and this was at Hopkins, and of course they didn't have access to any of these technologies. This was even long before the Human Genome Project, but now we're moving— then there was a phase in which you could use the genome to study, to find rare diseases or the mutations that cause rare diseases, And now you're using the genome to find genes, and often there are hundreds, there often may even be 1,000 genes, gene variants that cause common diseases, like, you know, a risk, a higher risk, elevated risk for breast cancer, an elevated risk for ovarian cancer, an elevated risk for cardiac disease, et cetera, et cetera. One question that has, has often occurred to me, um, not as a doctor— well, I should say as a humanist more than anything else— is that as we enter this kind of new phase of our lives, um, where the genome becomes a predictor of future disease, um, what do you think that does to the minds of people and to, um, and to society in general? I'm thinking of this idea of living under the sort of the anxious umbrella of the genome, and that anxious umbrella is constant. In other words, you know, you're born and you get your genome sequenced and out pops a series of diseases, common diseases. I'll give you two: breast cancer and the risk of dying of of a heart attack, that now you're, you're still a babe in the arms, and yet, yet you have the possibility of knowing a score, as you call it, of whether you will get this disease in your future life. Talk to us about how to manage or navigate that world and how you might want to navigate that world for your own children.

Yeah, I think it's a revelation that was a big part of our thinking when we started to do this, because I think at some level, understanding the genome was some— something about telling you something about what it meant to be human, but also something about predicting your future. And when we first were sequencing genomes, I don't think we were very good at predicting the future. But actually, we now have much bigger studies on which to base our scores. And so we now are in the position to apply these scores in a way that's helpful. But I think that while there's clearly an element of what you describe, and for some people they will find that they have variants in their genome of large effect that really are somewhat deterministic of what will— something really will likely happen in their lifetime that relates to that finding. Most of these changes that we talk about, and it can even be up to millions now spread across the genome that get incorporated into a score, what they do is change the probability. They move the needle a little bit. So if you go to your GP and talk about your risk for heart attack, They'll ask you, of course, about— they'll talk about cholesterol. You'll have it measured. They'll measure your blood pressure, talk about diabetes, your lifestyle. And all of that will get filtered into a score, which is online and available to anyone, and will come up with a 10-year risk of a heart attack. And if it's above 7%, there'll be a lot more focus on giving a cholesterol drug, for example. The thing is that we've known for years that for heart disease, half of it is nature, half is nurture, just about. About those kind of numbers. It's about 50-50. And right now, actually, the score we use in the US doesn't take any account of the nature part at all. It's all nurture. It's all about the environment, which is really important because those are the things you can change. But if you really want to know what your risk is, and that's the basis on which you're making the decision, you want to use the genomic information too. So it's a little bit more like that number becomes more precise. Instead of it being 5% with an 80% probability, or probably a bit lower, it's now 7%, but with a 90% probability. That's the way to think about the impact of the genome on these kinds of diseases. It's more of a probabilistic— it moves the needle a little bit. But that's not to say that, as you say, if we start to roll out full genome sequencing even early in life, there will be people who have a future that is— well, their assessment of the future will be changed as a result of knowing that information. And that's something— in the early days, we actually included— the first person I called, actually, after we realized that Steve Quake was going to come to the clinic was an ethicist, Hank Greely, who you may know, who's written very elegantly and recently about CRISPR. He has a book about that, and in the past about genome sequencing and these kinds of questions. So for example, something we hadn't thought a lot about in our clinic was if we do very targeted sequencing, then it's really mostly focused on that individual. But for Steve, for example, we were looking across his whole genome, including at genes that would cause potentially breast cancer and might be relevant for his daughter. So there are all those kind of questions that also relate to the family aspect of genetics, where we share not just our environment, but we share genes with our family members. And it's anything that we learn as a result of— or anything I learn as a result of sequencing my genome is relevant for my brother, for my sister, for my parents, for my kids.

And how do you— I mean, just to push you a little bit on that point, how do you negotiate that? Or what's— do you have a sense of how you negotiate that? So I mean, it will soon come to a point of time when that information, which is of course extraordinarily private to you because it's about your genome, becomes relevant to your sister. I've had this case many times in which I've had a person who has had a mutation in, let's say, the BRCA1 gene, and I've had to have a conversation with them if it's a woman or a man, you know, do you want to talk about this with your daughters or with your son? But with the daughters, it's very relevant because obviously their risk is more immediate. The son, of course, also has a risk, but that risk is passed— has a risk himself, but the risk is then passed across the next generation. How are you negotiating this in, you know, you're in your clinic, you're out to sequence, you know, some gene that causes cardiomyopathy and kaboom, you find that someone is BRCA1 positive. What do you do? How does one negotiate that information?

Yeah, well, the first and most important thing we found early on was to make sure that a genetic counselor was part of every single interaction that we had. And so early on, we actually hired someone straight out of genetic counseling school who was just a perfect fit for this kind of position, because those are not easy conversations to have. And the whole structure of understanding this kind of information at scale that's then relevant throughout a family is really a different skill set, even from the one that had been traditionally taught. And now, of course, everyone who comes out of these schools understands variant selection and variant curation and understands how to deal with genome-level data. But it doesn't make the underlying question any simpler. Simpler because what we're dealing with at the end of the day is families and family dynamics. And in many ways, these are issues that our family practice physicians have even, even more experience of than us in super subspecialized care. My father was a GP and growing up as a little kid, I used to go on house calls with him. Basically, that was when I— before I ever knew I wanted to do science or or do computers, which is what drew me into the data side of all this. I knew I wanted to be a doctor. It was partly because of going with him on house calls. And we'd go into people's homes, and there would usually be at least 2, sometimes 3 generations of people there. And so I think that it's not new to medicine, the concept that there's relevance from one generation to the next, whether it's environmentally transmitted or genetically transmitted. But it does make for difficult conversations, especially when these things have to date been hidden, and one individual has decided to shine a bright light on it. I think that's what's new. And so we obviously have to talk about it in advance, encourage them, as you did with your patient, to talk to their family members potentially in advance, and then make a decision that you hope is the right one for the whole family. And families that come in such diverse shapes and sizes, that's part of the joy of the medicine that we get to do, is interacting with these families and really getting to know them all. And most families find their own way to make this work, but it's not It's not simple.

Um, uh, if you don't mind, I'll ask you to read a little bit from your book because I like, um, I really like it when authors read from their book. If it's okay with you, I'd like you to choose a passage that is interesting or a story or— and then, you know, once you're done with that, tell us about why that story is in the book and how you thought about it.

Yeah, yeah, thanks for asking. And this is actually a relatively recent decision, but I think this little— this story touches on a few of the points that we've been making in conversation and has some relevance for your part of the world as well. I'll start actually— before I give you the background, I'll start with the beginning of the story, and then I'll give a little background, and then I'll read a little bit more. Please do, yeah. This is from Chapter 15, "River of the Land of Pine Trees." It was the sound. It was really like an animal sound, like a moaning, and I was trying to talk to her, but there was no response. There are no words that can begin to describe what runs through the mind of a mother who has found her teenage daughter blue and lifeless, lying in a pool of her own spit. A few minutes before, Susan Graham's daughter Leilani had been a normal 13-year-old. Vibrant, fun, sporty, very much like her parents. Susan's husband Chris was focused on the rising tide of personal electronics in kids' lives and a generation growing up increasingly sedentary. Both mom and dad were active in coaching and refereeing soccer and basketball, invested in lead-by-example encouragement to be active. But there was also more direct encouragement. There was a system in the Graham household: if you want to watch TV, you also run. On this particular night, Leilani had been working out as usual on the treadmill in the garage. Chris was out at a meeting. Susan was also supposed to be out, but her plans had changed, so she was sitting at home on the sofa, the back up to the wall next to the garage. Susan heard the treadmill whirring as usual, then she heard another sound. It did not, she recalled, sound entirely human, but whatever it was, she was clear about one thing: it was not good. She sprinted to the garage and found Leilani collapsed unconscious on the treadmill. Guttural moaning emanated from her slowly, the writhing body. Susan stopped the treadmill, called 911, and waited. Leilani was breathing but not coming around. When the paramedics arrived, she was still unconscious. Could it be drugs? Could it be a seizure? It wasn't immediately obvious what the problem was, but at the hospital it soon became clear. An ultrasound of her heart revealed an abnormally thickened heart. Leilani had a life-threatening genetic heart disease of the heart muscle called hypertrophic cardiomyopathy, and that event was not a seizure but a cardiac arrest. So this opens the chapter and tells a little bit about the story of Leilani, and then over the course of the next few pages, we talk about the history of the disease hypertrophic cardiomyopathy, and its hist— and the history of bringing people back to life using electricity. But I think the part I wanted to read a little more of was relating to the genome because that's the center part of the book and the centerpiece of our conversation. And so I'll read just another couple of pages that relate to how we discovered the basis of this disease and then how we worked out in her case what the genetic cause was. The first clue in the hunt for the genetic cause of hypertrophic cardiomyopathy came from an unexpected place. Next to the sweeping lush green Parc Laurentien in southeast Quebec, a few miles from the American border, and about 20 miles south of the nearest large town, Sherbrooke, is the town of Kotikook. The name Kotikook is of Abenaki origin, meaning "river of the land of pine trees." In French, the area is known as Perle des Cantons de l'Est because of its natural beauty. But Kotikook became known for something less beguiling: the Kotikook curse that took many of its citizens early with strokes and sudden death. In the fall of 1957, two brothers aged 39 and 41, originally from this town, were admitted to the Royal Victoria Hospital in Montreal with strokes. Both were found to have enlarged hearts. Both also came under the care of Peter Paré, a 6-foot-4 lung specialist known for his gentle, humorous teaching style. Fascinated by the coincidence of strokes and enlarged hearts in the two brothers, Paré became convinced there must be a genetic explanation. So much so that he dedicated the next few years of his life to understanding the cause of the disease in the family. The brothers, he learned, were not alone. In total, he studied 27 family members, asking many to come to the hospital for clinical examinations, some of them several times. In July 1961, he published his seminal paper, Hereditary Cardiovascular Dysplasia: A Form of Familial Cardiomyopathy, in which he described 20 living affected family members as well as another 10 who had previously died, and even traced the origin of the disease back to one particular immigrant who came to Quebec in 1650. In fact, in the 4 years it took him to complete the work, 5 family members died suddenly. If ever there were a family waiting for the right genetic technology to narrow down the cause of a condition, it was this one. Genome-scale sequencing, of course, was a long way off. The Human Genome Project wouldn't get started for decades. But the linkage mapping technology was invented in 1980, and it could help. With this technology and a family like the one from Kortykook, the pieces were in place. And the seminal work describing the genetic underpinnings of hypertrophic cardiomyopathy was led by a husband and wife team, Christine and Jonathan Seidman of Harvard Medical School. Together with Bill McKenna in London, they approached the family first characterized by Dr. Parry and invited them to weekend reunions where they could learn about the research study and comment and consent to join. More than 100 family members signed up for what was to become a tour de force of genetic discovery. The linkage mapping method was labor-intensive but extremely powerful. So powerful, in fact, that the Sideman laboratory estimated odds of 2 billion to 1 that the gene responsible was in the identified narrow region. Given strong odds that this was the right corner of the genome in which to look, the team started fine mapping by sequencing the critical genes in this area until they found one gene the cardiac myosin heavy chain, part of the molecular motor of the heart, that looked to be responsible. The genetic variant was found in every single family member with the disease. The Cotygoode curse turned out to be one misspelled letter in the genome. Where there was supposed to be a G, there was an A.

Great, and of course it makes me think about the pioneering work done by Nancy Wexler before this, where they ultimately mapped the Huntington's disease gene using techniques— and of course, similar techniques, I should say— and then ultimately found the gene responsible for Huntington's disease, which was, if I remember correctly, the first time that someone had moved— this was a big team— but had moved from you know, from the, from a mysterious disease that affected many members of a family to the actual misspelling of the genome itself. I think we should move to questions if that's okay. And there's some of that already up, if I'll start asking them and then, you know, we can go back and forth a little bit, but let me move to the questions. The first question is from someone anonymous. It says, can you elaborate a little bit more about the ethical considerations regarding the genome? And I suspect that they want to ask you about some of the things that we've been talking about, you know, once you know, you know, what are the ethical considerations? But, you know, let me push that point even a little bit further. It is now possible to biopsy, to create a, to biopsy a tiny fetus, and you can biopsy that fetus and you can determine whether even before that fetus is implanted So this is called pre-implantation genetic diagnosis, and you can, you can get extraordinarily detailed information about that fetus, or that, I would say, that collection of cells, that human being, before you implant that into a womb. Can you envision a future in which pre-implantation genetic diagnosis for a disease like hypertrophic cardiomyopathy that runs in a family, or perhaps even more something more complicated like BRCA1, etc., play a role in our society and what happens to a society where things become relevant to PGD, preimplantation genetic diagnosis?

Yeah, I know these are really important and big, big questions, and thanks to whoever put the question up for asking. First of all, I would say that in relation to our early genomes— and we talk about this in a few chapters in the book— when you first realize you have this level of information available, you need a way of, first of all, just analyzing what's there. And so a lot of our work was how do we make computers understand, and how do we have them spit out the information we want and prioritize it? But then in that question of priority comes our first ethical concern, because some people might want to know everything. And the early patients that we had were geneticists themselves who quite literally wanted to know everything, but most people probably do not. And so we need— we needed to create a method where people could almost like turn, turn dials for the sorts of things that they would want to know. For example, if there was a disease that was hidden in their genome, for which there wasn't a treatment, a lot of people didn't want to know that. And so we had to be able to set up our tools so that that was an option. But if there was something that, especially if there was a fairly straightforward treatment that could be implemented, if they only knew they had it, but would otherwise cut short their life, they definitely wanted to know that. And so then trying to make decisions about how we quantify that was hard. And so then as you talk specifically about PGD, this preimplantation genetic diagnosis of embryos, which is a really powerful technique, and it's of course not for everyone, it's a very personal decision, but for those with severe genetic disease running in their family, combining essentially in vitro fertilization with the ability to do a genetic test of the individual embryos, and so the way this would work is that there would perhaps be, if you're lucky, as many as 4 to 8 embryos that would each be taken to the point of, as you said, being a ball of cells, and then one cell could safely be removed from each of the embryos and tested, and let's say you have a disease that causes early fatality, you, you can test those individual embryos and then only implant the embryos that are— that do not have the condition. As with most genetic diseases, there's a probability that it's passed on, and so that's, that's what preimplantation genetic diagnosis is, and it's, it's been around for, for more than 20 years, although the technology to do single-cell sequencing is relatively recent, but be able to amplify DNA to be able to do that test, that's been around for a little while. So we've had patients in our clinic over the last decade to go through that. I think the other part that you're, I think, referring to now is, now that we can sequence a whole genome, is somebody going to suggest that we do that at the embryo stage? And what kind of a difficult and dangerous world would that lead us into if now we can have scores that predict all sorts of things, including maybe even intelligence, whatever that means, and however anyone has decided to measure it. Most people would agree that's not something that would be ethical to choose embryos on the basis of. And we're at the moment, of course, in the middle of a revolution in understanding what CRISPR can do for us, as opposed to what it has done for millions of years for bacteria, um, and whether it would even be possible to either change and gene edit a particularly deleterious mutation in an embryo back to the regular non-mutated form, or potentially enhance in some way an embryo. And this, of course, has been greatly in the news over the course of the last 2 years, and even more so with other books that are out currently focused on the CRISPR revolution, and those stories are being revisited right now. So it's clearly an area where the ethicists need to be very deeply involved and closely related to everything that's happening.

A quick question before we go on to some longer questions. Richard Zug wants to know what happened to the girl on the treadmill?

Well, so I— yeah, so I'm tempted to say buy the book and find out. But seriously, she is doing really well. And I'm actually next month having the pleasure of doing a sit-down like this one that Syd has been generous enough to do here tonight with her. And so if you'd be interested to hear more about her story from her, she is an amazing patient advocate. I won't tell you any more about her medical history, although it is in the chapters in the book, for want of spoiling the story for others. But she's doing great, and if you'd like to see her in person, then there's an event in about a month's time, and you can hear some more. Where is it?

Just give us a little bit about where that event is and how does one find it?

Yeah, so it's hosted by Stanford Health Library, and so there is actually now— yeah, so you can go to Stanford's— Google Stanford Health Library, or if you go to If you can type my name, there's a website now connected to my name, ewanashley.com, that has a list of the events on there. Great.

Um, Douglas Peck, um, has been waiting patiently. He wants to know, great book, by the way. One of the things that impressed me was how collegial and cooperative other scientists were in working together on projects. This is so unlike other technology areas. Do you think this, that's true? And if so, why? So many companies struggle to build teams and the ones that you described seem to work so well together, even when virtual. Is that true, Ian, or is that made up?

No, it's definitely not made up. And I'm so grateful for the question, because actually, that was one of the things that really inspired me to try to write this down. I would sit in these rooms with these just unbelievably talented and smart people. And kind of, I mean, inside me, I'm the guy who grew up in the west of Scotland who still can't quite believe that he's here. And I'm not quite believing that I'm in the room. And I look around, and there's just these giant minds that are coming together with this very collaborative spirit. And that's not to say I'm not completely naive about the world. There's plenty of competition out there, and there's plenty of times when people aren't always acting in good faith. But I feel I've been really lucky in the people I've been able to work with. And when I look around, I see mostly that. I see mostly people just bringing very high energy to an idea and working together with other people. And maybe it's the sunshine on the West Coast or something. I know that it's also the case in other parts of the world, but I really appreciate the question because I really feel so excited to be in those teams and to be able to describe them. And I think that, I think it really is like that most of the time.

It really is. So a technical question, but maybe you can unpack the words for us a little bit as you as you answer it. Um, Ross Feldberg wants to know, can you comment on the proportion, um, uh, um, of single gene polymorphisms with strong effects versus multiple gene forms each with weak effect, and can we actually detect the latter?

Um, yeah, that's a great question, and actually it mirrors very nicely the conversation that we've been having where the world of whole genome genetics has moved from being focused entirely on rare disease— and rare diseases are almost exclusively caused by variants of— small variants of large effect. Sometimes they're big variants, but in general, they're of large effect. And so rare diseases almost by definition relate to that. But we all have— and usually it's a few million when we sequence a genome. We end up with 3.5 million, something like that, single nucleotides, single base pair, or single letter changes, most of which are of pretty small effect. And we've really just learned over the course of the last few years that the answer for that is to add them up and that probably most of them have some effect somewhere, but the effect is overall very small. And I think that overall our studies have not been large enough until quite recently to get our arms around those multiple variants of small effect. We were looking at, let's say, the top 50 of those variants and thinking that's what we have, but we didn't want to get it wrong, so we didn't really go much beyond that. But some interesting techniques from a few different places, including a couple of great groups out of Boston, have really looked at so-called genome-wide scores. So really looking at taking 5 million variants, each of which has a— most of which have a very small weighting, very small weighting, but a small number of which have a larger weighting. But taking the weightings of all of those together really does push up. Just to put it in real context, that score, sometimes called polygenic risk score, is individually more powerful than any individual risk factor for heart disease. So it's more powerful than asking you whether you smoke, more powerful than knowing your cholesterol, more powerful than knowing if you're diabetic. Individually, those scores are now good enough that they individually outperform those individual risk factors. Of course, if you add all of them up together, you get the best possible prediction. And so that's what increasingly we're doing, and that's what we're rolling that out to Stanford. There's a clinic at MGH doing that now as well. And I think that really is the future of bringing the nature and nurture aspects of disease together. Into the best predictor so that we can make the best preventive care.

And let's pause for a second there, because as we get more and more refined in defining these polygenic risk scores, which of course I work a little bit about within, in the context of cancer, you know, you're having to make difficult decisions. How much, how much of it is reversible, do you think? I mean, so just talk about cardiac disease. Can we— if you get a real polygenic risk score that shows you that you have a very high risk of cardiac disease, can we do anything about it?

Well, traditionally, the thought has been that because it's genetic and you can't change your genome, you're kind of stuck with that part. But the first thing to say is that as a cardiologist, everybody— you know, I tell every patient they should be exercising more and eating better, and that goes for pretty much everyone in the world. That's— that you can— you can say that before the patient even opens their mouth, and I say that to myself. Every morning as well. So one of the things that, that was demonstrated though is that genetic risk can be reduced through good diet, exercise, through lowering the other risk factors. By itself, of course, we'd have to be talking about CRISPR and base editing, gene editing, in order to change the genome, and that is increasingly the case. I don't think we're even remotely close to thinking about changing thousands or hundreds of spots in the genome. We're thinking about trying to change one, and that's a big lift. But the good news is that even if your genetic risk is high, environmental intervention, changing your lifestyle, can reduce it significantly. That part is clear. So it's definitely malleable, and that's the inspiration, I think, for anyone out there who's thinking that it might be a bit fatalistic to measure this genome prediction and that you can't change it, but you really can. So in fact, in the Netherlands, they have some interesting work showing that— if you just tell people that they have high cholesterol and that they should exercise more, they do it at a certain rate. But if you explain to them through that their genome has made them predisposed towards a heart attack and therefore they should exercise more, they actually exercise at a higher rate. So whether that's a certain amount of— we sometimes call that genetic exceptionalism. People treat the genetic information differently. But in reality, of course, your cholesterol and your genome probably shouldn't be treated differently. But it turns out people really do. View it differently and often are more adept at changing their lifestyle in response to genetic information than just, uh, just environmental information.

Um, someone asks, um, are there any cases you detail in the book that are particularly stuck with you? Obviously don't give out the— don't give out spoilers, but, um, but, uh, but is there, is there, is there a case that, that, that particularly stuck with you? Don't Yeah, tell us as much as you want to tell.

Yeah, yeah. I mean, in many ways, it's hard to choose because obviously I had to choose for the book. I couldn't— you know, I would have written another 2 books this size with all the stories I wanted to put in. So there was already a little bit of a filtering. But, you know, one of the stories that really is emblematic of almost everything in the book is found in the middle, somewhere in the middle. And it relates to a little baby who first came to our attention when she was still in her mother's womb.— and her heart rate was too low, 70 beats per minute, and it should be about 140 if you're swimming about in your mother's womb. So we transferred her to Stanford Children's Hospital and delivered her by cesarean section. And within the first day of her life, the first 24 hours of her life, she had 5 cardiac arrests. So her heart stopped 5 times. And it turned out she had a genetic condition called Long QT syndrome. So that's a condition, the history of which I describe in the book, but where basically the resetting of the heart with each beat is prolonged, and so the next beat can't come in in time. And neither of those two things matter much, but it predisposes to this potentially fatal rhythm. And so her case really, really stuck with me for the obvious reason that there was just this little one who was only 37 weeks of age, including her gestation, and she'd had 5 cardiac arrests. She became one of the youngest babies ever to have a defibrillator, an electrical device, and which was implanted in adults, it goes under the collarbone. In babies, it goes into the abdomen, and she had that implanted at day 2, I think. And then we realized, though, that if we could understand the genetic basis of her disease, we could give her a precision therapy, because there's at least 2 forms of this condition, one where potassium channels are affected, one where sodium channels are affected, and if you know which one, you can give the right treatment, but we didn't. And the only way to find that out was to send a test that took 3 months. So we worked with Illumina, the company that does sequencing, and adopted their rapid protocol to get the answer. In 2 or 3 days and managed to make her therapy more precise just over the course of a weekend, basically. And so that sort of story for me— and she's doing great now, by the way. I saw a beautiful photo of her on her 5-year-old birthday just as I was fact-checking the book in the last stage. I got in touch with her mom and she sent me this amazing photo of her blowing out the candles on her birthday cake. And so in so many ways, That is emblematic of the stories in the book because it— it's a patient in need, it's technology at the cutting edge, it's groups from industry and academia coming together, everyone just doing everything they can for this little one, from pediatrics, from genetics, from all over Stanford, from multiple companies coming together and being able in the end to give information to the bedside team that could make her therapy more precise, what we are calling precision medicine now. So that's one of the stories I often think about as being kind of representative of the sorts of stories in the book.

I'm going to take the last two questions and actually combine them, and because they're very similar, one is from Damayanti Gupta and someone anonymous. They're interested in computation and computational genetics. We talked about polygenic risk scores And one thing that has struck me as very interesting is that the polygenic risk scores are additive. So in the sense that, you know, they have small weights and you add them up, and when you add them up, you know, you get a score. But as we explore deeper computational methods, it might turn out that there are non-additive factors, synergistic factors, where, you know, it's not A plus B, but it's A multiplied by B that will give us even deeper understanding into the genome. And of course, we enter the arena of AI where— or machine learning— where we can use even more complex methods to disaggregate this information. Tell us a little bit about sort of what computational— what new computational things are happening to help us understand the genome at an even deeper level.

Yeah, it's a great point. And I think that— I mean, we— you talked at the beginning about Moore's Law, this idea that technology moves so fast that every 18 months the number of transistors on an integrated circuit doubles. That was the idea from Silicon Valley in the '90s and 2000s of just how fast technology was moving, and we applied it to the genome and found the genome sequencing— the cost was dropping even faster than that. And so computation and genetics have been really tied together since the onset of the genome in a way that they weren't in the decades before, as you describe in going through the amazing history of genetics in your book. But yeah, for this last decade, computation and genetics have been inseparable, and in fact many of our genetics students now who come to Stanford for their PhD are They come because they want to do computational genetics. So absolutely, there's some very interesting approaches of artificial intelligence, of course. You don't need to go to deep neural networks to be able to get a major boost in our ability to work out what's happening with a given genome through computer methods. But it's actually even more than that. I think the combination of newer technologies, improvements in our sequencing technologies, and then improvements in our algorithms and computer methods not just in speed, but in accuracy— our ability to really characterize the difficult areas of the genome, both by sequencing in different ways and also having computer methods. I sometimes tell the story about the high schooler who's now— I think he is about to graduate from Stanford. But at the time, he came as a high schooler to my lab. And we're part of the Undiagnosed Diseases Network. That's part of the medical mysteries that are through the book, come from this network of doctor detectives He basically downloaded a new algorithm, basically pulled down a new algorithm, edited it somewhat, and then pulled all the unsolved cases. This is high school kid in the summer, and he solved one of the cases. This is a family that had spent months and even years with no answers. Now, with the ability to think about computers, manipulate computers, someone who is very, very talented, but in high school can come along and solve a medical mystery. And I think that's just part of the excitement that I try to put across in the book, that you— it's an incredible time to be going into genetics, incredible time to combine engineering skills like computer science with biology skills, and just, I think, a really exciting time for young people to think about this as a potential career.

Great. We are on the hour. And that there are no more questions. You know, I was totally intrigued by the book. I loved every story in it. I was, you know, I'm a big fan of case histories. I wrote a piece about— we were just talking about Oliver Sacks and how, you know, how he was a great fan of case histories. And what's amazing about these case histories often if you look at them from Victorian times or even before, is that some of them have now, years later, decades later, we're finding answers to these case histories. Back then they were case histories and described as, you know, medical mysteries that we never understood. A family that had, you know, sudden loss of hearing and all of a sudden, you know, 50 years later, 100 years later, you figure out that that family has a mutation in a particular gene, we now, you know, we remember the history of hemophilia in the royal family and how deeply it influenced human history. And so I just wanted to send, you know, I enjoyed the book, I thought it was riveting to read and I wanted to send you a word of congratulations and encourage people to read this book, which you hopefully will buy from an independent bookstore. But if not, I'm sure you can get your hands on the book through other sources.

Any last words, Ewan? Well, just a word of thanks to you. I mean, I think your writing has been just such an inspiration to me, and I think The Emperor of All Maladies was a masterpiece, and obviously I really was excited when you wrote about the history of genetics. And so if anyone hasn't read Siddhartha's books, then they need to go out and get hold of those, because I was actually a little embarrassed because all the books from my teenage years are here, but I read yours all on my Kindle. So I had to put them on the back here. But a huge fan and really grateful to you for the chat this evening.

Hope we have a chance to— My pleasure. Yeah. Thanks.

My pleasure.

Kate, the floor is yours. Well, I'll just wrap things up. Thank you tremendously, Ewan and Syd, for this fascinating discussion. And thank you, all of you out there, for joining us tonight. If you would like to learn more, copies of The Genome Odyssey are for sale on harvard.com, as well as Sid's books. So on behalf of Harvard Book Store, the Harvard Division of Science, and the Harvard Library, have a good night, keep reading, and please be well, everyone. Thank you.

Thank you very much.