When I was a young boy,

I used to gaze through the microscope of my father

at the insects in amber that he kept in the house.

And they were remarkably well preserved,

morphologically just phenomenal.

And we used to imagine that someday,

they would actually come to life

and they would crawl out of the resin,

and, if they could, they would fly away.

If you had asked me 10 years ago whether or not

we would ever be able to sequence the genome of extinct animals,

I would have told you, it's unlikely.

If you had asked whether or not we would actually be able

to revive an extinct species,

I would have said, pipe dream.

But I'm actually standing here today, amazingly,

to tell you that not only is the sequencing

of extinct genomes a possibility, actually a modern-day reality,

but the revival of an extinct species is actually within reach,

maybe not from the insects in amber --

in fact, this mosquito was actually used

for the inspiration for "Jurassic Park" —

but from woolly mammoths, the well preserved remains

of woolly mammoths in the permafrost.

Woollies are a particularly interesting,

quintessential image of the Ice Age.

They were large. They were hairy.

They had large tusks, and we seem to have

a very deep connection with them, like we do with elephants.

Maybe it's because elephants share

many things in common with us.

They bury their dead. They educate the next of kin.

They have social knits that are very close.

Or maybe it's actually because we're bound by deep time,

because elephants, like us, share their origins in Africa

some seven million years ago,

and as habitats changed and environments changed,

we actually, like the elephants, migrated out

into Europe and Asia.

So the first large mammoth that appears on the scene

is meridionalis, which was standing four meters tall

weighing about 10 tons, and was a woodland-adapted species

and spread from Western Europe clear across Central Asia,

across the Bering land bridge

and into parts of North America.

And then, again, as climate changed as it always does,

and new habitats opened up,

we had the arrival of a steppe-adapted species

called trogontherii in Central Asia

pushing meridionalis out into Western Europe.

And the open grassland savannas of North America

opened up, leading to the Columbian mammoth,

a large, hairless species in North America.

And it was really only about 500,000 years later

that we had the arrival of the woolly,

the one that we all know and love so much,

spreading from an East Beringian point of origin

across Central Asia, again pushing the trogontherii

out through Central Europe,

and over hundreds of thousands of years

migrating back and forth across the Bering land bridge

during times of glacial peaks

and coming into direct contact

with the Columbian relatives living in the south,

and there they survive over hundreds of thousands of years

during traumatic climatic shifts.

So there's a highly plastic animal dealing with great transitions

in temperature and environment, and doing very, very well.

And there they survive on the mainland until about 10,000 years ago,

and actually, surprisingly, on the small islands off of Siberia

and Alaska until about 3,000 years ago.

So Egyptians are building pyramids

and woollies are still living on islands.

And then they disappear.

Like 99 percent of all the animals that have once lived,

they go extinct, likely due to a warming climate

and fast-encroaching dense forests

that are migrating north,

and also, as the late, great Paul Martin once put it,

probably Pleistocene overkill,

so the large game hunters that took them down.

Fortunately, we find millions of their remains

strewn across the permafrost buried deep

in Siberia and Alaska, and we can actually go up there

and actually take them out.

And the preservation is, again,

like those insects in [amber], phenomenal.

So you have teeth, bones with blood

which look like blood, you have hair,

and you have intact carcasses or heads

which still have brains in them.

So the preservation and the survival of DNA

depends on many factors, and I have to admit,

most of which we still don't quite understand,

but depending upon when an organism dies

and how quickly he's buried, the depth of that burial,

the constancy of the temperature of that burial environment,

will ultimately dictate how long DNA will survive

over geologically meaningful time frames.

And it's probably surprising to many of you

sitting in this room that it's not the time that matters,

it's not the length of preservation,

it's the consistency of the temperature of that preservation that matters most.

So if we were to go deep now within the bones

and the teeth that actually survived the fossilization process,

the DNA which was once intact, tightly wrapped

around histone proteins, is now under attack

by the bacteria that lived symbiotically with the mammoth

for years during its lifetime.

So those bacteria, along with the environmental bacteria,

free water and oxygen, actually break apart the DNA

into smaller and smaller and smaller DNA fragments,

until all you have are fragments that range

from 10 base pairs to, in the best case scenarios,

a few hundred base pairs in length.

So most fossils out there in the fossil record

are actually completely devoid of all organic signatures.

But a few of them actually have DNA fragments

that survive for thousands,

even a few millions of years in time.

And using state-of-the-art clean room technology,

we've devised ways that we can actually pull these DNAs

away from all the rest of the gunk in there,

and it's not surprising to any of you sitting in the room

that if I take a mammoth bone or a tooth

and I extract its DNA that I'll get mammoth DNA,

but I'll also get all the bacteria that once lived with the mammoth,

and, more complicated, I'll get all the DNA

that survived in that environment with it,

so the bacteria, the fungi, and so on and so forth.

Not surprising then again that a mammoth

preserved in the permafrost will have something

on the order of 50 percent of its DNA being mammoth,

whereas something like the Columbian mammoth,

living in a temperature and buried in a temperate environment

over its laying-in will only have 3 to 10 percent endogenous.

But we've come up with very clever ways

that we can actually discriminate, capture and discriminate,

the mammoth from the non-mammoth DNA,

and with the advances in high-throughput sequencing,

we can actually pull out and bioinformatically

re-jig all these small mammoth fragments

and place them onto a backbone

of an Asian or African elephant chromosome.

And so by doing that, we can actually get all the little points

that discriminate between a mammoth and an Asian elephant,

and what do we know, then, about a mammoth?

Well, the mammoth genome is almost at full completion,

and we know that it's actually really big. It's mammoth.

So a hominid genome is about three billion base pairs,

but an elephant and mammoth genome

is about two billion base pairs larger, and most of that

is composed of small, repetitive DNAs

that make it very difficult to actually re-jig the entire structure of the genome.

So having this information allows us to answer

one of the interesting relationship questions

between mammoths and their living relatives,

the African and the Asian elephant,

all of which shared an ancestor seven million years ago,

but the genome of the mammoth shows it to share

a most recent common ancestor with Asian elephants

about six million years ago,

so slightly closer to the Asian elephant.

With advances in ancient DNA technology,

we can actually now start to begin to sequence

the genomes of those other extinct mammoth forms that I mentioned,

and I just wanted to talk about two of them,

the woolly and the Columbian mammoth,

both of which were living very close to each other

during glacial peaks,

so when the glaciers were massive in North America,

the woollies were pushed into these subglacial ecotones,

and came into contact with the relatives living to the south,

and there they shared refugia,

and a little bit more than the refugia, it turns out.

It looks like they were interbreeding.

And that this is not an uncommon feature

in Proboscideans, because it turns out

that large savanna male elephants will outcompete

the smaller forest elephants for their females.

So large, hairless Columbians

outcompeting the smaller male woollies.

It reminds me a bit of high school, unfortunately.

(Laughter)

So this is not trivial, given the idea that we want

to revive extinct species, because it turns out

that an African and an Asian elephant

can actually interbreed and have live young,

and this has actually occurred by accident in a zoo

in Chester, U.K., in 1978.

So that means that we can actually take Asian elephant chromosomes,

modify them into all those positions we've actually now

been able to discriminate with the mammoth genome,

we can put that into an enucleated cell,

differentiate that into a stem cell,

subsequently differentiate that maybe into a sperm,

artificially inseminate an Asian elephant egg,

and over a long and arduous procedure,

actually bring back something that looks like this.

Now, this wouldn't be an exact replica,

because the short DNA fragments that I told you about

will prevent us from building the exact structure,

but it would make something that looked and felt

very much like a woolly mammoth did.

Now, when I bring up this with my friends,

we often talk about, well, where would you put it?

Where are you going to house a mammoth?

There's no climates or habitats suitable.

Well, that's not actually the case.

It turns out that there are swaths of habitat

in the north of Siberia and Yukon

that actually could house a mammoth.

Remember, this was a highly plastic animal

that lived over tremendous climate variation.

So this landscape would be easily able to house it,

and I have to admit that there [is] a part of the child in me,

the boy in me, that would love to see

these majestic creatures walk across the permafrost

of the north once again, but I do have to admit

that part of the adult in me sometimes wonders

whether or not we should.

Thank you very much.

(Applause)

Ryan Phelan: Don't go away.

You've left us with a question.

I'm sure everyone is asking this. When you say, "Should we?"

it feels like you're reticent there,

and yet you've given us a vision of it being so possible.

What's your reticence?

Hendrik Poinar: I don't think it's reticence.

I think it's just that we have to think very deeply

about the implications, ramifications of our actions,

and so as long as we have good, deep discussion

like we're having now, I think

we can come to a very good solution as to why to do it.

But I just want to make sure that we spend time

thinking about why we're doing it first.

RP: Perfect. Perfect answer. Thank you very much, Hendrik.

HP: Thank you. (Applause)