This transcript has been edited for clarity.
Eric J. Topol, MD: Hello. This is Eric Topol of Medscape, with my cohost, Abraham Verghese. We are really pleased today to have a conversation with Dr Nathan Grubaugh. He is at Yale School of Public Health. We like to claim him as a product of Scripps Research since he did his postdoc here with Kristian Andersen.
He has a rich background in biotech at Johns Hopkins. He was at Colorado State. And he has done extraordinary work even before getting into the pandemic. Nathan, welcome.
Nathan Grubaugh, PhD: Thank you for having me. It's a pleasure to be talking to you today.
Topol: Well, you have been rocking it there at Yale. I've noticed you and your colleagues, which span the gamut from immunology with Akiko Iwasaki, who Abraham and I had a chance to talk with; to Saad Omer on wastewater; and the work you're doing with Anne Wyllie on rapid saliva.
You're covering every base of this pandemic. How are you doing it?
Grubaugh: Looking back at this over the past year, it's sort of crazy to think about all of the different things that we have done. It really stems from the beginning, going back to late February, early March last year, when a lot of us were realizing what was going to hit us. Maybe not everybody fully had an appreciation for it.
We just got together as a lot of different labs from the School of Medicine, from the School of Public Health, from the research standpoint, and also with the Yale New Haven Hospital to be able to supply basic support. At that time, hospitals weren't even able to do testing because there weren't any tests available.
So really basic stuff, and it just grew from there. We started enrolling patients early because there was obviously so much that we needed to know about: immunology, pathogenesis, and epidemiology. We formed a group called Impact. We shared clinical samples. We shared data.
Of course, Akiko has been rocking it on the immunology side and gets in further than what I understand. But it's just so great to be able to contribute to that. I think that they probably feel the same way on our end with the diagnostics and with the genomic epidemiology.
Topol: I think it's a case study of how a university can put together all of its brain trust in a transdisciplinary way; I try to follow what's going on out there. I haven't seen anything like what Yale has done. You're at the center of it and we really appreciate all your efforts.
Obviously, you've been a leader in outbreak analysis with genomics. You've been there with Zika, West Nile, and other pathogens. Last night, I got an email from the NBA. They wanted to tip me off that a preprint is coming out with the NBA players and staff that had B.1.1.7 compared with non-B.1.1.7, shooting down the idea that this was a viral load issue and could be an issue of clearance and prolonged infection.
Of course, I looked at the authors. You are one of them. Can you comment about this? This is very interesting. It's only in a limited number of people, although a fair good anchoring with the non-B.1.1.7 strain. What is your sense about this strain and that finding?
Grubaugh: There is an N of 7 here, so we have to consider that. We're finding that the duration of infection — at least by PCR testing — with these seven individuals, who are otherwise mostly healthy individuals, is between 10 and 17 days.
That is considerably longer than what we had been finding with non-B.1.1.7 infections that are more in the 7- to 10-day range. It forms this hypothesis that perhaps it's more transmissible because you're infectious for longer, not because you have a greater load at any one time.
We do need to look at this in other populations, obviously. The reason why we wanted to get this out quickly is because this could have an impact on our isolation protocols. Certainly, the NBA is looking at this and how long somebody might need to be isolated if they have this variant.
We thought that it would be a good way to get information out there to share with other groups that might be trying to generate similar data, to provide some guidance around how to control this.
Developing a More Tolerable Rapid Test
Topol: It wasn't the first intersection you've had with the NBA, which is also noteworthy because we don't think of professional sports as contributing to science. But because they have so much money, they can do so much testing. It's vital to keep their players healthy.
You were involved with the bubble, with the saliva testing. Can you tell us about what that experience was like?
Grubaugh: It seems like decades ago, but maybe it was April or thereabouts. We had just put out our preprint with Anne Wyllie on comparing nasopharyngeal (NP) swabs to saliva and finding that saliva was at least as good for detection as the swabs were.
We got a strange email from somebody at the NBA, saying they would be interested in talking to us about this because they were looking for ways to do testing on players. Of course, when people are getting daily tests, having something that is not as invasive as an NP swab is good.
We had several conversations with them and it led to this research project that became what was SalivaDirect. At the time, Anne and I were mainly interested in just telling people that you can use saliva. We weren't really interested in developing our own test.
They kept on saying, well, we'll just use your test. And we said, we don't have one. They said, well, can you develop one? Then we started thinking about if we were going to develop one ourselves, what are our values there? We put that into play. What was really fantastic about this is that because they were already going to be doing testing on players and staff every day, leading up to the bubble, it was then pretty easy to tack on a paired saliva test for this.
We ended up with some 4000 paired tests between swabs and saliva. It was a really large study. The funds from that helped us develop all the lab data we needed, as well as the clinical data. Much of what went into our emergency use authorization came from the hospital, but this was completely funded by the NBA.
As the study was going on, it turned out they were interested in doing some virus sequencing. Dave Weiss from the NBA had been asking around within the group about who could do this. Well, he was already working with someone who can do this.
We started doing sequencing with them and we became integrated with their epidemiology team, where there's just so much data. They've been so open about sharing that and wanting to promote research. There are several other outputs that I think are going to come from this that we've worked on through the bubble.
We've been working with them all season. We've done sequencing to identify whether cases are related or unrelated, which is a simple but fundamental concept when you're thinking about controlling outbreaks. We've been doing this with them for months and hopefully have a nice paper on the back end of that to describe how this works.
Topol: Those were the sequences that were also the B.1.1.7 comparison, right?
Changes in Transmission
Abraham Verghese, MD: You've done such a remarkable job, Nathan, of getting data, especially from Connecticut, but also much more broadly. For our listeners, just remind us again, if you would, about what makes B.1.1.7 so special. What makes the South African strain different?
It's groundbreaking to think that B.1.1.7's main issue is the prolonged duration of infection. It's a breakthrough in our understanding. Can you take a moment to describe it at a molecular level, if you will, just so we're on the same page here?
Grubaugh: B.1.1.7 has a constellation of mutations — maybe some of them individually, but likely a collection of them— that appear to enhance transmission. We see higher secondary attack rates and faster clade growth from the data that were coming out of the UK on this.
And now we have more data from Denmark and the US that suggest that this wasn't just a population effect. This is clearly something that's linked to the virus itself and makes it more transmissible. Also, in the UK, some of these viruses have obtained the E484K mutation, which I'm the most concerned with because this seems to have some effect on antibody neutralization.
That's really the primary concern now when we start thinking about B.1.351, which was first detected in South Africa, and also P1, which was first detected in Brazil and Japan — that it has these mutations. I don't think we quite know if those variants are truly more transmissible. We are now finding them popping up all over the place.
What happens when you have multiple, more transmissible strains that hit each other at the same time, the same location? They can't all be more transmissible. The fact that some of these variants carry mutations that could impact vaccines, I think, is really the biggest worry of all.
Verghese: As a follow-up to that, it seems like a wonderful act of serendipity that you happen to be testing the NBA players and could get the data that you just showed us on prolonged duration of transmission. What is the obstacle keeping us from doing that nationwide and having this kind of monitoring? Explain that to us, if you would.
Grubaugh: Well, one of those obstacles has been partially lifted today, February 17, with the announcement of $200 million from the CDC to start creating programs. To be quite honest, I think that we had been handcuffed by the previous administration because we didn't put effort into this. My lab and lots of academic labs and state labs had started sequencing right at the beginning of the pandemic.
I don't think most people, even in public health, at that time knew the importance of doing this work. That was just never prioritized. It wasn't prioritized at the top level and those funds weren't distributed down to build and support this work, so we were all supporting it in various different ways.
The CDC certainly knew about this. They gave out contracts. We have a contract from the CDC that started back in the fall to do more sequencing. But it takes time to build a program like this, and it's not just about money. We have the technology throughout the country to do it and we have the will to want to do it.
It's really more of the operational aspect of it, of getting the samples into the right place in the right format so you can do this at large scale, and then handling the data. It's not a simple positive or negative. There's a lot of data that goes with this. It's going to have to be analyzed in a coordinated way.
What we've learned by working with both the NBA and the NFL, doing the same thing with all the sequencing, is that the devil's in the details. Sequencing is easy. It's getting the samples there. It's getting the right data connected with the samples. It's getting them through the pipeline. It's having the data on the back end and interpreting it.
With the funds that are being distributed now, it's to create this program. It's not going to get us to 5% or so of the cases that we need to sequence across the country — the mark that some people have set as our standard of what we need to reach — but it's going to start to get us there. It's going to create the system to do it.
Once we have some more of this established, then it'll be much easier to increase capacity within it because we're going to have the framework.
More Pressure on Hospitals
Topol: I think it's important to note that the sample supply you had from the NBA and NFL put you in very good stead. The connection between sampling and labs is a big issue. Also, the reagents have been a bottleneck at times, even at the low level of sequencing that we're doing.
You touched on the E484K mutation, which is linked in South Africa as being part of the immune escape mosaic, if you will, but it's also been a mutation that's been found in many innocent variants. So it's hard to know when you have this combined with the UK variant, which is obviously one that's more transmissible and maybe more lethal, whether now you have the triple whammy, with immune escape.
I'd like to get your perspective of what we know now. Forget adding this unknown dimension of the recent fusion of a mutation and whether that adds— B.1.1.7, the UK variant, was bad enough.
What are we going to do now? Don't you think that's going to be the dominant strain in the United States over the weeks ahead?
Grubaugh: It's looking like it's trending that way, for sure. We're finding that with our data. We also recently posted a preprint showing the establishment of community transmissions throughout the country, and we see evidence for domestic spread, from New York, for example, into New Jersey, into Connecticut, into Michigan. The frequency is increasing.
Now, is it going to be a month and is it going to be dominant? It's hard to know exactly and it's going to be very heterogeneous across the country as well. It's certainly heading in that direction. Forget about E484K for a second. This could have very serious effects on our healthcare system.
I'm personally worried about my kids' schools. My 5-year-old, my kindergartener, wasn't in school the entire fall and just went back for the first time for in-person school in January. I'm already worried about what's going to happen as community transmission goes back up and we have to close that down again.
Even more so is the burden on the hospital system and what could happen there, and on top of that, the fatigue that we all have to go through again with another peak. I know that I'm tired. I know that my team's tired. I can't imagine what it's like for the folks in the hospitals who work these really long shifts — all the nightmarish things that they have seen over the past year. To put them through another spike again just seems really unbearable. There are many things we can do to prevent transmission. We shouldn't be arguing about whether or not B.1.1.7 will become more dominant or not. We should be doing these things just to help prevent that from happening.
Topol: You'd think that after what we've been through for over a year now that we'd be smart enough to know to plan for the worst.
But it looks like people are being lulled into the idea that everything's going down now. Cases are going down, hospitalizations. We're done. I think the cognoscenti know much better, but they can't affect behavior. Is that really kind of where we're stuck in this country? Is it that we just can't get people to be aligned to do the right thing?
Grubaugh: I think so. I really would like for the media to change their perspective around how they're covering variants. Right now, every interview request that I get is about how we're not doing enough sequencing. We've sequenced over 100,000 genomes in this country. It's an incredible feat. We haven't done enough, but it's hard to do.
The thing is, we know that the variants are here, and we know what we can do to slow the spread of them. There are some nuances in this that we might need to consider, like a longer isolation period or something, but they're still just really basic things like social distancing, mask wearing, and collectively doing things that we know prevent spread.
If you know that you have symptoms, don't go to the grocery store. If there is really high community prevalence, maybe we shouldn't have indoor dining and restaurants and bars. These are things that we know can work. If all of the media stories go from criticizing our surveillance system and put more toward emphasizing what works, it would be better: "Hey, we need to be working together. The end is near. There are vaccines. Yes, we wish they could be rolled out faster. We can all get there if we work really hard now." Nobody wants to see these spikes and overrun hospital systems again.
Verghese: Thank you for that. Speaking from the hospital perspective as a clinician, I think you're right. We're all quite burned out. We're also all fully expecting the surge by late March. But I'm hearing a sentiment from my colleagues — and I hope this is true — that maybe it will be different because we've managed, at least around here, to successfully vaccinate many individuals over 75 and over 65 years of age now.
What do both of you think? Might we have modified the picture any with the next surge?
Grubaugh: I would be really interested in Eric's take on this, so I'm going to let him go first.
Topol: I don't think we've put a dent in the vaccination yet to make a difference. It may be a small bit. But when you look at Israel and you see that they've countered B.1.1.7 so extraordinarily well, and then we're starting to see some of that in the UK for really advanced age, over 80... We are at a low level because 12% of our vaccinees in this country — many of those were healthcare workers and nursing home residents.
I don't know that we've had enough to even start to make a difference. We do have the background of having a massive spread of infections throughout the country. I'm not sure how much that's going to blunt this. I'm quite worried about it. Nathan, let's get your take.
Grubaugh: I am of similar mind that it would be great if we actually did vaccinate enough of our most vulnerable populations, where we can maybe escape part of it. There's still time to do this, right? There's still time to increase rollout. It's been so disappointing on how slow it's been.
But I understand, too, from the public health perspective here that the same individuals that are responsible for all the testing are now responsible for vaccine rollout. You put so much on a small, underfunded workforce to do all of these really important tasks, and we wonder why things aren't as fast as they could be.
Topol: I was going to ask you about that. The states that are leading the charge of vaccination are some of the most unlikely candidates, like West Virginia, the Dakotas, Alaska. Right up on the list of high performers is Connecticut. Is that because of what you're doing in New Haven? What's going on there?
Grubaugh: I am not actually sure how they're doing so well here. I know many folks in the health department. I know that there's a lot of coordination and plans to do it. I know that the folks that are running our New Haven health department are just fantastic. They put so much effort into this. It's just hard work and having a coordinated plan. But I think they would tell you as well that they wish they could be doing more.
Can Wastewater Surveillance Help?
Topol: It's been impressive to watch what you've achieved in Connecticut, in every respect. Now, another thing that we should be doing everywhere is wastewater surveillance. You said Omer and colleagues published a landmark paper in that regard, where you could tell a few days in advance whether an outbreak was starting to emerge.
Since that time, there have been papers that have come out saying that you could even pick up new strains and new variants. Are you doing that routinely in New Haven? Why aren't we doing this everywhere? Because it's so easy and cheap.
Grubaugh: First, I think that wastewater surveillance is just a fantastic supplement to clinical testing. We still obviously need the clinical testing. People need their results. I look at the wastewater data every day because I think it's so informative. There are some random blips that come in there because you get something that's more concentrated than not.
But if you look at the 7-day averages throughout, it's great. We have here at Yale, led by Jordan Peccia, with Saad helping to establish this. This is being done in six different catchment areas throughout the state. I think the number of times that I've heard this referenced randomly by people shows how much of an impact it's having.
Variant surveillance here is an interesting thing. I get asked about that often. Are we trying to sequence from wastewater to look for variants? Yes, but I will tell you that it's not an easy thing to say, "Yes, we're going to do this." Even though we can detect the virus in wastewater, it's not a large amount. This isn't clinical concentration level. It's usually down at those levels where, if I'm thinking about my background in intrahost sequencing, we look at population dynamics as they evolve within an individual, taking the same sort of mindset of this for wastewater. It's like, this is a population and we're trying to look for variants. The amount of virus there, if this was a clinical sample, I would say we're not even going to try this.
The other part of it is that it's wastewater. This RNA is degraded. It's not the easiest thing in the world to sequence from. When you combine these things and all the bottlenecks that happen in the lab — where you take a sample, and you take part of that sample, then a part of that sample, and a part of that sample — by the end, I'm not sure how many individuals that represents.
Even though this catchment area for wastewater serves 200,000 people or a million people, what you have in your sample might represent 200 people. But it can be done. There are ways around it. I'm hoping for some PCR methods that can help differentiate minor frequencies.
But so far, we haven't found the variants in the wastewater, even though we know they're in Connecticut. I think this is going to be a fantastic method once we really improve the method of tracking frequency changes over time. But I don't think it's the method to detect it when it's really rare.
Topol: Right. The take I get from you on this is that months from now, when we have this virus under relative control, hopefully by the summer, we're going to be wanting to know where it's cropping up — and in advance, before there's a significant outbreak.
Irrespective of being able to determine the variants, and you've given us some caveats, just knowing the titer of virus would be helpful.
Abraham, do you have wastewater surveillance in Palo Alto?
Verghese: I honestly don't know, Eric. I was just thinking of that. I don't have the answer to that.
Topol: We don't have it here. UC San Diego does, but only for their little network community. The idea that Nathan can look on his phone and see what's going on each day, and we're in the dark — are we prehistoric here? What's wrong with us? I just don't get this.
Grubaugh: It's set up in many different areas, but it's not nearly as widespread as it should be. I think some of it comes from the skill of collecting the samples and working with it in the lab. I sort of undersell, when I talk about this, the amount of work that Jordan Peccia's lab does just to extract RNA from these samples.
You get this big sample that you have to condense down and purify many times to get something that is a usable RNA sample for doing PCR. It's not like what we have in our lab, where we just get a tube of saliva, you pull some out of it, and you just do a basic RNA extraction. That's easy to do.
Wastewater is hard. One of the issues is that it takes investment to get something going. When there are just so many different things that you could be doing to help with the pandemic and few people who can do it, this is something that gets overlooked in some areas.
Less Invasive Testing?
Topol: Speaking of hard sampling, anyone who's gone through an NP swab — aka, brain biopsy — would be saying that they don't want to have that again or on a frequent basis. Your contribution with the saliva work has been transformative. Now all the meta-analyses surprisingly showed, for some of the naysayers, that saliva was as good or better.
Can you amplify about that? Why aren't we switching to saliva? Because still you have every place doing these deep nose swabs, and they're not exactly something that's user-friendly.
Grubaugh: Saliva has been amazingly useful. Akiko has a preprint showing that saliva titers are better correlates of disease outcome than these NP swabs are, which is really incredible. Hopefully we can get that validated in multiple locations and maybe it could actually be of more use in a clinical setting.
The saliva obviously has a lot of great advantages here. Take my 5-year-old, for example. She had to get a test. It was an anterior nares swab, so just at the base of the nose, not even the deep one. They did that to her once and she just freaked out and will not go again.
If I give her a tube and ask her to spit in it? She'll do it all day long.
The stories and conversations that I've had with other people about how they were afraid to get tested because of the NP swabs... And now with the saliva, they would be willing to do it. Why hasn't it become more widespread? I think one of the major limitations of using saliva is working with it in the lab.
Many of the labs would rather have the nasal swabs, which come in this nice media liquid where you just pull the swab out, you vortex it really quickly, and you use it. The tubes are set up for the systems. Our clinical labs are used to working with it because this is how we've done things with influenza and other respiratory pathogens for a long time.
Saliva takes a little bit more care. There is more variability within the sample so it's more work to do. I think that's been the major barrier. Even saying that, we're still seeing more and more saliva-based tests popping up. They're really popular for home tests or mail-in tests.
There's even some movement about using saliva for influenza testing. I think that we might see some carryover effects even after we are able to get out of the COVID pandemic, just for seasonal flu testing. It just takes time to change our laboratory testing systems to move away from things that are well established and make room for something like this.
Verghese: Going back to your thesis about the NBA and the B.1.1.7 infection being of prolonged duration, do you think that might have an effect on the nature of the neutralizing antibody production? Does it reflect the fact that it might not be as robust? Or might it be more robust at the very end? Are you collecting serum for that sort of stuff?
Grubaugh: To be perfectly honest, I don't get a lot of time to think very critically, even about our data. Everything is about generating data and putting together figures and making that available. Just the other day, with a neighbor, I had the longest conversation about virus evolution that I've had in a long time because it's more about do we find something? and not about how did this happen?
Why is it more prolonged and does that say something about their immune response to it? I think it's a good hypothesis, but I don't know. I'm going to guess that it's probably not due to neutralizing antibodies, from my very limited immunology background. I think that it has something more to do with its ability to infect cells.
Again, it might not be causing an increased viral load. There is laboratory evidence that suggests that B.1.1.7 has higher infectivity in cells and better binding to ACE2 receptors. Maybe it's just able to get to more places within the body and it lasts longer before it's cleared. Obviously, there's a lot that we have to learn there.
Verghese: I just wonder if the corollary to that might not be that this virus, if it becomes prevalent, might generate more robust immunity than we've been assuming for the other strains.
But What Does It Mean?
Topol: I like how you're looking at the sunny side, Abraham. That's good. One of the things that this brings up — and this has plagued the whole field of genomics from its origin — we're really good at sequencing, but we're not so good at functional assessment. And that's not just in pathogens.
Now here, this was a unique thing about the study that Nathan was involved in: You had this frequent sampling every day, where we never do that to understand the phenotype of the virus, right? We never do that, except in this rare circumstance. In fact, we can't even get these transmission studies of the vaccine effects because the companies, like Moderna, Pfizer, and all the rest of them, haven't really done the studies we need to do with daily NP swabs.
In addition to that, it's so hard to understand the functional consequences of the sequence changes. Nathan, how do you go about that? You wrote a classic paper on the D614G mutation in Cell because there, the question was, is that a more fit strain? It wasn't the very original SARS-CoV-2.
It was a change that became dominant. In the beginning, there was a group of people who said it's nothing. Then it's a little bit more fit. You cracked the case on that. Can you tell us about the phenotype, the assessment of the functional changes of a virus sequence?
Grubaugh: This is, of course, the hardest thing to do. For those of us doing a lot of sequencing, the observational data are relatively easy to get. We see things that are happening. Even going back to last February with Eddie Holmes and a PhD student in my lab, Mary Petrone, we wrote a paper on why we shouldn't worry when viruses mutate, which I regret the title of now.
The point of the paper still exists: that it's really difficult to determine the effects of mutations, the nontrivial amount of work to go into that. For most of what we study, when we're studying things not like this pandemic that affect everybody's lives, you have time to go through that and to do those assessments.
Here, we're in this really tough situation, where to understand the functional impact of a mutation, you need the observational data of the sequencing data itself to say, okay, we find that this is emerging. We can look at evolutionary pressures. We need the epidemiologic data.
For example, in this case with the NBA, we have daily testing. We understand way more than what we can actually reveal on that paper, too, about the epidemiology. We have these temporal testing components. You look at the UK, what they've done with B.1.1.7 and looking at secondary attack rate.
This serendipitous finding of the TaqPath assay, having a dropout with this variant, allowed you to look at the spread of it as well. Then you have to take that into an in vitro study. This is where the Jeremy Lubans and the Akikos of the world come into play to look at neutralization; to look at how this might affect the immune response in cells or maybe even in mice; to look at infectivity, growth kinetics, and all of these things.
The last part of it is that we need the clinical data. We need to have enough of this clinical data to make real assessments. It's not just one group that can do this to figure it out. It's so many different pieces of data. What you are seeing unfolding, for all of these different variants, is the assembly of all of the data to come up with an understanding of what's happening.
But what I have really learned over the past year about what public health is: It's making decisions with limited data. You're constantly being challenged because you don't know everything. But yet, you have to put something in place. And so we're constantly changing our minds; we're being influenced by the new data that are arising.
But we're having to put things in place and make recommendations up front on that. It's a really challenging thing to do.
Topol: We could go on for a few more hours because you're a wizard. We just love hearing your thoughts and your perspective. You've been groundbreaking in this pandemic, Nathan. We're so proud of your efforts. The fact that we could say that you were a postdoc with us at Scripps is something we like to brag about, seeing all your output.
I just want to reinforce what you've done at Yale as a model. Obviously you're collaborating with people all around the world, but you've assembled the parts and the sum is extraordinary. We're all watching that because if every university did that around the country and around the world, we would all get smarter faster.
One of our most memorable conversations we've had was with Akiko. We spoke with her early on, Abraham and I; I think we had immunology from 101 to 505. She's just lighting it up out there and so are you. We're really grateful because we know that this is 25-hours-a-day stuff that you're doing, you and your team.
Thank you for all of your efforts, and thanks to Anne and Howie Forman and every one of you. You're just amazing.
Verghese: Thank you so much, Nathan.
Grubaugh: Thank you. I really appreciate all the sentiments about that. I would say that one of the bright spots from this past year is to be able to work with many of these individuals and really have this strong connection with the School of Medicine and the hospital, the state epidemiologists and the local epidemiologists — many individuals that I never got to work with before.
These collaborations now, I think, are going to be really useful for anything going forward. We have this established. We know how to do it, and hopefully others can replicate this as well.
Topol: We sure hope so.
Eric J. Topol, MD, is one of the top 10 most cited researchers in medicine and frequently writes about technology in healthcare, including in his latest book, Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again.
Abraham Verghese, MD, is a critically acclaimed best-selling author and a physician with an international reputation for his focus on healing in an era when technology often overwhelms the human side of medicine.
Nathan Grubaugh, PhD, has been working on infectious diseases for more than a decade, with a focus on mosquito-borne viruses, including Zika, chikungunya, dengue, and West Nile. Even before COVID-19, his work was based on collaborations and a diverse team whose skills include field work, computational biology, and genomics.
Medscape © 2021 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: What the NBA's Collaboration With Yale Reveals About COVID Variants - Medscape - Feb 23, 2021.