The story that I'm going
to tell you today,

for me, began back in 2006.

That was when I first heard
about an outbreak of mysterious illness

that was happening in the Amazon
rainforest of Peru.

The people that were getting sick
from this illness,

they had horrifying symptoms, nightmarish.

They had unbelievable headaches,

they couldn't eat or drink.

Some of them were even hallucinating --

confused and aggressive.

The most tragic part of all

was that many of the victims
were children.

And of all of those that got sick,

none survived.

It turned out that what was killing
people was a virus,

but it wasn't Ebola, it wasn't Zika,

it wasn't even some new virus
never before seen by science.

These people were dying
of an ancient killer,

one that we've known about for centuries.

They were dying of rabies.

And what all of them had in common
was that as they slept,

they'd all been bitten by the only mammal
that lives exclusively on a diet of blood:

the vampire bat.

These sorts of outbreaks
that jump from bats into people,

they've become more and more common
in the last couple of decades.

In 2003, it was SARS.

It showed up in Chinese animal markets
and spread globally.

That virus, like the one from Peru,
was eventually traced back to bats,

which have probably harbored it,
undetected, for centuries.

Then, 10 years later, we see Ebola
showing up in West Africa,

and that surprised just about everybody

because, according
to the science at the time,

Ebola wasn't really supposed
to be in West Africa.

That ended up causing the largest
and most widespread Ebola outbreak

in history.

So there's a disturbing trend here, right?

Deadly viruses are appearing in places
where we can't really expect them,

and as a global health community,

we're caught on our heels.

We're constantly chasing
after the next viral emergency

in this perpetual cycle,

always trying to extinguish epidemics
after they've already started.

So with new diseases appearing every year,

now is really the time

that we need to start thinking
about what we can do about it.

If we just wait for the next
Ebola to happen,

we might not be so lucky next time.

We might face a different virus,

one that's more deadly,

one that spreads better among people,

or maybe one that just completely
outwits our vaccines,

leaving us defenseless.

So can we anticipate pandemics?

Can we stop them?

Those are really hard questions to answer,

and the reason is that the pandemics --

the ones that spread globally,

the ones that we really
want to anticipate --

they're actually really rare events.

And for us as a species
that is a good thing --

that's why we're all here.

But from a scientific standpoint,
it's a little bit of a problem.

That's because if something
happens just once or twice,

that's really not enough
to find any patterns.

Patterns that could tell us when
or where the next pandemic might strike.

So what do we do?

Well, I think one of the solutions
we may have is to study some viruses

that routinely jump from wild
animals into people,

or into our pets, or our livestock,

even if they're not the same viruses

that we think are going
to cause pandemics.

If we can use
those everyday killer viruses

to work out some of the patterns

of what drives that initial, crucial jump
from one species to the next,

and, potentially, how we might stop it,

then we're going to end up better prepared

for those viruses that jump
between species more rarely

but pose a greater threat of pandemics.

Now, rabies, as terrible as it is,

turns out to be a pretty nice
virus in this case.

You see, rabies is a scary, deadly virus.

It has 100 percent fatality.

That means if you get infected with rabies
and you don't get treated early,

there's nothing that can be done.

There is no cure.

You will die.

And rabies is not just
a problem of the past either.

Even today, rabies still kills
50 to 60,000 people every year.

Just put that number in some perspective.

Imagine the whole West African
Ebola outbreak --

about two-and-a-half years;

you condense all the people
that died in that outbreak

into just a single year.

That's pretty bad.

But then, you multiply it by four,

and that's what happens
with rabies every single year.

So what sets rabies apart
from a virus like Ebola

is that when people get it,

they tend not to spread it onward.

That means that every single time
a person gets rabies,

it's because they were bitten
by a rabid animal,

and usually, that's a dog or a bat.

But it also means that those jumps
between species,

which are so important to understand,
but so rare for most viruses,

for rabies, they're actually
happening by the thousands.

So in a way, rabies
is almost like the fruit fly

or the lab mouse of deadly viruses.

This is a virus that we can use
and study to find patterns

and potentially test out new solutions.

And so, when I first heard
about that outbreak of rabies

in the Peruvian Amazon,

it struck me as something
potentially powerful

because this was a virus that was jumping
from bats into other animals

often enough that we might
be able to anticipate it ...

Maybe even stop it.

So as a first-year graduate student

with a vague memory
of my high school Spanish class,

I jumped onto a plane
and flew off to Peru,

looking for vampire bats.

And the first couple of years
of this project were really tough.

I had no shortage of ambitious plans
to rid Latin America of rabies,

but at the same time,

there seemed to be an equally endless
supply of mudslides and flat tires,

power outages, stomach bugs
all stopping me.

But that was kind of par for the course,

working in South America,

and to me, it was part of the adventure.

But what kept me going

was the knowledge that for the first time,

the work that I was doing
might actually have some real impact

on people's lives in the short term.

And that struck me the most

when we actually went out to the Amazon

and were trying to catch vampire bats.

You see, all we had to do was show up
at a village and ask around.

"Who's been getting bitten
by a bat lately?"

And people raised their hands,

because in these communities,

getting bitten by a bat
is an everyday occurrence,

happens every day.

And so all we had to do
was go to the right house,

open up a net

and show up at night,

and wait until the bats tried
to fly in and feed on human blood.

So to me, seeing a child with a bite wound
on his head or blood stains on his sheets,

that was more than enough motivation

to get past whatever logistical
or physical headache

I happened to be feeling on that day.

Since we were working
all night long, though,

I had plenty of time to think about
how I might actually solve this problem,

and it stood out to me
that there were two burning questions.

The first was that we know
that people are bitten all the time,

but rabies outbreaks
aren't happening all the time --

every couple of years,
maybe even every decade,

you get a rabies outbreak.

So if we could somehow anticipate
when and where the next outbreak would be,

that would be a real opportunity,

meaning we could vaccinate
people ahead of time,

before anybody starts dying.

But the other side of that coin

is that vaccination
is really just a Band-Aid.

It's kind of a strategy of damage control.

Of course it's lifesaving and important
and we have to do it,

but at the end of the day,

no matter how many cows,
how many people we vaccinate,

we're still going to have exactly the same
amount of rabies up there in the bats.

The actual risk of getting bitten
hasn't changed at all.

So my second question was this:

Could we somehow
cut the virus off at its source?

If we could somehow reduce the amount
of rabies in the bats themselves,

then that would be a real game changer.

We'd been talking about shifting

from a strategy of damage control
to one based on prevention.

So, how do we begin to do that?

Well, the first thing
we needed to understand

was how this virus actually works
in its natural host --

in the bats.

And that is a tall order
for any infectious disease,

particularly one in a reclusive
species like bats,

but we had to start somewhere.

So the way we started
was looking at some historical data.

When and where had these outbreaks
happened in the past?

And it became clear
that rabies was a virus

that just had to be on the move.

It couldn't sit still.

The virus might circulate in one area
for a year, maybe two,

but unless it found a new group of bats
to infect somewhere else,

it was pretty much bound to go extinct.

So with that, we solved one key part
of the rabies transmission challenge.

We knew we were dealing
with a virus on the move,

but we still couldn't say
where it was going.

Essentially, what I wanted was
more of a Google Maps-style prediction,

which is, "What's
the destination of the virus?

What's the route it's going
to take to get there?

How fast will it move?"

To do that, I turned
to the genomes of rabies.

You see, rabies, like many other viruses,
has a tiny little genome,

but one that evolves
really, really quickly.

So quickly that by the time the virus
has moved from one point to the next,

it's going to have picked up
a couple of new mutations.

And so all we have to do
is kind of connect the dots

across an evolutionary tree,

and that's going to tell us
where the virus has been in the past

and how it spread across the landscape.

So, I went out and I collected cow brains,

because that's where
you get rabies viruses.

And from genome sequences that we got
from the viruses in those cow brains,

I was able to work out

that this is a virus that spreads
between 10 and 20 miles each year.

OK, so that means we do now have
the speed limit of the virus,

but still missing that other key part
of where is it going in the first place.

For that, I needed to think
a little bit more like a bat,

because rabies is a virus --

it doesn't move by itself,

it has to be moved around by its bat host,

so I needed to think about
how far to fly and how often to fly.

My imagination didn't get me
all that far with this

and neither did little digital trackers
that we first tried putting on bats.

We just couldn't get
the information we needed.

So instead, we turned
to the mating patterns of bats.

We could look at certain parts
of the bat genome,

and they were telling us that some
groups of bats were mating with each other

and others were more isolated.

And the virus was basically following
the trail laid out by the bat genomes.

Yet one of those trails stood out
as being a little bit surprising --

hard to believe.

That was one that seemed to cross
straight over the Peruvian Andes,

crossing from the Amazon
to the Pacific coast,

and that was kind of hard to believe,

as I said,

because the Andes are really tall --
about 22,000 feet,

and that's way too high
for a vampire to fly.

Yet --

(Laughter)

when we looked more closely,

we saw, in the northern part of Peru,

a network of valley systems
that was not quite too tall

for the bats on either side
to be mating with each other.

And we looked a little bit more closely --

sure enough, there's rabies
spreading through those valleys,

just about 10 miles each year.

Basically, exactly as our evolutionary
models had predicated it would be.

What I didn't tell you

is that that's actually
kind of an important thing

because rabies had never been seen before
on the western slopes of the Andes,

or on the whole Pacific coast
of South America,

so we were actually witnessing,
in real time, a historical first invasion

into a pretty big part of South America,

which raises the key question:

"What are we going to do about that?"

Well, the obvious short-term
thing we can do is tell people:

you need to vaccinate yourselves,
vaccinate your animals;

rabies is coming.

But in the longer term,

it would be even more powerful
if we could use that new information

to stop the virus
from arriving altogether.

Of course, we can't just tell bats,
"Don't fly today,"

but maybe we could stop the virus
from hitching a ride along with the bat.

And that brings us to the key lesson
that we have learned

from rabies-management programs
all around the world,

whether it's dogs, foxes,
skunks, raccoons,

North America, Africa, Europe.

It's that vaccinating the animal source
is the only thing that stops rabies.

So, can we vaccinate bats?

You hear about vaccinating dogs
and cats all the time,

but you don't hear too much
about vaccinating bats.

It might sound like a crazy question,

but the good news is that we actually
already have edible rabies vaccines

that are specially designed for bats.

And what's even better

is that these vaccines
can actually spread from bat to bat.

All you have to do is smear it on one

and let the bats' habit
of grooming each other

take care of the rest of the work for you.

So that means, at the very least,

we don't have to be out there vaccinating
millions of bats one by one

with tiny little syringes.

(Laughter)

But just because we have that tool
doesn't mean we know how to use it.

Now we have a whole laundry
list of questions.

How many bats do we need to vaccinate?

What time of the year
do we need to be vaccinating?

How many times a year
do we need to be vaccinating?

All of these are questions
that are really fundamental

to rolling out any sort
of vaccination campaign,

but they're questions
that we can't answer in the laboratory.

So instead, we're taking
a slightly more colorful approach.

We're using real wild bats,
but fake vaccines.

We use edible gels that make bat hair glow

and UV powders that spread between
bats when they bump into each other,

and that's letting us study
how well a real vaccine might spread

in these wild colonies of bats.

We're still in the earliest
phases of this work,

but our results so far
are incredibly encouraging.

They're suggesting that using
the vaccines that we already have,

we could potentially drastically reduce
the size of rabies outbreaks.

And that matters, because as you remember,

rabies is a virus that always
has to be on the move,

and so every time we reduce
the size of an outbreak,

we're also reducing the chance

that the virus makes it
onto the next colony.

We're breaking a link
in the chain of transmission.

And so every time we do that,

we're bringing the virus
one step closer to extinction.

And so the thought, for me,
of a world in the not-too-distant future

where we're actually talking
about getting rid of rabies altogether,

that is incredibly
encouraging and exciting.

So let me return to the original question.

Can we prevent pandemics?

Well, there is no silver-bullet
solution to this problem,

but my experiences with rabies
have left me pretty optimistic about it.

I think we're not too far from a future

where we're going to have genomics
to forecast outbreaks

and we're going to have clever
new technologies,

like edible, self-spreading vaccines,

that can get rid of these
viruses at their source

before they have a chance
to jump into people.

So when it comes to fighting pandemics,

the holy grail is just to get
one step ahead.

And if you ask me,

I think one of the ways
that we can do that

is using some of the problems
that we already have now,

like rabies --

sort of the way an astronaut
might use a flight simulator,

figuring out what works and what doesn't,

and building up our tool set

so that when the stakes are high,

we're not flying blind.

Thank you.

(Applause)