News

Fireside Chat with NIH Director Francis Collins and ISB Co-founder Dr. Lee Hood

ISB Co-founder Dr. Lee Hood hosted a fireside chat with NIH Director Dr. Francis Collins

The two renowned scientists talked about their early careers and long friendship, the challenge of COVID-19, the preceding scientific work that led to the fast development of COVID vaccines, and much more. 

The conversation was part of PMWC 2021 Virtual, a five-day event that took place June 14-18, 2021. Learn more at https://www.pmwcintl.com.

You can watch the fireside chat here, or by clicking play on the video above. For a transcript of the conversation, please scroll down this page.

Transcript from the Lee Hood-Francis Collins fireside chat

Lee Hood: (00:00)

Francis, what a pleasure to be able to talk with you today. And I find it particularly fascinating because your career has involved a fascinating evolution. I remember meeting you first at Yale when I think you were working with Sherm Weissman on chromosome, working in chromosome jumpiness as I remember at that time. And to see you then evolve from, if I remember correctly, physical chemistry, PhD, getting interested in biology and then through a series of stages, moving to human genetics. And I think that’s where I first met you. When you were exploring techniques for being able to isolate genes with Sherman Weissman to generate everything. And then later moving to Michigan and doing exciting things. And then finally getting started in your administrative responsibilities as director of the National Human Genome Research Institute, which evolved into the Human Genome Project. And later you even moved up one additional step when you became director of NIH.

Lee Hood: (01:22)

So it’s a rich and diverse career. And actually, I should say the other stage, albeit it qualifies as a stage, is COVID-19. Because it’s been so unusual and so dominating in science this past year and a half. And so let me ask you some questions about COVID-19. I be curious, from your point of view, what do you think are the most important contributions that NIH made to the whole quest to understand and deal with COVID?

Francis Collins: (02:04)

Well, thanks, Lee. There’s a lot we could say about that, and it’s a pleasure to be here with you in this interview and yeah, we go way back as colleagues, as friends. You remember me as a post-doc working here in Sherm’s lab, trying to figure out how we could travel across distances in the genome, when there was no genome sequence to help us know, click of the mouse to say, “Oh, just go over here. And that’s where it is.” It was pretty tough back then, wasn’t it? But my goodness, and now our graduate students can’t imagine how you and I did what we did back in those days. It was really difficult and now it’s so straight forward. But now we have new Mountains to climb and certainly COVID-19 has been an amazing challenge. For me as the NIH director, presiding over this largest supporter of biomedical research in the world, the NIH, mostly from right here in my home office, where I’ve been kind of as a homie for the last year and a half, trying to manage all this safely. It has been an amazing ride.

Francis Collins: (03:03)

And of course the first thing to point to is the vaccines. To be able to deliver in 11 months of vaccines that are shown to be safe and effective. And granted emergency use authorization, is about five times faster than it has ever happened before for a human infectious disease. And much of that built upon 25, 30 years of preceding really important, basic science work, evolving people, understanding how messenger RNA might actually be potentially a way to develop a vaccine quickly. And then the vaccine research center at NIH in that early January 2020, when the sequence of SARS-CoV-2 was first put on the internet, within 24 hours designing the vaccine. And within 65 days that first test dose going into somebodies arm in a phase-one trial. And again, people shouldn’t think that just happened because somebody thought of it in January 2020, it’s built on this long history of understanding. I guess you could go back to understanding what messenger RNA was all about in the 50s. And one more example of how science really is a process that build step-by-step upon the work of many people, many of them doing basic science that you wouldn’t have realized it was going to be so profoundly important.

Francis Collins: (04:28)

So the vaccines are a big part of this, and it is wonderful now to be able to say that they’re closing on to 70% of the country that has had at least one dose. And of course we need to move quickly to get the same immunization opportunities out there for the rest of the world. And president Biden is now very committed to making sure that happens. So that’s one. Another has been therapeutics. And when we looked back in March of 2020, it was a very scattered shot effort there in terms of how we could figure out whether there’s some therapeutic that might benefit this viral infection, ideally quickly, maybe we could repurpose compounds that had been used for something else.

Francis Collins: (05:13)

And so that meant getting our act together and put together a partnership with about 20 pharmaceutical companies, the FDA, the CDC, the veterans administration, and we all work together to try to prioritize which therapeutics ought to get tested first, design master protocols. So that could be done rigorously and pull up a clinical trial networks together to quickly mount those trials. And that meant more than 20 therapeutics have now been tested. Some of them have been shown to be beneficial. Remdesivir, various monoclonal antibodies for outpatients, a lot of them turned out not to work, but you need to know that too. Like Hydroxy chloroquine, which was such a focus of so many trials. And now we know really it wasn’t going to be the answer. So that was a big effort and it’s not over. And now we’re moving, Lee, into the more specific antivirals like drugs that are designed to specifically target the viral lifecycle, like anti proteases that are now in our trials from Pfizer. That was a big effort.

Francis Collins: (06:14)

And then thirdly, it was diagnostics. And you’re a technology guy and I am too. So when Congress gave NIH an additional billion dollars back in April of 2020 and said, “We need more tests and we need them out there in three or four months.” We decided, okay, let’s pretend we’re a venture capital company. And let’s open the door to people who’ve invented interesting technologies that need to be really tested, validated, and then scaled up and see what you got. And we were astounded at the response. We got more than 700 applications, mostly from small businesses. And we put them through an innovation funnel to see what was going to work. And right now, today, now there are 32 of these technologies out there, collectively contributed about 2 million tests a day, that otherwise would not have happened without NIH in this additional funding. That was quite an experience of doing things very differently than the usual grant cycle. And it wouldn’t work for a lot of what we do, but for this, it worked.

Lee Hood: (07:16)

Beautiful. What’s your feeling about vaccines and their ability to deal with variance? So that’s one part of the question and the next part is, as a former immunologist, I’ve always wondered what role cellular immunity played in these vaccines. And it seems to be something we know very little about at this point in time.

Francis Collins: (07:41)

No, you’re quite right about that. And it is something we really need to understand better. In terms of the variance, we are worried about those. I guess they all now have Greek letters, although I still keep remembering the numbers, but alpha, beta, gamma, and delta, about four variants of concern that have popped up in various parts of the world. So far we are looking okay that the existing USA-approved vaccines work well enough to protect against those. Although maybe the titers of antibodies that are neutralizing are a little lower against the South African variant or the variant from India than they are from the original, but they’re good enough, but I don’t know how long they’ll be good enough. The other question is, yeah, what’s the decay function look like of your antibody levels and they clearly do drift downward a bit over the six or so months that we’ve had a chance to measure them.

Francis Collins: (08:35)

And when will they actually drop below a protective threshold? We don’t know so far so good. Nobody is saying you need a booster today, but boosters might very well be in our future at some point. And they might be here sooner if other variants pop up that are even more different than the original spike protein. And therefore aren’t as well covered by existing vaccines. But Lee, I’m glad you brought up the immunology here. I’m not an immunologist, you are, but I am aware that there are certainly some data to suggest that T-cell responses might be just about as important providing your protective immunity from the vaccines. And yet we do a poor job of measuring that. So it’s much harder for us to say whether that also is decaying or whether it’s holding up.

Lee Hood: (09:24)

And you know, I think the important thing about T-cell immunity is if the viruses are actually internalized, it’s the only way we can deal with those cells that have been so infected. But I think that’s a fascinating question. But another fascinating issue about COVID 19 has been long COVID. The idea that you can go out three months afterwards and still see maybe 30% of patients with diverse syndromes that range from respiratory to liver, to cardiac, to GI and especially to mental. I mean, it requires a kind of a multi-disciplinary medical approach to these individuals. What is your feeling and the potential for the enormous numbers of individuals that may go on to the healthcare system and require major kinds of medical support? What’s your feeling, A, about what really is the costs of all of this and B, how are we going to deal with it effectively?

Francis Collins: (10:40)

I’m glad you’re bringing it up because this is a really serious issue. And we don’t understand it. We can categorize that this happens to maybe 10, maybe as many as 30% of people. It seems to be more common in those who have more serious illness, but even young people who’ve had relatively mild illness can get long COVID and not be better weeks or months later. I don’t understand. And I don’t think anybody does, what is the pathogenesis here? Is this a leftover of the immune system’s reaction, that’s making you feel badly? Is this because of the propensity to blood clots, which we know can happen in little vessels all over the place? Is that what you’re seeing here? Is it some persistent, viral infection in some reservoir that we can’t find? You can’t find virus there, but maybe it’s hiding somewhere. We just don’t know. So NIH now has a very large scale program where we aim to follow 20 to 40,000 COVID survivors and track in a very precision medicine kind of approach, what exactly is going on and how do people differ and who are the people susceptible to this, and how do we actually treat it more effectively? Because right now there are clinics popping up all over to take care of these people, but they’re desperate for answers about what they should be doing. Because right now you can’t see the symptoms and that’s it.

Lee Hood: (12:04)

Well, one of the things we’ve with done ISP is very deep immune phenotyping, where we analyze patients and at each blood draw off each patient, we take 5,000 white cells and we do single cell analysis, complete transcriptome cell surface secreted molecules. So it means for each of those 5,000 cells, we can define cell type and stage of activation, stage of differentiation, the whole business. And this gives us a view of the immune system that in some way is on long go of that patients, is really beginning to lead us to the conclusion. It may be fundamentally auto-immune in nature, a really fascinating hypothesis that clearly can be tested.

Francis Collins: (12:53)

I think that’s exactly the kind of work that needs to be done. And I’m glad you bring it up, the single-cell approach. That’s been such a revolution of everything we’re doing and it certainly fits nicely here.

Lee Hood: (13:04)

Yeah. Yeah. So what is your prediction of where we’ll be vis-a-vis COVID in one year?

Francis Collins: (13:13)

I wish I had a crystal ball that wasn’t quite so cloudy. I think at least in the shorter term, we’re going to see an improvement in the United States situation going through the summer. I do worry though, Lee, that come the fall and the winter as people get more crowded together, the weather changes. If there are places that didn’t get their vaccination levels any higher than 50%. And there certainly seem to be some that are not doing very well to even get to that point, that we will see other outbreaks. And certainly I will worry, especially if there are other variants coming to our shores or maybe home grown that are also more infectious and more dangerous. I don’t think we’ve seen the end of COVID-19 in the United States. And then the rest of the world, of course, desperately needs those vaccinations. Haven’t for the most part gotten even started. And so the virus will continue to rage in other places for at least the next year, maybe two years, depending on our ability to get him innovation out there and for people to accept it.

Francis Collins: (14:16)

So we’re going to be in this for a ways, and nobody should imagine that we can now kiss this one goodbye. It’s going to be vexing us as a global problem. All the more reason why we need to work really hard also on the therapeutics and not give up on that because people are going to get sick and we need something for them.

Lee Hood: (14:34)

So my last question in this area is, what lessons have we learned to prevent the next pandemic?

Francis Collins: (14:43)

I think we have learned a lot and I just hope we don’t slip into complacency and forget it as we end up in a somewhat less frightening space. I think we have learned that messenger RNA vaccines are really powerful, really effective, and we could start priming our pump for the other pandemic pathogen candidates that are out there and not wait for that next emerging problem. Virologists can actually sit down and make you a list of about 20 of what they think the likely prototypes are for the next pandemic. Why don’t we go ahead and design vaccines that would be as broadly applicable as possible to members of those viral families and carry them all the way to a phase-one trial and have that part of the work done so that you can start almost immediately to move forward into therapeutics and maybe have a vaccine, not an 11 months, which was still an amazing world record, but in half that time.

Francis Collins: (15:40)

And similarly with therapeutics, we should be, I think really pulling out all the stops to identify those vulnerable targets in those common viruses that could be subjected to a careful structurally-based drug design approach to see what we could have already on the shelf or in the freezer compounds that had good activity and good safety profiles against those major categories of the next pathogens that are lurking out there.

Francis Collins: (16:06)

There’s no reason without, well, some money will be required that we couldn’t do that and be ready the next time.

Lee Hood: (16:13)

Great. So let me ask you a question I’ve long been curious about in your scientific career, Francis. I noticed you got a PhD in chemistry from Yale and then you switched into a career trajectory that took you ultimately to human genetics. And I just be fascinated about why did you make the change? What prompted you? Was there a mentor or was it reading a paper or what was it?

Francis Collins: (16:46)

I got first excited about science as a high school kid in a 10th grade chemistry class by an incredibly gifted professor. And I thought, “This is so much fun. I must be destined to be a chemist.” Because that’s what he’s teaching me. And biology on the other hand, I got to say in high school was just awful. It was like there was no principle kind of basis at all. It was just memorized. It’s tough and it was boring and it seemed murky. And it was completely intellectually devoid of interest. So I avoided biology all the way through college. I focused on chemistry and physics and math and went off to get this PhD in quantum mechanics, which was a lot of fun, but I did kind of figure out, maybe I missed something. I’ll tell you what it was. I’m working as a graduate student. It’s two o’clock in the morning, I’m back from the computer center, with my box of Fortran cards. And the other person who’s in that chemistry lab at that point is a graduate student who was working on DNA in Don Crothers lab, which was a DNA lab at Yale [crosstalk 00:17:53]the building.

Francis Collins: (17:54)

He and I were the only ones up at 2:00 AM. And so I would ask him, “So, what are you doing?” And he would ask me what I was doing. And he didn’t think what I was doing was very interesting, but what he was doing, I thought I had no idea. Biology sounds pretty interesting after all. It makes sense. I missed something here. So I took a graduate course very quickly in molecular biology and taught by Peter Lengyel and got totally fired up and realized, ah, that’s the kind of science that I was really looking for. I just didn’t know I had kind of missed it. But Lee, I was all shaken up about, “Well, wait a minute, what am I really supposed to be doing? I know I want to study life, but I’m not quite sure in what way, I’d better go to medical school.”

Francis Collins: (18:41)

A great answer to that admission’s committee question. Like, why do you want to be a doctor? But I guess it was good enough because they took me at UNC and I loved it from the first day. And I loved it, especially when I started hearing about patients that had diseases that were caused by genetic misspellings, things like sickle cell disease or cystic fibrosis. I thought that’s how I could put it all together. I liked the math. I liked the information content, but I like something about humans that I might be able to help someday. There you go.

Lee Hood: (19:15)

Your illustrious career in human genetics accomplished many things, but I’d like to ask you, what are you most proud of that you did in human genetics, if you could pick out one event?

Francis Collins: (19:34)

Well, I’ve been really fortunate to have a chance to usually in a team effort, play a role in something in the way of an advance beginning with cystic fibrosis, gene discovery and moving on to what I’m working on right now with this rare disease of premature aging called progeria, where we have hopes of being able to cure that disease with gene editing. But I think I’d have to say the privilege of being the international leader of the Human Genome Project. That had to be the top of anybody’s career, to oversee this amazing team of 2,400 scientists in six countries who all agreed to work together to read out that first copy of the human DNA instruction book and did it, and did it really well and gave it all away. That was just amazing.

Lee Hood: (20:20)

Yeah. So how do you think human genetics is evolved since you first stepped into it up till today? I mean, obviously the human genome transformed what you could do, but I’d be curious if… So the question is how is genetics matured as a discipline in the last 20 years?

Francis Collins: (20:49)

Well it’s matured and it’s basically spread into everything in the last 20 years. If you walk into any molecular biology lab, whether they’re working on humans or some other organisms, they’re using genomics as approach, because it’s kind of become the center of the center of everything. It is after all the foundation of how a life system works. Is the instruction book. And it is so facile now because of the technologies, like the one you particularly got going called DNA sequencing that people can ask and answer questions that are really quite profound. Even as simple, small team, can do things that maybe we wouldn’t have imagined possible 10 or 20 years ago. So it’s powerful and it’s fundamental. And therefore, in terms of the basic science, I am blown away by everything that has been possible because of genomics. And we just talked about things like single-cell analysis, using RNA-Seq or looking at chromatin with ataxia. All the things that we can do now at amazing scale that a single graduate student can do in a given week. It’s just amazing.

Francis Collins: (22:00)

In terms of the medical implications, those have not come as quickly as maybe I would have hoped being a physician, but they have come. And certainly cancer is the one you can point to that has been completely transformed by the ability to look in any given patient and say, “What are the drivers of that malignancy at the DNA level?” And I mentioned earlier how this has transformed in pediatrics, the ability to make diagnoses of mysterious conditions, that’s certainly the case. You mentioned forensics. I think we could say pharmacogenomics is getting there, but it’s not quite there, but the idea of what I had dreamed of and still do, have the prevention approach where each of us has our complete genome sequence, and within it our cues about things we should pay attention to and maybe other things we don’t have to pay that much attention to. So we get away from one-size-fits-all prevention to real precision prevention. We’re still on that road. You and others are helping us find that path. And certainly the all of us program that I’m really excited about is going to give us a lot of that needed data, but we still have work to do.

Lee Hood: (23:07)

Do you want to tell us a little bit about all of us since it is one of the major efforts of NIH these days?

Francis Collins: (23:15)

Indeed. Well, it is coming really well. And now with Josh Denny coming on board as the CEO, we are up to almost 300,000 fully enrolled participants. Who’ve given bio specimens, electronic health record access, answered a gazillion questionnaires and have physical measurements. Those are all there. And the researcher workbench is open and people are using it, are already papers are getting published right and left on that, but we want to get to a million and we got slowed down, as you might imagine by COVID-19. So enrollments basically went to a slow boat for a bit, but they’re cranking back up now and we do expect to get there. It’ll take another couple, three years at the pace we’re going. But we’re constantly adding new ideas to this wearable sensors to keep track of a lot of information about our participants and they’re our partners. And so they’re completely willing to be engaged in follow-up studies, they are pre consented for recontact.

Francis Collins: (24:12)

So this is going to be an amazing engine for answering an awful lot of questions. And the other thing I’m really pleased to see is we have managed to live up to our promise of making this the most diverse cohort that’s ever been assembled. A 51% of these participants are racial and ethnic minorities. And on top of that, there’s a lot of rural residents and a lot of people in lower socioeconomic status that have joined up through their community health centers. So if you’re interested in health disparities, this is going to be a phenomenally powerful way to try and get answers.

Lee Hood: (24:46)

It does indeed sound like an exciting project. And one related to one that I’m very interested in. In your view, it seems to me a really unique opportunity for NIH, is ARPA-H. This idea that… Can we create something similar to what DOD has done with DARPA, where you bring in people who are invested for a few years and they’re given enormous freedom to invent the future. How do you view that as a possibility? And one question is, does that need to be protected from the routines of NIH itself?

Francis Collins: (25:37)

So I’m very jazzed about this, and it is a very high priority for president Biden. And for Eric Lander, the white house science advisor, who has now officially been put into place as a cabinet member, no less, because this is the right time to do this. I think the experience from COVID-19 further underlines that there are areas of science that are amenable to this kind of DARPA approach, but maybe don’t get served as well by the typical RLN mechanisms or even cooperative agreements where you want to have these program managers who have a lot of flexibility, a lot of resources. And they go out and recruit the participants they need, even for companies that might never think of writing a grant to NIH and just make it work. And they set milestones that are rigorous and they pull the plug on things that are failing, hopefully early. All of the things you’d want to see kind of like what I said we did for COVID 19 diagnostics. It was very much a DARPA kind of project giving me more confidence that NIH can do this. But you’re right, it has to be sent in a separate part of the institution. It can’t just be diffused out amongst all the institutes and hope that it works. It’s got to have its own culture, its own absolute determination to be nimble-

Lee Hood: (26:51)

And that’s what we did with the Human Genome project, by setting up. And that made it work.

Francis Collins: (26:56)

Exactly. I’m glad you mentioned it. Because I do think we have a precedent here of how to do this. It’s just been a minority of what NIH has done. It will still be a minority, but let’s see if we can encourage it to go faster than it otherwise would. So I’m hopeful with the president having proposed this in the budget for next fiscal year, that the Congress will agree and we’ll be able to launch this sometime by the end of this calendar year maybe. That would be great.

Lee Hood: (27:22)

So we’re coming unfortunately toward the end of our conversation, but I’d like to maybe close by the question of what you see as the biggest challenges ahead for NIH.

Francis Collins: (27:40)

Well, there are certainly plenty of them. Our most critical resource really is not the buildings. It’s not the equipment or the fancy technologies. It’s the people. And particularly we have to think about our next generation of leading scientists. And it hasn’t always been easy to see that path forward for people who are early in their career and we need to do everything we can to nurture and encourage those leaders of the present and the future. We have worked really hard at NIH to prioritize those applications. If you’re an early stage investigator, who’s not had an NIH grant before, you’re going to get a special boost in that opportunity to get funded. And we’re going to continue to do that. I particularly worry about those who are coming from less advantaged backgrounds. We need to work really hard on the diversity of our workforce because it is not representative of our country. And that’s holding us back as far as productivity is. Anybody can tell you who looks at the relationship between diversity and productivity.

Francis Collins: (28:47)

So we have at NIH now, a big emphasis on new programs to try to recruit and retain talent of all sorts. And that certainly includes women. Where we’ve done a better job, I think in doctoral training, but not such a great job in an academic tenure track positions and having the opportunity for women to advance all the way to large roles in high level leadership, we are losing women along the way. One of the things we’re doing right now is to provide more opportunities for flexibility in schedules and in childcare. And we need to do as much of that as we possibly can. That’s my greatest concern. Is that we need to have this amazing workforce for the future for all of the science opportunities.

Francis Collins: (29:28)

And we’re not fully taking advantage of that by the way in which our workplace is designed. On top of that, we got to have resources. I mean, I got to say, for me having had the chance to work with the Congress, especially over these last six years where NIH’s budget has gone up by 42%, thanks to bipartisan strong agreement that this is one of the best investments the government makes. I don’t want to see that momentum lost.

Lee Hood: (29:55)

That’s for sure.

Francis Collins: (29:56)

Yeah. The idea of inflation plus 5% as a good multiplier for each year’s budget. I love that. I maybe encourage that a bit and I hope we can see that sustained with some of the heroes that we have in the Congress, although they’re always turning over. And so we need new ones all the time.

Lee Hood: (30:13)

All the time, yeah. Maybe let me ask you then, a very final question. And that is, as you look back over your diverse career, is there one thing you wished you’d done that you didn’t do?

Francis Collins: (30:31)

Oh boy, there’s a long list. I’m not sure if I can pick just one because I know I’ve screwed up so many times. I will tell you, when I first came to the job at NIH of being asked to lead the Human Genome Effort for the United States, I had no clue what I was doing. And I think the first couple of years were a bit of a floundering experience. And I’m glad people like you and the others who were trying to make this work were tolerant of the fact that I had never run anything bigger than my research lab. And now here I was stuck in the middle of a government bureaucracy and it took me a while to figure out how to make that work. So yeah, if I could go back and suddenly have all of the experience that I do now about how to make this government engine actually do amazing things, we probably could have had an even better first couple of years. But you know, it worked out because of the people involved. That was again, what did it.

Francis Collins: (31:29)

When we started that genome project Lee, we made a promise, or Jim Watson made it for us, that we were going to have it all done in 15 years. And nobody had a clue about where that number came from or how we were possibly going to get there, but it was the inventiveness and the determination and the shared vision that got everybody to sign up, and then look what happened. It was just-

Lee Hood: (31:50)

We got it done early and on budget.

Francis Collins: (31:54)

Ahead of budget, I might add. 400 million bucks less than what had been projected. I’m fond of reminding the Congress of that from time to time.

Lee Hood: (32:02)

Yeah. Well look, Francis, this has been a delightful interview. I’m sure you’re going to inspire many scientists, both young and older, so I wish you continued success. And I look forward to seeing where your leadership is going to take us next.

Francis Collins: (32:20)

Well, Lee, it’s great to be able to have this conversation with you. Thank you for all your leadership over these years in getting us to where we are, but with much more to come. So many thanks.

Recent Articles

  • Nir Barzilai on Health Span, Life Span and the New Science of Longevity

    Longevity and aging researcher Dr. Nir Barzilai participated in a fireside chat conversation with ISB Co-founder and Professor Dr. Lee Hood. The two renowned scientists talked about Barzilai’s study of 750 centenarians, how aging research has changed over the years, and what exciting developments are coming.

  • Microbiome and weight loss

    Can You Lose Weight? Ask Your Microbiome

    The strongest associations with weight loss success or failure – independent of BMI – are found in the genetic capacity of the gut microbiome. These new findings open the door to diagnostic tests that can identify people likely to lose weight with healthy lifestyle changes and those who might need more drastic interventions.

  • Drs. Jim Heath, Yapeng Su and Jihoon Lee

    Metabolic Changes in Plasma and Immune Cells Associated with COVID-19 Severity, May Predict Patient Survival

    Researchers from Institute for Systems Biology (ISB), Fred Hutchinson Cancer Research Center and other organizations have uncovered underlying metabolic changes that regulate how immune cells react to COVID-19. These findings are associated with COVID-19 severity and may predict patient survival. The work was published in the journal Nature Biotechnology.