Ep 5: Please don't kill the bats! (with Barbara Han)
How do diseases spread from animals to humans? Is it possible to forecast where disease outbreaks will occur and when they will blow up into major health crises?
Tune into this podcast to hear Marty and Art talk to Barbara Han about how we track infectious diseases and whether we'll ever be able to predict outbreaks. Han studies the conditions leading to disease outbreaks in humans at the Cary Institute of Ecosystem Studies in New York. Who knows, her research on forecasting disease outbreaks may even help predict and stop a zombie pandemic (think World War Z type scenario) from ever starting.
Follow Barbara on Twitter: @bahanbug
-
AW = Art Woods
MM = Marty Martin
BH = Barbara Han
MM: I'm Marty Martin.
AW: And I'm Art Woods.
MM: Welcome to the Big Biology podcast. Today we're talking with Barbara Han, a disease ecologist at the Cary Institute of Ecosystem Studies. Barbara studies how infectious diseases affect the behavior, ecology, and evolution of their hosts, what factors make some species and places especially prone to harboring the bad guys, and how parasites sometimes spill over from wildlife to affect us. Today we talked to her about such zoonotic diseases.
AW: Recall that we do two versions of each conversation. This is the longer, more complete version of our conversation with Barbara, and it runs almost an hour long. If you'd like to hear a much shorter version, check out our website bigbiology.org, or your feed on iTunes or Google Play.
MM: Barbara, it's exciting to have you on the show today, thanks for joining us.
BH: Hi Marty! Yeah, thanks, thanks for having me!
AW: We had a series of questions about your work on diseases from other animals, and I just wanted to first ask you about what kinds of diseases come from other animals, and just how that process happens.
BH: Yeah, that's a great question. So, the majority of diseases that we consider to be what we call emerging in humans right now come from animals, and that can include things that we've heard about recently, like Ebola or Zika virus, things we have in the United States pretty commonly like plague, those are all examples of diseases that have, that live in animals and kind of persist in animals but spill over into humans.
AW: Great, yeah. And so, you, you guys had a set of terms that you use for this, so, zoonosis, zoonoses, zoonotic diseases. So, what, what does that mean?
BH: Oh yeah. We love our terms. It basically, zoo means animal, so a zoonotic is just an animal-borne disease, that's all it is.
AW: Okay, great.
BH: It's our special word.
AW: Awesome. So, so, how, how does it happen? How does a disease jump from another species into, into humans?
BH: Yeah, well, jump is kind of an... I don't know, I don't know if I like that term. Marty, you probably have opinions about this too, but, but you know, we call it spillover, we call it jumping. But really what happens is there are these events where the, the microbe or whatever is causing that disease you know, gets into an animal that it's not intended to live in, right? So, a pathogen like Yersinia bacteria that causes plague, it'll get into a human and then 99 times out of 100 probably won't do anything, like it's not in it, in the host that it, it wants to be in, it's not adapted for that host, and so most of the time when a pathogen ends up in a different host, nothing happens. But sometimes, sometimes, that pathogen will mutate a little bit or, or adapt a little bit, and with repeated spillover events, that adaptation eventually catches, and then that pathogen can now suddenly infect a human, and then cause, cause damage in a human. And so, that spillover process is actually like, it's a repeated event that happens over a long period of time, and then eventually, you know, a disease will, will catch in a human and stick.
02:58
AW: Is, is that why you don't like the term jump, because it's not like a single, you know, big leap across a chasm, it's more just like a repeated thing that statistically in the end is going to infect humans? Is that the idea?
BH: Yeah. And there's a lot of, you know, there's a lot of, there's a lot of chance involved and you know, probability, and so, jumping kind of makes me think of like okay, I'm going to prepare to jump and I'm going to make this decision and then I'm going to jump and I'm going to make it. And that's definitely not how it works, and spillover too kind of gives me this idea of like, you know, you keep filling up a water glass, filling it up, filling it up, and then eventually when you're like so full of pathogens like something spills over into like the thing that's standing next to it, and, and that's also kind of, I don't know, it's not exactly the right picture, but they get the general idea across I guess, so yeah.
MM: Alright, so this is our chance here to just coin a brand-new word. If you don't like jump and you don't like spillover, do you have the perfect term that you've never laid on anybody yet?
BH: Oh no, that's a hard one. No, I don't have one. I guess we're going to stick with those two because we don't have an alternative.
MM: Okay, well spillover is a, it's a neat idea. There's a cover of a, a cover and title of a really cool book...
BH: Yeah, David Quammen.
MM: ...that attracted a whole lot of readers, you know, it's a great resource for this topic too.
AW: So, I just want to circle back to something you said just a couple of minutes ago. You were talking about the, the microbial diversity and microbes mutating, and perhaps evolving in a way that allows them to get into, to human hosts. Is there also something happening from the human side where there's some people that are more susceptible or more compromised in some way, and those people can be doorways into, you know, the human species for, for these diseases?
BH: Yeah, that's a great question. I mean, I think the easy answer is of course, you know, of course there's... I mean we can all think of that person in our lives that's constantly sick, right? Or if you have a small child and they're in daycare I mean, all parents have this. We've all experienced this trauma. But I think that, that there's also this really interesting behavioral, these behavioral patterns too, right? People who have a lot of contact with, with hunting, you know, if people who were hunters, whether you're in Africa or in North America, you're, you're interacting with wildlife and the landscape in a really different way, and some of that interaction can give you a higher risk of acquiring pathogens that are carried by those species that you interact with. So, yeah, I think that there's a lot of diversity in the kinds of people that are maybe doorways as you put it, to, to pathogens spilling over and establishing in humans. That can be behavioral, that can be you know, just your sort of innate susceptibility to picking things up.
05:32
AW: And, and what, what kinds of interactions are higher risk, you know, versus lower risk?
BH: I mean, I think it comes down to contact most of all. If you are in a profession or if you're, you live in an area where you're sort of intimately connected with animals that, with other animals, so, if you live in the, in the rainforest for example, or if you're a subsistence bush meat hunter, if you're a livestock farmer, if you're sort of farming ducks and pigs, and also you have fruit bats hanging over your fruit trees and that kind of stuff, if you're sort of in this mix of lots of animals interacting with each other, that can give you a different kind of a risk profile than somebody who's like a Wall Street banker and has a very sterile sort of lifestyle like, you know what I mean? They're two very different ends of a continuum, and it's not necessarily that the banker's going to have less exposure to, to disease, but certainly less exposure to spillover.
AW: Right. And, and potentially less exposure to sort of the blood, guts, and you know, out there cutting up their own meat, that kind of stuff, so, yeah.
BH: Yeah.
MM: One would hope.
AW: Alright.
06:35
MM: So, Barbara though, I mean you said context, but sometimes things are not transmitted person to person or into the environment to be picked up. What about the, what happens when it's, when it's vectors?
BH: Yeah!
MM: And everybody, everybody says I'm the, I'm the guy to have it, the cookout, because the mosquitoes are always attacking me, and not biting everybody else because of it. Is there any, anything on...
BH: Ooh, are you one of those people Marty?
AW: You can come to my next cookout.
MM: I am one of those people, my poor daughter is as well.
BH: So, I get to stand next to you when I'm in the rainforest?
MM: Yeah exactly, that's the only reason I get invited to parties, but it's something.
BH: Yeah, yeah. Vectors, vectors, insect vectors like ticks and mosquitoes just add a whole new dimension to the whole picture, right? So, even if you're not in, in direct contact with something like a primate, like few of us are really in this country, but you know, if you live in sort of an urban environment in Central and South America let's say, but you are regularly seeing monkeys that kind of go into the forest and come out, and, and are sort of commensal with humans, living in the environment where you are, there are mosquitoes that sort of hang out and opportunistically feed on whatever they can get their mouth parts into, right? Anything that's warm, that's around, that's in high numbers. And so, a mosquito that can bite a primate that you're actually never going to be in direct contact with can then finish off its blood meal by biting you, and that could lead to a transmission event. That's how a lot of these zoonotic vector-borne viruses have, have emerged into humans in the past.
AW: So, so, I just also had a question about the overall frequency of, of zoonotic diseases, you know? We obviously get a lot of things from other, other people, but like, how, how important altogether are zoonotic diseases compared to you know, other stuff that's already established in humans? How many people every year get, get zoonotic diseases?
BH: Gosh, I don't know about the human count, that's a good question. I mean, I assume that, I mean if you kind of just look at the, the charts, things like diarrheal diseases sort of are chart toppers in terms of how many people they infect, and I think that, you know, diarrhea is caused by lots of things, including things that are, that have an animal origin, and things that, that don't, that are just sort of in the environment all the time. So, I think it's, that is a difficult question to answer directly, but I think the general sense is that out of the things that kind of scare us, right? So, the things that are increasing in frequency, increasing in number in humans over time, the latest research suggests that the majority of those things are animal-borne. And the majority of the things that are, that are emerging in humans actually come from mammals. So, there's this increasing interest in you know, what is it about mammals? Is it, what is it besides the fact that they're sort of more closely related to us than things like fish or frogs? Yeah, so that's where a lot of the research buzz is these days.
MM: Yeah, I guess, let's, let's talk about the mammal side of things that terrify everybody and make their dream, their worst nightmares come true, that everything that do worry about...
BH: I know. I'm all sunshine and rainbows at parties when people ask me, "what do you do" and I'm like, do you really want to know?
MM: So, I'll suck up the mosquitoes, and then you can tell people horror stories, and everybody would be happy. So, so what about the bad guys? Among all the mammals, what are the species to worry most about? You know, take that any direction you want in terms of where on the planet, or the types of things, or you know, I want to get to all of it anyway.
09:51
BH: Yeah. Well, so, there's a couple of different answers, ways to answer that question, right? Like, there are certain places in the world where maybe statistically you might be more likely to come into contact with something that's bad for you. So, southeast Asia for example is sort of lit up on the map in terms of how many zoonoses have emerged from that region in the recent past and how many animal species that we know of right now to carry zoonotic diseases, there's a high diversity of those "reservoir species" that hang out in the same region of southeast Asia. But also if you just sort of think about the mammals as a group, I think there's been some consensus building over the past couple of years that bats are something to be worried about as a group, and that's because a lot of pretty virulent zoonotic viruses have emerged from the bats. And then rodents have sort of been an age-old, you know, they have a bad rap for carrying lots of diseases and I mean, that's not without evidence, right? There's, they actually carry more than, than, than many other groups. And that's because they're so species rich I mean, there's over 2200 species of rodents so, if you kind of just statistically speaking like, there's more of them, there's going to be more things that they carry, and in fact I think the most recent number is something like 13 different zoonotic diseases can be carried by one single species of rodent.
MM: Oh wow! Do you want to tell us where that lives so everybody can immediately move out of that area?
BH: It's one of the rats. Yeah, it's one of the rats, and I think it's one of the ones that's really widely distributed, so I don't know if there's any escaping it really.
MM: Oh boy, what is it, the Norvegicus, the Norway rat, or the black rat, something like that?
BH: I think it might be Norvegicus, yeah. I think it, that is Norvegicus, yeah.
MM: Oh excellent, great, great, the rat that we all have around.
BH: Oh yeah, our favorite.
MM: Wow. So, what is it about these critters that you think makes them so dangerous for us? I mean, with rodents, I know I've heard stories that it is the long history that we've had with them, and lots of time our, our sort of lives overlap them, especially things like the black rat. But in general, what is it that makes these guys, this taxa, so problematic?
11:51
BH: Yeah, I mean I think there's, the, that's one question that we tried to answer with some data-driven approaches, right? We won't have to kind of look at all of the rodents together and say, well, we know that some of these are really problematic, we all know which ones those are, and then there's other that we don't have any information about or they just don't seem to carry anything and they're kind of just as common as the other one, so what is it about the ones that are bad that make them special or different from the other ones? And when we did that analysis what we found was that if you just look at their traits, right? Just things like their behavior or what they eat or sort of where they live, their social structure, things like that, things that are sort of intrinsic about them, the picture that emerged was something that has a fast life history pace, like a fast lifestyle. So, you can, you can think about rodents as you know, some rodents have lots of babies starting early on. They don't live very long, but they certainly pump out lots of babies in a very short amount of time. They have faster metabolisms, they might be smaller in size, and that compared to something like, you know, a [word?], or something that's much bigger that doesn't live as quickly, doesn't have nearly as many offspring per year, they might gestate for longer, kind of spend more resources growing a bigger baby. So, those are kind of two ends of the continuum, and what our analysis highlighted was that the species that carry a lot of diseases, or the species that carry zoonotic diseases at all are more likely to be those that are fast living. And there, you know, I think the mechanism there is sort of, the jury's still out on that. There are some hypotheses that need to be followed up with empirical tests, experiments that are controlled and sort of run in the field and in the lab. And, and Marty actually, you've done some of the groundbreaking work on this, kind of thinking about immune profiles, and how does immunity differ when you consider rodents that kind of live in, in areas that are really seasonal, versus places that aren't so seasonal. Like, rodents that have to deal with a lot of environmental stress, their immune profiles are going to be really different, right? The strategy that they have evolved, and they adapt, and use to fight off these infections that they're kind of encountering all the time is going to be really different from something that's slow living, that takes its time, that's investing its resources in, in living a long life. And you can imagine that those immune strategies would be really different, right?
MM: Right, right. So, what about the, the possibility that with something like a rodent where, you know, the average mom is having 12 to 15 different rodent babies, that's plenty more opportunities for the, the pathogens to get in, right? Lots more opportunities to infect those susceptible, I mean that's part of the story too, right?
BH: Yeah, definitely. Yeah, their population density is a big, is a, is a, is something to consider, right? So, if there's more of you, and it could be that things with higher population density also kind of hang out in groups, so there is some added risk there in terms of if one of you gets infected, will all of that whole group get infected? So, there's a bit of a density effect there for sure.
14:51
AW: I just wanted to ask, does the pace of life idea apply to bats too, or is something else going on with the bats?
BH: Not so much. So, for bats we haven't yet completed the analysis where we just kind of look at all bats as a group and ask that very general question of how many things do you carry and what makes you different. What we did for the bats was we were interested in the bats mostly when... We asked a more specific question I guess I should say, and that was when the Ebola outbreak was happening in West Africa and we thought, well, gosh it's been 40 years and we still don't know what the reservoir is, and how do you even find the reservoir when there's, I mean that's a biodiversity hotspot, right? The Congo basin is just, I mean, talk about a needle in a haystack, like, you have to be in the right place at the right time and get the right animal almost, like the right individual to really find the smoking gun, to find the reservoir species. And, and that just seems like an impossibly difficult task and so we thought, well, to try and narrow the boundaries there of what's possible, maybe we'll just start with the bats. And you know, there's 1100, over 1100 species, maybe we'll just take the ones that have come back zero positive for filovirus, for any, any of the viruses that are grouped together with Ebola, and that includes things like Marburg virus, which also infects humans and causes a hemorrhagic fever, and then things that aren't so, that aren't infectious to humans but are closely related, so Reston virus and things like that, right? So, we took all of those filoviruses and looked at the bats and said, okay, well we know some of you guys are positive for these things and the majority of you aren't, so what separates the positives from the, from the we don't knows? And when we did that analysis, we, we found that it wasn't... So, the, the story that was coming out in the media was that you know, it's the fruit bats, it's the fruit bats. But actually, we didn't find that signature at all. It was actually equally split, fruit bats and insectivorous bats, and there wasn't, there wasn't that profile of like, it's the fast living bat, because really, all bats are slow. For their body size, they don't have very, there's, they have one baby per year basically, and it doesn't matter what species you are. Some populations of some species have, they can squeeze out like one extra. Like so, it works out to be like 1.5 per year. And those do tend to be a little bit more likely to be filovirus positive than the other ones, but yeah, we didn't see that signature. We did find a very strong signature of bats that had, so, within the range of the geographic range of a bat that carries filovirus, you're much more likely to carry filovirus if you overlap with lots of other mammal species. So, if your mammal diversity is super high in your range, then you are much more likely to be filovirus positive. And interestingly, that was the total opposite pattern that we found for rodents. So, for rodents, rodents that had really few, like really species poor ranges, were much more likely to carry zoonotic diseases. So, that, I think that corroborated a couple of things for me. When you think about rodents, you think, oh they're so good at living, you know sort of, in the nooks and crannies of the environments that humans create, and so that allows them, for us to have such high contact with them. And so, it makes sense, when humans move in, we sort of like, de-faunate everything, and lots of things die out or get hunted away, or we destroy their habitats, so the diversity really drops, and the things that are left over are the rodents. And so, it kind of makes sense that they would carry more diseases than the other animals. And in contrast, the bats that carry filoviruses, we've always sort of known intuitively that, that Ebola virus is hanging out in a rainforest, these mega-diverse regions that have lots of, lots of species and not only that, but these species are sort of moving in and out, tracking resources that change with the seasons, and so, your chances of finding them are, finding the reservoir are really low, and the bat range covers that area where there's tons of species, lots of, lots of richness in, in many ways, and it's sort of I guess highlighted for us that we're sort of picking up the signal that lots of other scientists have picked up in the past that this is a forest-dwelling pathogen. We just need to figure out which species is carrying it.
18:58
MM: So, to back up to the, the [word?] that you're talking about Barbara, when we were talking about the Ebola in bats tending to be, you know, more likely to show up for filovirus, and they show up in places with communities more diverse, yet in rodents it tends to be the other. Is that specific to a particular parasite for the rodents? Because I remember, you know, one of the things I was reading that in rodents, I think they have most of the zoonotic bacteria and viruses, was that right? Like, it was another paper that you were, you put out. So, I'm just wondering whether disparities also might have something to do with the types of parasites we're talking about.
BH: Yeah. No, that's, that's a great question, and we don't know the answer to that yet. I think that, I suspect that that's true, there's got to be something there with the, you know, the proportion of different types of pathogens you carry, so if you were like a rodent that carries lots of worms but carries fewer viruses, versus a rodent that carries tons of viruses and bacteria but, but fewer worms, I mean, what does that mean for your ability to transmit something to humans? I think that those patterns are, are worth following up on, and I don't know of anyone who has done that yet.
AW: So, I think one thing you didn't talk about was this aspect of bats, some species roosting together in very large numbers, so, so, but not all bat species do that, so, is there a correlation between those that aggregate and those that are big disease reservoirs?
BH: Yes, yeah. For filoviruses, you're much more likely to carry them if you hang out in really dense population groups. So, if you're hanging out with lots of other members of your group roosting in these huge roosts, then you're much more likely to be positive.
AW: And does, that must arise from these diseases kind of sweeping through big groups of, of bats. Are there just repeated big die-offs of bats that live in big groups like this, and you know, we're looking at the sort of long evolutionary outcome of the, those that survive but still carry the disease?
BH: Yeah, I know for a wildlife pathogen. So, there's this fungus called white nose syndrome that causes a disease called white nose syndrome in the bats that does do exactly that. It sweeps through a population, kills a bunch of them, It'll basically knock out an entire cave of bats in one season. But those bats are sort of hibernating together. So, they're sort of experiencing a different metabolic, like they're in a different phase of their life, right? They're sort of, they're hibernating, so their metabolic rate is super low, they're kind of checked out for the winter, whereas filoviruses are tropical. So, the species that are carrying these filoviruses and roosting in these big population numbers, they, they're not, they're not hibernating, they're, they might do some sort of diurnal torpor where they match their background temperature at night, but they're not truly hibernating. So, the biological processes there are similar, but they're different. Like, we don't have any examples I don't think of bat zoonotic diseases that have wiped out a bat colony. At least I can't think of any, do you know of any, Marty?
MM: I can't think of any either, no. I think it's more the other way around, what the bats carry don't really seem to matter to the bats most of the time, but make our lives pretty miserable, yeah.
BH: Yeah, yeah, so that's another question, right? Like, well what is it about, like why are bats able to carry these things that are crazy virulent to humans? Like it just, and I think that jury's still out on whether bats really do carry the things that are super bad to humans, or if it's just a perception or a sampling issue. I think we're sort of narrowing in on that answer, but, but yeah, bats are certainly getting a bad reputation, which I don't know, I don't know if it's deserved but, but they've got it.
22:36
MM: So, how much do we know about that, that side of it Barbara, with regard to the why, you know, things that really cause massive problems for us don't matter for, for much of, for many other species? It's just so weird to think of bats flying around with these viruses circulating, and somehow everything is okay for these guys, but the first time that they poop on fruit that the pig eats, and then the person ends up eating or coming into contact, it's all bad. How does that work?
BH: Yeah, yeah. Gosh, I don't know. Bats are, ugh poor bats, bats are such awesome creatures. I mean, I think we know so little like basic information about bats. So, bats are so weird, right? They're heterothermic, which basically means they're kind of in between like a cold-blooded animal and a warm-blooded animal, because some of them hibernate like cold, you know, because some of them hibernate, and they always match their background temperatures, even if they're in the tropics. So, they do this weird torpor thing. But then they fly, so they have, have to have really high metabolic rates when they're flying, and then, but they, and they're small, but they live as if they're huge animals. They only have one baby a year. So, they're like very, that's a very unique group of animals. And I think that, I don't know if it's because they're hard to catch, or because it was always sort of a niche interest for biologists, but they're just really understudied compared to some of these other animal groups, right? And so, we don't know basic information about their immune profiles, like you know, what, what immune cells do they have, like how does, what is their immune system, how does it function? There's almost no comparative immunology comparing different species of bats against each other in terms of their immune, immunity. So, I mean, I think these questions are kind of fundamental if we want to understand why is it that bats don't seem to suffer any ill consequences, no pun intended.
MM: That was good, that was good!
BH: You know? And I, I'm like sort of, I'm like kind of on the edge of my seat, like looking at the literature going, who's going to do the experiments that it takes to figure out what's going on here with the bats, but I think it's, we have a long journey ahead of us. There's a lot of unanswered questions there for sure.
24:38
MM: Do we have other, other groups of mammals that are that way? I mean, with 2000 species of rodents, we know a lot about a couple of domesticated critters, but I mean, the rodents are part of that, but what about other species? I'm thinking the shrews, they probably don't get a lot of attention either, right?
BH: Yes! Yeah, I don't know why nobody wants to study shrews, they're so cool! But like the insectivores in general, we did this big sort of dumb analysis, it wasn't dumb I don't want to call it dumb, but we basically just counted, right, the number of zoonotic diseases that each of these groups carry, and we found that like, given how many insectivores there are in the world, they actually, a high, a large proportion of them carried zoonotic diseases. And so, if you kind of do the, if you extend that line across those few points you have data for and you say, well we have so many other species that we have yet to sample, chances are that insectivores are actually carrying lots of things that, that are, that could possibly infect humans, but we have, they're so understudied. I mean they, a lot of them are underground, so that makes it a little harder, but...
MM: Tougher, sure.
BH: Yeah, yeah.
25:40
AW: So, so, if you stumbled on, say a species of shrew that had never been studied and you wanted to figure out if it was carrying potential zoonotic diseases, how would you even tell that? Like, like, what...
MM: You just ask it, don't you?
AW: ...what if it carried diseases that, you know, we didn't even recognize as diseases but yet they would be transmissable under the right circumstances?
BH: Yeah, I think, I mean, there are metagenomics approaches now that you can take to kind of like take a tissue sample, or a fecal sample and just kind of, you know, sequence whatever, everything that's in it, and kind of match it up with things that are known pathogens to humans. So, there are more sophisticated, there have been technological advances that allow us to do some of that now, but I mean, you kind of also in some sense, if you want to be specific about it, you kind of have to know what you're looking for. And therein lies the difficulty, right? If you know that you are looking for the reservoir of Ebola, then you can look for Ebola, and you can be really specific about whether you've found Ebola or not. But if you're looking at an animal and you're saying, I want to know everything that you carry, that's a little bit trickier of a question to answer, although we're, we're kind of getting close to it now.
AW: So, so you can tell from the DNA sequences that it has microbes that are potentially pathogenic to human?
BH: Yeah, you can. Yeah, and a lot of these viral surveillance group studies now just kind of go out and do exactly that where they'll take the, they'll sample as many different animal types as they can and sequence what they've got in them and match them up with known viruses, and then you'll see all these papers come out that say, oh we found this virus that was like a 60% match with a human herpes virus or something in a bat, and what does that mean? I don't know, but it's worth knowing, I guess, right? So, I mean, I think that those kinds of studies are worth doing because it's always worth knowing what's in your environment. But getting from what's in the animal that's new to whether it's potentially pathogenic to humans is a more difficult question to answer for sure.
27:36
MM: So, I think we'd be irresponsible if we didn't ask you right now, with the two hot diseases still on most people's radar, where's the thought process for where things are coming from, Ebola and Zika especially? Who are the conspicuous critters right now?
BH: Oh, for you mean like carriers for Ebola? Well, so there was, now I can't remember the name of the bat... But recently, so, right after the Ebola crisis was sort of starting to wind down a bit, we published this paper that showed the results of our machine learning analysis where we were like trying to figure out what the other species were, right, that should carry filovirus, and we made predications like, you know, here are the species in rank order, at the top of the list is the one that should carry it like, you know, we have, it's in the 90th percentile, it should carry it, and then you know, so on down the list. So, we have kind of like our top 10 or our top 12. And so, we published those results, and then like six months later, I stumbled on this paper that found that number 5 on our list was found positive for filovirus. And it wasn't, it wasn't like a known filovirus. This is a brand-new filovirus. So, the, which was, I mean I don't know whether to cheer or be like oh no. Because on the one hand you're like, oh our algorithm's working, it's right, like our analysis is like, it was onto something, but then on the other hand you're like, oh dang it, like this is, this is exactly what we thought, this is awful, this is bad news. And the thing about this filovirus is that they found in the lungs, like it was a lung tropism so, which, you know, makes everybody worried that it could actually be aerosolizable, and what does that mean for transmission of this filovirus, and all those kinds of scary questions. And it was in China, it was in southeast Asia. So, yeah, which is unexpected.
AW: So, I want to jump in here. So, you mentioned this, this process that you use to rank these species. You mentioned machine learning. So, can you explain what machine learning is...
BH: Oh yeah, I said the "M" word.
AW: ...in sort of simple language and tell us how you use it?
29:31
BH: Yeah. So, machine learning, well you guys know what machine learning is, right? Like, you shop on Amazon and listen to Spotify and get your data collected by unknown entities, and you know, get ads on your websites for things that you want?
MM: Evil Mark Zuckerberg.
BH: That's right, but that's all machine learning, basically. So, what all machine learning does, at least as far as the work that I do is that it learns patterns from data, and then it uses those patterns to predict things that I want to know. So, in the case of our work, we train these algorithms on data that describe disease reservoirs, carriers of Ebola virus, or rodents that carry lots of zoonotic diseases and we say, okay algorithm, take the data that I give you and learn what a good reservoir looks like. And then, I'm going to give you the rest of the species that you know, we haven't studied yet, and I want you to pick out the ones that look just like the known reservoirs. And the, the traits that I give it to learn on are things that are really intrinsic and differentiate one species from another. So, things that describe differences in their biology, or their ecology, or their distribution in the world, what they eat, their behavior, things like that. So, yeah. It would be like if you were shopping for something on Amazon and you're like, you know, I really want blue socks and so you type in socks, and then you, and then you know, there's all these different features of socks and you give it all those features, and Amazon eventually will like narrow in on the things that, that you want to buy based on your search terms or your, your purchasing history and things like that.
AW: So, so, once you train an algorithm on a particular data set, you know, to find a particular kind of disease, can you then go in and look at that algorithm and figure out what it's actually focusing on? Like look into the brain of the machine that's been doing the learning. Or is it, does it remain a black box?
BH: Yeah, no you know, I think, I don't know if I like this term black box, because it implies that it's like unknowable, right? But in our case, it's, that's not true, so, okay. So, well maybe it, maybe it is a little bit true. So, the way that the algorithm works, it's very like, it's like a recipe, right? So, it's not, it's not like a, it's not mysterious how it's working. Like, it's a very specific set of rules that it's following to arrive at this thing, this pattern that it finds in the data. And in terms of like after you, on the, on the other end of it, so you feed this machine a whole bunch of data and it does its thing in obscurity, and then on the other end you get these like very crisp predictions that it makes, and it has like a nice probability that assigns each number. Well the way that it assigns those probabilities is by looking at the traits. And so, if you tell, if you ask the algorithm, okay well I want you to give me the profile of like, what are the traits that were the most important for you to make the decision that this species is at the top instead of at the bottom? You can get that profile out, and the machine, I mean, you can, you can get those summaries from the algorithm, yeah. So, in some sense you do learn, like the algorithm is learning, is using the traits to learn which ones are disease reservoirs, and then it will also, you can also have plots, you know, make plots that show like, here are the features that were the most important for that algorithm being accurate, and being able to classify these things in two different groups. It's pretty fun.
32:44
MM: I mean, I'm an ecologist and you're an ecologist and on some level this is kind of a silly question, but if this is so bad, I think, or sort of, we have some sense of where the diseases are coming from, species and places. In terms of the species, why don't we just go out and eradicate them all?
BH: Oh my gosh, that would be like the worst possible thing.
MM: Do you get, do you get this one?
BH: I like, this is what keeps me up at night, right? So, when we were doing this Ebola paper I thought, you know what's going to happen is I'm going to publish this list and then people are just going to start at the top and start killing off all the bats one by one. It was a huge fear of mine, and so, I think that there's a balance that needs to be struck here, right? When we perturb the environment, we sort of poke it and prod it and push on it, and eventually you're going to, I don't know, I don't know if this is a good analogy, but annoy it, and then something's going to spill over. You're going to, you're going to perturb the balance there, and I think that that, this mentality of like, okay the easy fix is just to kill all of the primates that could be carrying Zika, or kill all of the bats that could be carrying Ebola, you're just going to perturb the system in a way that you're probably going to make the viral titer shoot way up and it's going to make everything bad for everybody, everything worse for everybody, right? So, yeah, I don't, I would not recommend that as a, as a realistic strategy. I think it's...
MM: Not a good strategy, okay, okay alright we'll take that off the list.
BH: ...probably cheaper and better for everybody to, to do a better job of managing our, our relationship with the environment.
AW: Would there be alternative ways of depressing the, the microbial titers in these species that didn't involve killing them, so going out and dosing them, or giving them vaccines, or something like that?
BH: Yeah, absolutely. So, yeah, that's my favorite, right? I don't know why we don't have animal vaccines, like, I mean I'm sure there are pretty important reasons why, I don't know, it's less expensive than developing human vaccines, so in that sense it seems like a no brainer that you know, if we're going to go through animal trials to get on the way to a human trial, but we can kind of do it in a smart way to target the species that we think are actually carrying these viruses, then why not just use that virus that that vaccine at that stage and vaccinate the animal that's the actual problem and then prevent the spillover process. So, that's one, I don't know if anybody's really...well, so, for rabies, we do that for rabies.
MM: Right.
BH: But I think we should do it for lots of other animals too. I mean I think, personally I think we should develop a primate vaccine for Zika.
35:08
AW: So, so, you mentioned pets, and this maybe brings up the broader class of just domesticated animals in general. So, how important are they in transmitting zoonotic diseases and are they a greater risk than wild animals?
BH: Oh gosh. Yeah, so greater risk than wild animals...I think some in terms of contact. So, I guess it depends, right? So, if you're, if you are in very close contact with livestock, and you have livestock that like pastures outside and are sort of sharing environments with wild animals, and the pathogen transfer there with the wildlife, the greater wild, is going to be higher, and you're kind of in intimate contact with your livestock and so, the chances of contacting a pathogen and picking something up are higher, but yeah, I, in terms of the relative numbers of diseases carried by livestock versus everything else, like everything else definitely wins. There are definitely more diseases that are not livestock based diseases than, than the other way around. But, yeah, I don't, I think that it all comes down to contact. Your risk is going to be some function of like your probability of getting infected and your contact rate. So, if you're constantly contacting something that isn't probably infected with anything but you're contacting them a lot, like your dog, then your chances of picking up something from your dog are probably like a lot higher than if, yeah if you're like outside in the wild and there's lots of animals that you're in very low contact with but everything carries something, right? So, it's a matter of how you look at it.
36:35
MM: So this, this brings up something, it brings something to mind that we really haven't talked about that I think it would be interesting to sort of just you know, get your perspective on. There's a difference between carrying around the bug and being able to transmit the bug and actually being sick from the bug, right? I mean, everything that we're talking about so far, sometimes we're emphasizing one thing and sometimes we're going the other way. I mean, do you want to, do you want to speak to what we're doing, or how you can kind of incorporate that into thinking about and managing disease?
BH: Yeah, we have to be really careful, I think. I mean, I think as ecologists we try to be careful in the literature and you know, like making more precise the question that we're trying to answer. But when we talk the wider public, I think those terms get really confused. So, like you can talk about an animal that is a reservoir, like a true reservoir is something that is an animal that can allow that pathogen to persist even if humans were not part of the picture. So, if there was no human in sight and you had that animal and that parasite, then that parasite and that host species would persist as long as that host was around. So, that's like, and the reservoir is also able to then, if there is a human around, facilitate the transmission of the pathogen from that animal to a human. So, that's the definition of a true reservoir. But then there are these weird in-betweens like you have animals that get sick, that get super sick from a pathogen that is technically zoonotic because it's carried by an animal, but it's not really a reservoir because it's getting sick and dying just like humans would, right? So like, Ebola is an example of that, great apes, great apes get really sick and die from Ebola, actually at rates that far surpass human mortality. I mean like, ninety something percent mortality in great apes when they get infected with Ebola. But we wouldn't consider them a reservoir because honestly like, it's so bad, it's so lethal to great apes that Ebola could not persist in that population, so that can't be a reservoir. Yellow fever is another example. There's lots of monkeys that sort of drop out of the trees, fall down dead from yellow fever, which is a zoonotic, and it does persist in other monkey species, but so, we have to be careful, like the term reservoir is different from the term host, which is different from lots of other terms. There's lots of ways you can slice that pie up. But what we're really worried about is the species that not only carries it, but carries it for long enough that it can pass it on. And I think from a public health sort of disease management perspective, that's really the thing that we care about. Of course as ecologists we care more about, you know, also conserving wildlife, conserving biodiversity, and you know, improving human health and, and reducing that transmission. And if you're thinking about that you think, okay well, what do we need to do to manage great apes better so that they're not infected with Ebola so that we can conserve their populations, and we don't get the spillover risk to humans from you know, that hunt them. Yeah, so that's a, it's sort a big mixed bag. We have to be careful about how we talk about these things. There are really important biological nuances there.
39:30
MM: Alright, so, we talked about species predominantly, but Art and I, we've, we were talking about this before we had you on. We really wanted to talk about Typhoid Mary, you know, and the possibility that there are lots of typhoid [word?]. So, do you want to talk about who that was, or your favorite other patient zero? And, you know...
BH: Oh, Typhoid Mary. Yeah, she's a special case, because she was kind of a character too, right? She was like belligerent and like would not be confined, like people could tell that she was, there's something about her that was a problem, and they kept trying to like, keep her from, from spreading, from cooking for people, and then killing them eventually. But she just wouldn't, she wouldn't have any of it. So, yeah there's lots of special things about Typhoid Mary. But yeah, she is the, she is the quintessential patient zero for, for infectious diseases. So, the idea of super spreaders I think is really interesting. The idea that like a single individual could, could account for the vast majority of, of a number of people that eventually end up getting infected, right? So, it's this idea that like, either behaviorally or, I think mostly behaviorally. There's something about that individual that gives it more connections to lots of other people. And if you can identify who that super spreader is early on in the process, then by breaking the chains of transmission around that person, then you might minimize the impact of that infection on the whole population. So, that's really what the idea was with you know, with Typhoid Mary and the example of the super spreader, like if, if we could have been more successful like keeping Typhoid Mary from other people early on, typhoid probably wouldn't have been as big of a, or have killed as many people.
AW: So, on this topic, so, in general, would you say that in most species there are individuals who act as super spreaders? Is that a general thing, or is that a, like a human specific thing?
BH: Yeah, I don't know if that's a general thing. I mean, I do think that there are super spreaders in animals and in humans. I think it is a thing in both groups, but, and I do think that there's enough individual level variation in behavior that is completely not being studied, right? So, we tend to study diseases in populations, and we think about the transmission processes being between two individuals, but it's almost like to go from, from understanding transmission as a concept where two things have to come into contact and then something gets transmitted onto that second individual, you have to really consider like the behaviors of those two people, of those two individuals. And that's really hard to do, because it, because it requires a lot of observation and data, and to go out into the wild and to observe something and say, okay I know you're infected, without like poking you and touching you, and like, short of you like glowing red when you're infected with something, there's just no way to, you know, there's no way to know if somebody's infected or not. And then to, to watch that one individual contact like ten individuals, and then count how many of those got infected, like you can't know if those people were infected or not. So, it's...
42:40
AW: Starting to sound impossible.
BH: Yeah, yeah. It's pretty difficult. I mean, folks have gotten close to doing it before with very controlled systems, but I think it's a very general phenomenon in behavioral ecology, or in disease ecology, in infectious disease in general that's really hard to study and is not very well understood, but it's a huge source of variation that I think we need to do better at quantifying.
AW: Yeah. So, so, Barbara, I want you to look into your crystal ball and tell us, are these, are zoonotic diseases going to be a bigger deal in the future, and if so why, and what can we do about it?
BH: I mean, I think that they are proving to be a big deal now, and the recent trends suggest that they are an increasing problem. So, yeah we need to, we need to get better at dealing with them. I think we already have made great strides in doing that, in improving our ability to deal with them. I mean the first step is just recognizing that they are an increasing issue, which I think was a matter of controversy for a little while, but I think, you know, Kate Smith at Brown and others have done really great studies to really kind of nail home the point that like, yeah it's actually increasing, it's not an artifact of us looking harder, it's not an artifact of the internet. It is actually that the numbers of new diseases that are emerging in humans is increasing over time. So, that's point one. So, yeah. I think zoonoses are going to be a bigger problem moving forward. And then, and then you wanted to know what to do about it?
AW: Well, actually before we get to that, can you just say why, why is it? Why, is it because human populations are expanding and we're getting into new ecosystems that we've not spent much time in before? Is that...?
BH: Yeah, yeah. I mean, I think the human, the global society's more connected, so, you know, you've seen those maps where there's just a bajillion lines crisscrossing the globe, and the amount of airport traffic and stuff and so, things that you know, originate in one part of the country can make it to the other side of the country, or the country or the world in, within 24 hours. So, connectedness is huge. The amount of contact between humans and animals is increasing as we encroach into their habitats, and more industrialization, things like that. So, I mean, there's a, there's a slew of different reasons why we think that zoonotic diseases are increasing, and I think that they're all partly true.
AW: Okay, so, so, is there anything to do about it? Because you know, a lot of the factors you just described are things that describe good things, right? Like it's great that I can get on a plane and fly coast to coast in one day, but, and we're not going to stop that obviously, so, so, what can we do about the spread of these diseases?
45:10
BH: Yeah, I think that so, our vaccine development and our, well, let me just back up and zoom out a little bit and say like, our response times to emerging crises has gotten faster. So, I think that's one thing that has come with experience, right? We get burned by something like Ebola, and we're going to take a really hard look at how we responded and how to improve that in the future. And so, I think with, with multiple of these kinds of exercises in modern times, we have become faster and better adapted at deploying resources and people when we need to. But I think the other thing that we need to really work at to be better prepared for future zoonotic diseases is to really kind of quantify the risk, right? So, what we can never do, right, is say, I'm not going to be able to pick a species and pick a spot on the globe and say, this new undescribed pathogen is going to emerge from this rodent species in this city in 2025. I mean, that's like crystal ball stuff, that's never going to happen. But what we can do is say like, okay well if we're interested in figuring out what virus is going to emerge, we should look at what viruses have already emerged, and which animal groups have the most viruses, and where those are distributed. And by gathering those types of baselines, you can start to kind of draw boundaries around the possibility of, like what that risk space looks like. And you're never going to be able to quantify all of the risk, like you're never going to be able to do that finger pointing exercise that I just said, but you can at least say like, well if I have five animal groups and these are the most, you know, these are the five groups that I think are the most, the most likely to carry viruses, well, the majority of the animals that carry viruses that we know of are in this country in southeast Asia. So, maybe we should start looking there first and figure out what our risk is there. I mean, I think it's these exercises, this perspective of we can never tell that, you know, fortune tell in that way, but we can, we can do a lot better at quantifying the risk, and the better we do at quantifying risk, more steps ahead we are able to take in terms of heading things off before they turn into these fire situations where you know like, something emerges on the landscape and now it's a wildfire before you even realize that it's happening. Like, we should kind of be building up like horizon scanning a little bit better, I guess, in terms of surveillance and vigilance.
47:25
AW: Yeah. So, so, you implied that there's kind of a, like a disease SWAT team that can jump into action when, when these new things appear on the landscape. So, who are these people or organizations, and you know, what do they do? How do they operate?
BH: I mean, I think public health, ministries of public health in every country, that's like sort of the first line, right, first line of defense. Or I guess first line of defense is like clinicians that like notice if something weird is up with their patient and they report it on things like Pro Med [?]. I mean like, shockingly like, Pro Med is sort of on a volunteer and it's kind of a, it's not like people pay into it. It's a, it's a service that's run as a non-profit, and it's, it's a huge service to the public health industry, like, like globally. Like, everybody reads Pro Med. And so, the little sort of like telltale signs that something is brewing in that far corner of the world, like a lot of those reports come through Pro Med mail. It's just like a list serve that people get where clinicians can report weird cases, or... And so that's kind of, that's one example of a way that communication has really improved our ability to stay connected and gather information in a timely way. But also like, when things get kind of, when things blow up a little bit, the World Health Organization, that's like their responsibility to kind of jump in and make decisions that are hard, and assess the risks, and figure out what the supply chain's going to be, and... But they're also extraordinarily underfunded, I mean, we're supposed to all as governments pay into it, but I think only 20% of the member states pay in to WHO like, as they're supposed to. And so, what happens when there's a huge outbreak that threatens, you know, threatens as an international concern? Like, it's not like they're not going to respond. They're going to respond, and they're going to kind of throw everything they've got in a smart way at the problem, but it's kind of, it's hard to sustain, and it's sort of a global good that we need to be aware of and pay into, but, but not everybody does, and so I think there's a resource limitation there for sure.
49:18
MM: I want to, I want to, Art uses this terminology 20,000 feet I think you call it?
AW: The 20,000 foot view, yeah.
MM: You fly low in the jumbo jet. 20,000 foot view. So, if you had, I mean, let's hear, this podcast, the title is Big Biology. Let's hear your craziest big picture prediction or appeal for study in you know, research dollars or, I mean, what do you think the best way forward is in disease ecology research?
BH: You know what, I think we could knock out a bunch of, I mean, disease ecology research is, it's ongoing, it's this huge enterprise, it's not a huge enterprise, it's actually a small enterprise. There's not that many of us doing it, right? Relatively speaking.
MM: Not nearly enough.
BH: Not nearly enough, yeah. Especially in, especially now. But, I think that we're kind of on the cusp of being able to, to develop something that's like, like something similar to weather forecasting, right? Where if you think about weather forecasting, you're taking all of these data streams and you're putting them together into these models that give you with a fair degree of accuracy, there is still some inaccuracy around, and you know, stochasticity as we call it in ecology, I mean yeah, these random factors you know, that kind of throw off your prediction a little bit, but, but the fact that we can pretty much say the chances of rain are 70%, that gives us a little bit of leeway. That gives us some leetime, right? It's going to influence my decisions about what I'm going to do that day, it's an influential thing to have weather forecasting. And I think that we're close to being able to do something like that for infectious disease. We're close to being able to take a bunch of, I mean, we're going it now, right? We're taking a bunch of data streams that already exist, we're cleaning them up, training an algorithm to recognize things that should be a threat, and that should inform how we survey those things in the future, and then what we do about management. I mean, that pipeline is just one fraction of what we could do with the data that we already have available, and I think that it would be worthy of our investment as a global society to be able to really formalize that and make it a forecasting system.
MM: That sounds fantastic, although the nightly news is going to be totally different. You get your weather forecast and you're a little depressed, and then you get the disease forecast and it's all bad.
BH: Yeah, it's going to turn all of us into total hypochondriacs, it's going to be awful. It's going to be all my fault.
MM: All your fault, good job, how dare you.
BH: Well, that's my big idea, that's my dream.
MM: Well thank you so much for doing this! I hope it wasn't too painful.
BH: Oh no, it was super fun! I always like talking to you, and I think this is a cool project, I'm really excited for the podcast to start.
MM: Okay, bye Barbara, thanks!
AW: Thanks Barbara!
51:53
MM: Thanks to Matt Blois for production help, to web master Steve Lane and [name?], and to Haley Hanson for editorial help. For more episodes, check out our website bigbiology.org. See you next time!