Dr Pardis Sabeti - Sentinel

Dr Pardis Sabeti and her longtime collaborator, Christian Happi, built Sentinel on a simple premise: most “emerging” diseases aren’t emerging, they’re just not being tested for. We sat down with her to find out more about their team’s work on hunting viruses, what we’ve learned from Ebola outbreaks and COVID, and why the systems designed to stop the next outbreak are now under threat from cuts to global health funding.

A conversation about what’s at stake, but also what we already know works, and how science, technology and local networks could help us build a world that catches a pandemic before it becomes a crisis.

Okay, let me set the stage for your audience. 2014 was the heart and the heat of the Ebola outbreak in West Africa. In March of that year, an outbreak was declared in Guinea. Ebola hadn’t really been seen in West Africa before, and it was terrifying for a lot of people, because by the time anyone picked up that this mystery disease circulating was Ebola, it had already reached a large part of the country.

We were working in Sierra Leone at the time. My team at Harvard, our colleagues at Tulane, and partners in Nigeria and Sierra Leone were all working together on a different virus, Lassa, another hemorrhagic fever that circulates in that part of the world. We recognized that both countries were now at risk of Ebola, and that if it came, our hospitals would see cases. So we quickly set up diagnostic testing, and the teams we worked with on the ground were among the first to detect it in both Sierra Leone and Nigeria.

The more testing we set up, the more cases we identified, and we started to realize that “emerging disease” isn’t really the right idea. These aren’t emerging diseases. This is emergent detection of diseases that are already circulating. I think Donald Trump once said, if you don’t test for COVID, you don’t have it, and I think that’s actually quite poignant. It’s a true fact. If you don’t test for something, you never see it. As soon as you set up testing, you start seeing it.

That’s basically what we were doing in these hospitals, and we kept picking up more cases. People assume every Ebola patient is bleeding out of every orifice, this dramatic thing you see in Hollywood, because the only time anybody bothers to run an Ebola test is when that’s happening. You get what you look for. If you only look for Ebola in patients who look like that, you will only see Ebola in patients who look like that. Extending that logic, if you looked at more people, you’d find a lot of others walking around with Ebola who you’d never otherwise know about.

So you decide this is a problem worth solving. Where do you even start with something like that? Do you have to build something new, or plug into existing systems? What’s the balance between those two things?

We started small. We started with my Harvard start-up fund and a call to Christian Happi, my colleague in Nigeria. I was actually a human geneticist mainly, although I always worked in infectious disease. I’d developed an algorithm that could mine the human genome for signals of adaptation, the kind of thing you see with sickle cell anemia, which became common because having one copy of the trait protects you from malaria. A classic signal of adaptation.

I was studying that process, traits that help humans survive, and the strongest signal I found was a gene called LARGE, which is critical for the entry of Lassa virus. It was under evolutionary pressure in Nigeria, exactly where Lassa is found. That made me think: wait, what is this disease? Why have I never heard of it, even having gone to medical school? Why is it one of the major drivers of our evolution?

So I set out to ask a simple question: does having this trait protect you from Lassa? I didn’t know that much about the disease going in. But when we started working in Irrua, Nigeria, we realized this disease was everywhere. There were a huge number of cases, and none of it was being detected, because there were no diagnostics in the country. It wasn’t that there’d been some alert that Lassa was a problem. It was a weird signal in the human genome that made me ask whether this disease might be more important than anyone realized. It turned out that it was.

We went to this rural hospital in Irrua and helped them build capacity to test more cases and figure out what was going on. I spent seven years helping that community build testing infrastructure for Lassa before I ever started asking a single research question, because we had to get the fundamentals in place first. What we found was that the more capability we set up, the more cases we found. That hospital used to be described as “the cemetery,” because you’d go there and there was no diagnosis and no good treatment. It just witnessed what the disease did to you, without any way to intervene. Once we set up diagnostic testing and got people into early treatment, the hospital transformed.

We ended up writing a paper asking the question: is this an emerging disease, or emergent detection of a disease that was already circulating? Our argument was that if we helped local hospitals build the capability to test for what’s circulating in their region, we wouldn’t just help those communities and the doctors on the front lines, we’d create sentinel sites for the whole world. That was the premise, and it started small: local communities, clinicians on the front lines, meeting them where they were with the needs they actually had. A much bigger understanding of the problem grew out of that.

This different view of disease response, one that’s about setting up a surveillance network and catching the signal before an outbreak happens, rather than just responding to disease, how much of your hunch turned out to be correct? What ended up surprising you?

Honestly, nothing. I’m trying to think of what we didn’t know, and we actually knew everything. That’s the part that’s hard to explain, because people ask, how did you know COVID was coming? And I say, because you’re in the field, and it’s obvious. It’s a matter of time before one strain gets away and goes out of control. We wrote a paper in 2012 saying Ebola was circulating all over West and Central Africa. There was obvious evidence for it. You just had to look.

So none of it was actually a surprise, and that’s the frustrating part, because we see all of it and we know exactly what needs to be done. We can see that another outbreak, and a major pandemic, is coming, because it’s obvious. We’re in an environment where these viruses have more opportunities to jump from human to human than at any point in history, at the same time as we’re cutting surveillance and countermeasure capacity. I recently connected with a researcher at the University of Michigan, Jocelyn Ackerman, who’s been telling me fascinating things about how humans behave when they’re sick, how we hide it from each other because we trust each other less. So we’re also going to be walking around hiding our own infections. It’s just a matter of time. None of it is really a surprise.

We need to start working now to deal with that, because otherwise we’ll only decide to act again once we’re in the middle of another pandemic. And this time it won’t be one that half the population can write off. It’ll be one we all have to contend with, because it could be fatal to major swaths of the population.

So you saw this problem and decided to do something about it. In 2020, you launched Sentinel through the Audacious Project. The timing was extraordinary, launching in February 2020, before most people had heard of COVID. Take us from that moment to now. What does Sentinel look like in 2026? What have you built?

What’s interesting is that the first time we got significant funding around this premise, an early-warning system for pandemics, was in March 2014. We had the thing in early March, and by late March we were setting up diagnostics for Ebola, and we were in it.

Then we got the Audacious Project prize. We submitted the application in December 2019 and got it in February 2020. By March, we were in the COVID pandemic. I remember thinking, uh oh.

You must be cursed when it comes to grants.

Yeah, yeah. Well, now we’ve got this MacArthur prize, and I have to say, I feel like there’s something brewing again. We don’t know what yet, but it’s going to sideline us. So watch out if we ever get a grant, we might be in trouble. The other way I describe it: you can’t always get what you want, but sometimes you get what you need.

Sentinel is really the product of about two decades of work between Christian Happi and me, on Lassa and malaria before this. The premise is simple: detect, connect, empower. You need to detect pathogens wherever they occur, whether that’s characterizing them in genome centers or through at-home testing for circulating threats. And you need everyone with a phone able to say, I see a cluster of symptoms, or I have these symptoms, and to have that information get them what they need.

Then you need to connect all of that, because infectious disease is communicable. It spreads person to person, village to village. You have to connect the data for it to mean anything. Finding out that three people who went to the same party each got sick somewhere else doesn’t get you anywhere on its own. But once you see a cluster, that’s where you start to get the intelligence that lets you respond quickly.

Something I say a lot: unlike other existential threats, where you can tell someone they should vote, but if they don’t live in a swing state, it’s hard to convince them it matters, pathogens spread exponentially. One case becomes two, becomes four, becomes eight. Because of that, a single person can launch a pandemic, and that means every person can also help stop one. Every person truly matters. There’s no way around it. One person missed and you’re in a world of danger. So within that, we think we have to empower every actor in the system with the information and the tools they need to stop a pandemic. A lot of what we do is built around education.

I love that. This has worked, right? I think Sentinel played a fairly large role, not the only role, but a real one, in reducing the impact of the Marburg outbreak in Rwanda. Tell us about that. How did you use it? Why did it work?

The biggest thing that went right is that we were there at the service of Rwanda. That’s the thing we always do. People ask how we work well in different countries, and I say, we just ask them what they need from us. We never come in saying “we’ve got this.” That’s not what we do. We find the terrific people in the country already leading the charge, and we give them the resources to do what they’re doing as well as possible.

Christian’s team in Nigeria are the ones who roll out to the different sites and work directly with partners, for the most part. We go there too, but their team leads. They went on the ground, supported the country’s sequencing efforts, helped establish reporting, and worked closely with Rwandan teams on how to stop the outbreak, identify cases, and understand transmission. What was great was that they stopped transmission quickly and got a very low case fatality rate, because they were able to identify cases fast and get patients treated before they could spread the disease further.

The current outbreak in the DRC is now the fastest-growing Ebola outbreak ever, and I think it’s either become, or is very close to becoming, the second-largest outbreak on record. The UN and WHO are warning that if it continues at this pace, it could become the largest. So what went wrong? Why didn’t we catch this one? How does it go right somewhere like Rwanda, but here we’ve got a real problem on our hands?

Honestly, it’s a perfect storm. In regions with conflict, there’s a dual problem. Conflict creates all the circumstances that let an outbreak escalate: displaced populations, poor sanitation, chaos, the decay of healthcare systems. All of that makes an outbreak more likely to happen, and also more likely to go undetected. As the world seems to be moving toward a constant drumbeat of conflict, we’re going to have to figure out how to build systems that can weather that and still stop outbreaks from escalating.

So the biggest reason we’re where we are is the conflict in the DRC. But on top of that, there’s the loss of global health funding, the removal of peacekeepers, and the continual difficulty of funding science, vaccines, and diagnostics. It’s a real perfect storm, and it’s hard to see a way out without real international cooperation, which isn’t in great supply right now.

People intuitively understand how hard it is to provide health services in a conflict zone. What’s maybe less obvious is the slashing of foreign aid and global health funding that started about eighteen months ago, and we’re really starting to see that come through now. What does that actually look like on the ground? What did it look like specifically for Sentinel?

So much disappeared from the network at once: frontline testing, health workers, peacekeepers. I had colleagues who had worked very successfully in the DRC for many years. Now there are kidnappings, much greater risk. There’s a sense that there’s no longer an international force protecting the region, so you’ve got a lot more gangs, warfare, and terrorist activity. A lot of people lost the funding to work there, and a lot of people who still had funding could no longer safely do the work. Honestly, it’s no surprise this happened, given how many people I know who either withdrew or had to withdraw because of what was going on.

When you strip that funding away, which is exactly what’s happened, what’s still standing on its own? What does the response actually look like on the ground in the DRC right now?

Honestly, I don’t have the best sense of it, and I’d rather be upfront about what I know and don’t know. I’m not with the WHO, so I don’t have visibility outside our own group. What I can tell you is that our own team has been sending people to the front lines in rotation, supporting the teams already there. But there’s a lot more friction and restriction than there used to be. Our teams in Nigeria used to go to the ground without a second thought, and our US teams could join them. Now every part of that is harder.

You mentioned this is something that’s in the news, that everyone knows about. Is it, though? When was the last time Ebola was on the front page? For what we’re looking at, it’s really not being followed that closely. There’s a lot other news competing for attention, and there’s a lot we still don’t know about how this is going to play out.

I’m curious what the slashing of global health funding has meant for you more broadly, for the Sentinel network. This Ebola outbreak is serious and worrying, but it’ll eventually be contained. I imagine what actually keeps you up at night is a serious global pandemic, another one on the scale of COVID, but with a very different profile. What does your funding environment look like right now, and how have the cuts affected you over the last eighteen months?

We’d just gotten this big prize from Audacious, a five-year project. TED works with us closely on sustainable funding for the longer term, so things don’t just go out overnight, and at that point we had a broad, diverse portfolio that would let us keep going once the Audacious funding was completed. Then the cuts came, and it wasn’t just cuts to current funding, it was a lot of things on the horizon that got withdrawn entirely. Suddenly we were on the side of a cliff with no way out. A lot of philanthropists were watching, but everyone seemed paralyzed, trying to figure out what was going on, so there wasn’t much movement.

In the middle of that, I had to do one of the hardest things I’ve ever done in managing my lab. There came a point where there was no funding, and no cut I could make, that would get us there, so I had to lay off a third of my own team. I went to other agencies asking for help, and they’d say, we just laid off 80 percent of our own team. When everyone you go to for help has already cut most of their team, there’s nothing left to ask for.

What I told each person I had to let go was: I don’t want you to think the training and the background you got here isn’t valuable. It is, even if we don’t have the funding for it right now. And sadly, you’re in a growth industry, because we’re breeding pandemics all around us, and at some point it’s all going to come home to roost, and you’ll be back in action.

Then I went back to the rest of the team. Even at the start of all this, I’d told everyone: I don’t know what’s going to happen, so let’s get our papers out and finish our work, because we don’t know what tomorrow brings. Let’s live every day like it’s our last, professionally. We were already doing that. Then I did the same thing again with the team that was left: all things must go. We wrapped up a lot of work and got it published.

I wasn’t sure what would happen next. We were finalists for the MacArthur 100&Change prize, out of more than 860 applications, with one winner. We were up against unbelievable talent and unbelievable organizations, and honestly didn’t think we’d get it. It was a shot in the dark, maybe a twenty percent chance. Then, several months later, we got the prize, and our fortunes completely changed.

Going into that meeting, which was billed as a “status update,” I assumed that name meant they were about to say thank you, but no thank you. So I’d prepared a concession speech. My colleague Christian was much more optimistic. He said he’d prepare the acceptance speech, and I said, go ahead, I’ve got the concession speech covered. When the moment came, he was so shocked he froze, so I ended up giving my concession speech anyway. Which turned out to be strangely fitting, because it was really about knowing what it’s like to live with uncertainty, to be a finalist and not get it, because that’s how I’d been living.

I was given this unbelievable, almost karmic, gift, and I have a karmic debt to pay off now. We have to figure out how to help the whole community, because I know there are people out there thinking, why you? And honestly, I don’t know, why me? So we have to make sure it’s clear that this is good for everybody, and figure out how to make that true.

That’s a strange kind of philanthropic survivor’s guilt. I haven’t heard that framed quite like that before. What was the size of the grant, and why has it changed your fortunes so dramatically?

A hundred million dollars, to change something meaningful in the world. It’s obviously the largest grant I’ve received, or will receive, in my lifetime. And having it arrive right as I was watching everything crumble was almost more than I could process. I was low for a while afterward, honestly. People ask how you celebrate something like that. I carry things. Some people talk about having FOMO, fear of missing out. I have it in reverse: fear of others missing out. I know a hundred million dollars is a lot, but we’re working across five countries, and when you look at what we’re actually trying to do, it’s not going to replace the public health system. So you have to think hard about how you make things better with it.

What’s now possible? What does the next four years look like? What capability do you hope to have by 2030? I mean, “stopgap” undersells it, this is pretty significant, and hopefully enough to tide you over until there’s a real shift in attitude and money starts moving again.

The committees reviewing us always asked some version of: what’s the point of this money, given this is a multi-billion, even trillion-dollar problem? How could it possibly make a difference? Our answer is that reducing cost is the big thing we work on. We believe governments should put billions toward biosecurity and infectious disease, but they may never do that, so the way to make the problem manageable is to reduce the cost of everything in the system: using information intelligently so you test the right people for the right things, instead of running hundreds of millions of tests on healthy people. Smarter, cheaper testing.

Then you reduce time. You make the data systems work faster, so intelligence gets out quicker and outbreaks get stopped before they escalate exponentially. Everything we’re doing right now is about reducing cost and time in outbreak response. There are others working in this space too, and it matters to us to be collaborative. We don’t want to turn anyone down as a partner. We want to help however we can and support good work broadly. I think we’ll see real progress there.

At the same time, biosecurity risk is escalating out of control. I think bioweapons are going to be a real thing. We’re going to see agricultural attacks, human attacks. We’ve moved into a new kind of warfare, no holds barred, anything goes, the ends justify the means. We have to be ready for that. I think we’re moving at a good clip, but we’re fighting a rising tide, so we have to be deliberate and careful in every part of it.

What needs to be true by 2030 for Sentinel to still be the answer to this problem? What conditions does this work actually depend on?

We have to build tools that meaningfully change response time, and get that technology into the hands of more people. That’s really what we’re working toward: high-quality technology built around the frontline worker and responder, understanding where they’re coming from, making sure it actually works for them.

The bigger thing we need to do is change the culture of outbreak response itself, the way people interact. It doesn’t matter what technology we build if we don’t build a community that’s actually working together to stop outbreaks. I think one of the great tragedies around the lab-leak theory, and I have my own opinions, but I don’t think it’s worth weighing in publicly on whether I believe it was natural or leaked, is this: whatever happened, nobody wanted it to leak, if it did leak. In the worst case, it was a mistake that got out of control. But the way we responded, as if we needed to interrogate everyone to the ground, destroy people, treat everyone from China as guilty, has left us in a scenario where, regardless of whether we believe this outbreak was natural or created, we’ve moved from a world where a bioweapon was impossible, to possible, to plausible, to probable. We’re now in a world where it could actually be done. Whether anyone did do it is a separate question, but that it could be done, by any of us, even a high school student in a village somewhere, that’s the real issue.

So why are we vilifying and fighting each other when we all have that capability? Isn’t that exactly the point where diplomacy has to become the number one priority? Shouldn’t we have gone straight to something like truth and reconciliation? It doesn’t matter who’s at fault. We need to get back to a place where we’re all working together.

At the end of the day, we’re all humans on one planet. Nothing short of an alien invasion would unite humanity the way a deadly virus that could kill all of us should. If we care about the human race and its future, we need to stop the escalation, stop the arms race, and figure out how to work together.

My personal hope is that it doesn’t take a global pandemic for people to start cooperating with each other again. But I understand you’re describing a very real possibility, and that you and a lot of other people are working on a response that heads this off, or at least catches it early enough. That gives me some reassurance, even with all the question marks still hanging over it. Do you still want to be doing this in ten or fifteen years? Is this a problem that will ever actually be solved?

During COVID, I almost stepped away, because by nature I’m curious, and I’ve got three other research projects that have nothing to do with any of this. I started in human genetics, did a lot of work in anthropology, went to medical school. I run a whole cognitive science arm of my lab. I’m not really an infectious disease researcher, that’s not where my career started. I think of myself as a student of humanity, someone who likes using science and technology to make things better for people.

During COVID, I thought, okay, now everyone in the field has got this, everyone recognizes the problem, I can step back. I’d done something similar once before: my entire early career was in human genetics, nobody knew me for infectious disease at all, but once the Human Genome Project wrapped up and the floodgates opened, there was a lot of gold left to mine, and I thought, someone else can mine that gold, let me go find a new patch. That’s kind of how I live. Once something’s being mined properly, I go look for the next thing.

I thought that had happened here too. But then I saw the dysfunction in how we were relating to each other and responding, and decided I needed to spend more time on this. That said, I’m always thinking about succession planning. I’m always looking for the next big thing. We have an incredible team, and part of why empowerment matters so much to me is that I want to build the next generation of leaders who do this without needing me. With my postdocs, the people who train in the lab and go on to become professors themselves, I’m always telling them, take this piece, it’s yours now. I can’t hand things off fast enough, honestly.

But I’ll stay in it as long as the problem needs something I can give, and as long as I feel like I’m actually contributing. I’ve got other side quests coming. But most of it, for me, comes back to using science to make humanity better, and asking why we’re doing what we’re doing in the first place.

I love that answer. That’s a pretty good place to end it, unless there’s anything else we missed?

We could go on, you’re clearly very curious and ask great questions, and I could talk for a long time, but I’ll end it there. As I always say, none of this work is ever really finished, but this is as good a stopping point as any.

That’s great. Pardis, thank you so much. We really appreciate everything you’re doing, and we’ll be keeping tabs. Hopefully we won’t see you in the news for the wrong reasons.

That would be great. That would be great, yeah. My biggest success would be that you never have to hear from me again.