We publish our conversations with inspirational individuals in the healthcare industry to promote the innovation discourse and support framework. We spoke with Preston Estep III, Ph.D., Director of Gerontology and an adviser to the Personal Genome Project, Founder of the Innerspace Foundation, Co-Founder of the Mind First Foundation, and the Chief Scientific Officer of Veritas Genetics.
Dr. Estep describes his journey from Cornell University to his research at Harvard Medical School, explains his approach to emerging technologies, and offers promising advice to students regarding technology. A transcript of our conversation is below.
Healthcare Innovators: How did you cultivate your interest in science and genetics?
Dr. Estep: I’ve been interested in science pretty much all of my life. I didn’t really become serious in my interest until college. I was a business and economics major in my early years of college. I thought of it as being very practical but it wasn’t my true passion. I knew business and economics were important but they didn’t answer some of the deeper questions about which I was curious. They did not answer the questions I thought were important to have a more complete view of the world and the way it works. I knew I needed something more than just a generic framework for business to become truly passionate and engage in what I was doing.
After a couple of years in economics and basic business training, I took some time off from school to really find my passion. I started reading more popular science, press books from scientists, and a series of books about genetics and its molecular aspects. There were changes happening in paleontology based upon the genetic inferences that were being made about primates and their relationships to humans. I got intrigued by that, not by the paleontological discoveries, but by the general idea that genetics was becoming a harder science like physics or chemistry because of the advent of DNA and computational analysis. This is the early days in the 1980s when everything was primitive. I could see that, in the future, we would have efficient computational power and sequencing abilities to get into lots of genetic data to start to really understand the important features of genes that shape human traits. That fascinated me enough to take me back to school, back to Cornell, to get my undergraduate degree studying primarily Neuroscience, although genetics was always at the core of my interests. Those are the two passions I have in the life sciences: genetics and the neurosciences.
When I was done with my undergraduate degree, I decided to go to graduate school, not with the purpose of necessarily becoming a career academic scientist, but for the purpose of creating new things. I had an entrepreneurial interest. I was already focused on getting into the business of genetics and genomics although it was very simplistic at the time. It was obviously going to be a career opportunity in the future. I applied to graduate school at Harvard thinking I would have some very good opportunities there, not knowing what I wanted to do or with whom. I applied to other graduate programs too and had some other good opportunities, but I thought Harvard would have the greatest opportunities for me, have forward thinking people, and would allow me to integrate my interests in business and science. Lo and behold, I didn’t find any of what I thought may be there — there was no real apparent interest in the merging of the entrepreneurial aspects of science with the cutting edge work that was going on at the time.
I worked for the first year and a half just exploring basic science. I was very naive going into graduate school; I did not know what the state of the art was. I spent the first year and a half getting a greater foundation in the advanced sciences. After my first two laboratory rotations, I was not entirely satisfied. I was convinced I did not want to become an academic scientist after seeing the nature of the game, the struggle, and the constant pressure of constantly getting grant funding.
I went to the program administrator and asked her if there was anybody that she knew who was a super, outside-the-box thinker, maybe interested in entrepreneurial and computational work; someone doing things that everybody else thought are crazy at the time. She said, “Yes, there is one guy named George Church and if he’s not far out enough for you, we have nothing for you here.” I said “Ok, where can I find George Church?” She said, “His office is right around the corner.”
I made a beeline for George’s office and happened to catch him in his office, which wasn’t too rare at the time because he was a pretty young investigator with a pretty small lab. No one knew who he was. He had fewer than 10 people in his lab at the time. He and I started talking and after about 5-10 minutes, I knew I had found the guy, the person with whom I was going to do my PhD, and the guy who would help me crystallize my nebulous vision of the future and bring it to fruition. We talked for an hour and a half to two hours. It was one of the greatest conversations I’ve ever had with anybody. It was one of those moments where you just know your life has changed. You are so excited and you’ve put form to a lot of the unclear and nebulous ideas that you’ve got in your head. It all takes shape all at once.
I left his office extremely excited and knowing a lot of things that I’d thought were possible and crazy. I knew somebody else shared my passion and vision. George was interested in forming companies and creating new technologies. I joined his lab as a rotation student and started working on a really visionary, potentially crazy, project called nanopore DNA sequencing. We were the first people trying to create what the company, Oxford Nanopore, now commercialized. We tried to sequence the DNA by passing it through a pore in a membrane and measuring the electrical conductance changes as the DNA transitioned across the membrane. That was over twenty years ago. At the time, it was very pie-in-the-sky, and obviously, very difficult. After I was done with my laboratory rotation, I figured that was not a fruitful thesis topic and that I would be fortunate if I received a PhD. It’s important for people who are pursing technologies to think about their viability.
This is the most difficult part about being a pioneer and breaking into a space ahead of everybody else. You must be realistic about the technical hurdles that are required to bring a technology to commercial fruition. The timing has to be right. At the time, I wasn’t sophisticated enough in my scientific thinking to know whether or not nanopore sequencing was three or twenty-three years away. It was more so the latter in reality. George and I agreed it was a very challenging and difficult technology, and that I probably should pursue some other, more tractable project for my PhD thesis. Once I got later into my PhD thesis, I did have a more sophisticated idea of what technology and commercial products may be viable, but even then, sophisticated investors and business people do not understand markets, opportunities, and the timing. All of that only comes with experience and trying something that you think is right at the right time. Maybe it won’t turn out that way, and often it doesn’t in biotech.
There are no really good statistics on how many biotech companies fail — people use a general ballpark of 90% of all biotech companies fail. That’s partly because of the emphasis on therapeutics and a lot of times, research and development of therapeutics is a real crapshoot. I am very cautious about getting into the therapeutics space because it is a huge gamble. I’ve always known that the analytical and diagnostic space is not as lucrative but is more predictable and tractable. The development of a drug, for example, is extremely lucrative but not predictable. That’s why I’ve focused my efforts on the analytical and diagnostic space. Drugs for me are a stopgap measure between what we have now, our growing genomic analytical capacity, and our growing ability to edit the genome while completely circumventing the need for drugs by completely changing the genes. That’s the beginning-to-end story of how I’ve gotten to where I am and the thinking that has guided me in recent years, for how I made the choices which commercial enterprises to be personally involved and invested in.
Healthcare Innovators: What an incredible story. How much impact has regulation played into your own ventures? Do you consider regulation when deciding your diagnostic ventures, or is that a byproduct that you consider only after you are done with your work?
Dr. Estep: It is always important to consider. The FDA is very vigilant and active in our space. Everybody who knows the space knows the story of 23andMe. There are similar stories, some not as well known, so people and companies need to be aware. Research projects even need to be careful and aware. It’s important to recognize the FDA, at least from my perspective, is a very important agency and serves a very important function. I certainly don’t want my loved ones taking drugs or inaccurate diagnostic tests that will send them down a dangerous path that regulators haven’t approved. I do not want to do that myself. I want to know that a test or drug that I take has been vetted and analyzed by experts. The critical problem right now is that, as we move into sequencing whole genomes and returning data to people to individual consumers, it is really important to engage and return results to the public even though the genome is vastly more complicated and messy than a single diagnostic assay.
For example, even a cholesterol test is accurate. We know there is a small range of error that is possible in a test. If you get a total cholesterol reading of 250, you know it’s highly unlikely that your true total cholesterol is 125, or 150. That’s because regulatory agencies like CMS that regulate the CLIA certification process and agencies like the FDA, make sure that there are standards that companies have to meet to offer these products. For a long time, the meaning of the cholesterol test was not clear. It was not until several years after the tests became widespread that the ratio of HDL to LDL became a more meaningful metric than total cholesterol.
The genome is vastly more complicated than that one test or combination of tests that we get today. The genome is billions of nucleotides with 30,000 to 50,000 genes. We haven’t sequenced enough genomes. In the early days of cholesterol testing, people had their cholesterol levels tested and cardiac, dementia, and mortality outcomes measured so we could know what those cholesterol levels actually meant. In the case of genomes, we haven’t sequenced enough genomes and paired those genomic data with health outcomes yet to know what all the genetic determinants are of those outcomes.
There’s a lot of uncertainty in the genomics business; that’s the way it will remain for the foreseeable future. It’s important to understand we are going to be living in a world of a high level of uncertainty with lots of gray areas for a very long time. For companies like Veritas Genetics and for projects like the Harvard Personal Genome Project, we just need to be allowed to keep operating while educating the consumer, layperson, and non-scientist about all of these gray areas and uncertainties. That’ll allow us to accumulate enough data in an efficient and rapid way so that we get to the point where we are not as uncertain about some of these discoveries.
Healthcare Innovators: You have started a lot of companies and initiatives. Can you give us some insight into your own innovation and thinking process? How do you come up with new ideas for new technologies given the rapid pace of science today? What are you involved with now?
Dr. Estep: I’ve gotten approached by a lot of people to start companies. I’ve been very, very selective. I’ve focused most of my investment of time and resources in companies that, to me, are very transformative. I am interested in understanding genes and genomes, primarily so that we can engineer genes and genomes to improve the lives of people. I’m not in this business for curiosity or to make money. I am a techno-progressive. I am a realist about the evolutionary baggage that everyone carries with them.
We’ve been hearing about preventive medicine for decades and we still don’t live in an era of true preventive medicine. You can go to your doctor today and get a completely clean bill of health in an annual check-up, yet die of a heart attack the next day, or within 5 minutes of walking out of the doctor’s office. It is terrible we live in a time like this. We often look back in horror at medical and social practices. We think of how primitive things were and how terrible it must’ve been to live in those times. In 10 years, we’ll look back at the time we’re living in now and say those same things. We are going to look back and think, “Wasn’t it absolutely and inexcusably terrible that you could walk out of a doctor’s office and die after an annual check-up?” It’ll be like bone saws from the First World War. For example, if you got an infection, the only cure was to saw your leg off to prevent the spread. We’ll look back at this time and look at our inabilities in the same way.
I’ve always been heavily focused on pushing the boundaries and really getting involved in the areas that I would think accelerate progress the most towards a better future for most people with less disease, less suffering, and fewer physical and mental problems generally. One of the things that we live with that we’ll look back on and think is the most inexcusably primitive feature of our healthcare now is how invisible mental health disorders are. We deal with mental health disorders in a way that is completely different from physical health disorders. They are incredibly pervasive and far more present than people realize because you don’t see them. If somebody is depressed, has ADHD, or has alcohol/drug dependency, it isn’t obviously manifest when you see them. The mental health realm is woefully underserved. One of the things we do at the Personal Genome Project and the nonprofit I am affiliated with is bring funding to those efforts so that we can at least get a handle on these problems.
I just wrote a book about dementia and the prevention of dementia called The Mindspan Diet. One of the things I learned about dementia is that the CDC estimates for prevalence of dementia and the incidence of dementia are so far off base that the funding of dementia research is woefully inadequate. They estimate it’s the sixth biggest killer in the developed world. I think realistically, dementia is the first biggest killer and kills more people than cancer, which is now tied with heart disease as the official number one killer of the developed world. I think we are very unrealistic about our estimates of mental health problems. We haven’t even figured out how to measure it accurately and how pervasive it is.
Healthcare Innovators: How did you get into the computational neuroscience space?
Dr. Estep: I got involved in neurotechnologies that are promising for helping disabled people. The interesting thing about neurotechnologies is that they are similar to genomic technologies. Genome-editing technologies aren’t only limited to helping sick and disabled people, but they also have potential for helping everyone. One of the really interesting challenges we face as a society is acknowledging that everyone could be healthier, fitter, smarter, have better memory, be more rational, be compassionate, and so on. The human attributes we have are limited by evolutionary processes, so we think of the upper bounds of normalcy as being sufficiently good in our set of human attributes.
I’ve always been of the mindset that, if you have a problem that you need to solve, there are two elements in the solution to that problem you need to consider. If you think about doing it on a computer, it is the difficulty of the problem and the power of the problem-solving infrastructure you have. If you’ve got a very complicated problem and an underpowered computer, it will take a long time to solve that problem to the point you may never solve it. The first thing that you think of when you have a complicated problem to solve is to get a computer of sufficient power. It is interesting that when we make the leap to human problems and when the computers are our brains, we only focus entirely on the complexity on the problem. We never think about the other side of our equation.
Our brains are really underpowered to be dealing with the problems that face us today. For example, we make predictions about climate change and try to understand the difficult physics of planetary science and make projections over decades. That’s a very uncertain and complicated business. There are lots of problems like that which we are trying to solve, like our economy. The economy is incredibly complicated and composed of so many moving parts that no one individual with their brain can understand any particular economic development. You have to use all sorts of powerful computational models. There are problems in physics and in neurosciences. Understanding the brain itself is a daunting challenge.
This brings me to the paradox of our age. One of the most important challenges of our age is understanding our brain well enough so we can engineer it to make it function better. Of course, that takes a huge amount of problem solving capability and insight of natural brains. We need to get into that bootstrapping cycle, which we are already in, with powerful, simpler technologies. We need to get into a concerted cycle, where we recognize what the current bottlenecks are, and fix them to bootstrap our way into greater levels of intelligence and better problem solving. The immediate use for that is understanding the brain even better so we can make even better improvements. This is a really important problem and we can start with the people that are the most mentally disabled.
We have a large number of people who are mentally disabled for a variety of reasons. I think business, academia, and organizations should focus more dollars and efforts on the neurosciences and neurotechnology. That area is potentially such a fruitful area of investment that it would pay so many dividends in the long-term. Homelessness is a chronic problem in every country regardless of how developed the country. In the United States, for example, a large proportion of homeless people have some form of mental illness. It is one of the unrecognized burdens and probably one permanent barrier to solutions for homelessness until we get people to have stable emotional and mental lives, like non-homeless people do. I think there are entrepreneurial opportunities if we can at least begin to realize what the problems are.
Healthcare Innovators: What advice would you give to students now?
Dr. Estep: I can give some very practical advice and some very pie-in-the-sky advice. The practical advice I can give is a very interesting challenge that your generation faces. Science and math, for example, are very difficult for people. Human brains are not wired to understand those subjects, which is why they are almost universally regarded as difficult subjects. They require serious focus, study, and immersion. The thing that worries me is that one of the fruits of the kinds of technology is that it is pervasive and everywhere.
When I was learning science and math and learning the difficult concepts, I didn’t have a smartphone, Internet, or email. I didn’t have these distractions that prevented me from immersing myself in difficult material and learning and mastering it. I was able to go into my office for hours at a time, shut the door with my book, and just read and think. I am as distractible as anyone. I am in a position now where I’ve already learned all of that difficult stuff and can be more selective in my learning. I know when to turn my phone off but I know a lot of people, students in college, and undergraduates, that are literally addicted to their devices. They can’t seem to turn them off and disengage.
This is worrisome for me because to learn difficult, challenging material, you have to do that. You have to block off time so your brain can change, so you can get enough information to your brain in an uninterrupted flow so that your brain actually adapts and changes as you study. That is my practical advice. For people who are addicted to their devices, they’ve got to find a way to get a sufficient block of time to immerse themselves in the material so they can learn these difficult concepts. At the very minimum, to get even more practical, turn your device off for as long as the task will take. Say for example you have an hour-long test coming up in a month. Every day or night in preparation for that test, study uninterrupted with your device off, for at least one block of time as long as that test. That will put you in the mindset of being uninterrupted and having total focus for that length of time. When you get into the habit of having uninterrupted focus for that length of time, you’ll be much more comfortable, be in a groove, and be sufficiently focused on the task.
For more pie-in-the-sky advice, try to think about the very long term — in other words, try to emphasize long-term goals over short-term ones. Plan for a future that enables you to work on increasingly powerful technologies, like genome editing or more direct and better human-device interfaces than touch screens. But beyond any specific technology, try to imagine the challenges of working in a world where people are even more immersed in their devices, and think about ways to make that experience safe and fruitful. For example, at the moment we see an increase in car accidents as the result of device distraction. Self driving cars are one possible solution to that but there are other costs of device immersion that occur off the road, but those might be improved greatly or even turned into positives with imaginative applications of better technologies. Again, don’t limit yourself to thinking just about the problem or the hardware, think about how brains and minds might be changed so that they are better suited for modern challenges. Study and education physically change the brain, but don’t limit your options or toolkit for changing the structure and function of the brain for the better.
And prepare quantitatively as much as possible. I think if you are capable in mathematics and computational, quantitative stuff, take as much as you can. It is important to have a quantitative and analytical background. I’ll say something that will sound heretical to some undergraduate curriculum committees, but I think that it is more important to focus on probability and statistics than it is any other kind of math. Probability and statistics should be the core mathematics rather than calculus at this point. Almost every scientist I know uses probability and statistics routinely, probably every day, and the business people I know use it. One of the things we see most frequently in articles that are interesting and quantitative are statistical reports. Even in elections, we hear about polling and statistics.
Statistics and computer science are very important with a focus on medical and biological sciences. A lot of quantitative people I know learn and go into building an app or getting involved in a computer science or web-based startup company. I would really like to see people be more patient about the development of their careers and take that quantitative foundation into biology, medicine, and neurotechnology. Those areas are more interesting and potentially more fruitful than getting into a crowded space like building apps and web sites. More quantitative thinking is needed in the medical sciences. The combination is extremely powerful in our modern, preventive healthcare structure that we are trying to build. One of the biggest parts of our business at Veritas is building genomic interpretation and the analytical engine that teases information out of the genome, medical traits, and outcomes.
Healthcare Innovators: You make a good point about the emphasis on probability and statistics. I didn’t realize how less I use calculus in my non-calculus coursework, asides from physics, perhaps.
Dr. Estep: Yes, most people don’t use rates of change calculations. Even many engineers, like software engineers, rarely use that. They use a lot of probability and combinatorics. Calculus is important for civil engineering, mechanical engineering, and physics, but even so, the majority of undergraduates would benefit from a solid background in statistics.
I took statistics as an undergraduate and a separate class in probability. My colleagues who are in charge of college curricula admit that statistics is far more useful. Academia is so mired in conventional thinking. Nobody wants to be the first to part from tradition. I think it’s time to make the change from calculus to statistics and probability. It’s so much more important.
Healthcare Innovators: We’ve noticed a lot of the opportunities within biotech and genetics even require advanced degrees or PhD candidates. How can we become involved before we begin such an advanced program?
Dr. Estep: I think it is beneficial to do undergraduate research before applying to graduate school. I did that and it was a great experience. I learned a lot of basic ways of thinking about science realistically. There is textbook science, classroom science, and hands-on science. You realize that a lot of what you’re learning in the classroom and in textbooks is helpful for the background, but when you finally get in the lab and on a computer to solve a problem, you start to think in a very different way and you start to become practical. I encourage every undergraduate to do research and start his or her research experience by doing research into what research is available to him or her. Go visit labs, go talk to principal investigators, and see what they actually do. Stand in the lab and watch people while they are working. It’s a real eye-opening experience. There’s a huge range of different kinds of things, like engineering, robotics, and computer science labs.
Healthcare Innovators: Thank you so much for your amazing advice. This has been tremendously helpful.
About the speaker
Dr. Preston Estep is the Founder of numerous companies and non-profit organizations. He currently serves as Chief Scientific Officer of Veritas Genetics. Dr. Estep earned his Ph.D. in Genetics from Harvard University and conducted his doctoral research in the lab of Professor George M. Church at Harvard Medical School. He most recently published a book called The Mindspan Diet.