-Good morning, everybody. And welcome to the Radcliffe
Institute for Advanced Study. I'm Liz Cohen. I'm dean of the institute. And I am delighted to welcome
you to our annual science symposium, which will engage
us all today in probing the past, the present,
and the future of DNA. The Radcliffe Institute
for Advanced Study is a distinctive
intellectual community where scholars, scientists,
and artists work at the forefront of
their disciplines and across disciplinary
boundaries. In fact, you might
say that dedication to interdisciplinary work is
deeply embedded in our DNA. And you can see its expressions
in today's list of speakers, which includes a chemist,
an anthropologist, an entrepreneur,
and an explorer, as well as the prominent
geneticist you might expect. I hope you'll take
advantage of the opportunity to talk with our
speakers by joining us for a reception at Fay House,
right next door, immediately after the conclusion
of today's symposium. There you will also
find DNA related posters prepared especially
for this conference, chiefly by students. These posters will be on
display in the Sheerr Room of Fay House during
the lunch hour, as well as during the reception. Radcliffe has a long and
deeply embedded commitment to the sciences and to sharing
advance scientific work with the public through
conferences like this one. We are proud of
our rich tradition of science programming,
hearkening back to Mary Ingraham Bunting. She was a microbiologist,
the fifth president of Radcliffe College,
and the founder of the forerunner of today's
fellowship program, which was called the Bunting Institute. We continue Mary
Bunting's legacy through our yearly
science symposium and through a related
science lecture series. I hope you will come back for
the lectures on DNA beginning on Tuesday November 3 at 5:00
PM when Dan Barouch, professor of medicine at
Harvard Medical School and the director of
the Center of Virology and Vaccine Research at Beth
Israel Deaconess Medical Center discusses, and I quote his title
here, prospects for a vaccine and a cure for HIV. For more information on this
series and many other upcoming events, I hope you'll
check out the calendar cards that we've put on your
seats, as well as our website. Today's conference
has been organized by our biological
science faculty director, Janet Rich-Edwards. And I'm very grateful to
Janet for her leadership. We're here today to
consider the past, present, and future of DNA. And the very title
of today's conference alerts us to how timely,
but also in some ways how timeless, a topic
DNA actually is. Swiss scientist Friedrich
Miescher first isolated DNA in white blood cells in 1869. Its discovery revolutionized
the science of genetics and also inspired an entire
literary genre of microbiology as dramatic narrative. In 1926, The Microbe
Hunters by Paul de Kruif ignited the imaginations of an
entire generation of budding scientists. Half a century
later, The Eighth Day of Creation by Horace
Judson did the same for another generation. For decades, scores of
biologists and geneticists have attributed their
decisions to become scientists to reading The Microbe Hunters
or The Eighth Day of Creation when they were young. In subject matter, both are
books about microbiology. But in form, they are
adventure stories. As one reviewer put
it, they feature, and I quote, epics
of science that come to triumphant fruition. The Eighth Day of
Creation, for example, portrays James Watson and
Francis Crick discovery of DNA's double helix
structure in 1953 as the culminating event
in molecular biology. It's a suspenseful thriller,
and discovering the double helix is the climactic moment. The book was published
35 years ago in 1979. And yet, since then DNA research
has continued to advance and has helped us to see
life on Earth holy anew. It has challenged
what we thought we knew about neanderthals,
cancer, crime fighting, human personality, and more. Its discovery has
changed medicine and spawned whole
new industries. Even in my own field
of history, which would seem to be as far
removed from cutting edge molecular biological research
as you can possibly get, it is being fundamentally
altered by DNA. For example, advances
in genomic archaeology have provided us with
better understanding of why famines affected
populations differently. Some human remains
contain genes that show unusually high prevalence
of lactase, an enzyme that breaks down lactose, the sugar
that occurs in dairy products. The more lactase a
person's body can produce, the more a person can tolerate
and derived nutritional value from milk and butter. Populations with high levels
depended less exclusively on grain for subsistence. And therefore could better
withstand poor harvests in the short term. Because they were
already accustom to supplementing low
grain yields with milk. On the other hand,
those same populations were much more vulnerable
when epidemic diseases struck cattle. Very close to home,
my own colleague in the Harvard history
department, Mike McCormick, is reevaluating the
Anglo-Saxon invasion of Britain on the basis of DNA research. Mike is not exaggerating
when he calls human genetics, and I quote him, an
extraordinary window on a vanished world of
migrations and matings that take you back literally
to the dawn of human time, end quote. The study of DNA then is
more relevant to the study of history than might
at first be obvious. At the same time, how scientists
study, understand, and apply genetics also changes over time. So while DNA changes history,
it also has a history. And we are living today
in a watershed moment of that history. Let's look quickly at a case
of DNA research now evolving. The emerging field of
epigenetics studies how environment can alter
the molecular processes that affect how genes get expressed. Researchers have shown
that toxins, chemicals, and some kinds of
trauma can alter how chromosomal code actually
gets suppressed or articulated. Recent studies raise the
intriguing possibility that social interactions
might do that as well. For instance, research indicates
that some genetic diseases progress differently in
socially isolated people than they do in people with
robust support networks. Learning more about how DNA
replicates and expresses itself therefore might not
necessarily reduce us to dehumanized chains
of polypeptides, as some ethicists worried
when the double helix was first discovered. It might instead help
us to better understand exactly how and why the
people in our lives, our friends and our families,
have a transformational impact on us. The study of DNA in short began
raising profound questions as soon as Miescher began
isolating white blood cells in 1869. It continues to do so today. And it will persist doing
that into the future. It is thus a most fitting
topic for consideration here at the Radcliffe
Institute, where we embrace open ended,
complicated, and cross disciplinary investigations. To get us started, I would
now like to hand things over to Janet Rich-Edwards. Janet is a Radcliffe alum
and an epidemiologist with research interests
centering on maternal and child health. She holds dual appointments
at the Harvard Medical School and the Harvard T.H Chan
School of Public Health. And she also serves as director
of developmental epidemiology at the Conners Center for
Women's Health and Gender Biology at Brigham
and Women's Hospital. Her research interests
include the implications of pregnancy complications for
future cardiovascular disease and the impact of childhood
abuse on chronic disease outcomes. Janet? -Good morning. Thank you Dean Cohen. And thank you all for
coming here today. And special thanks
to our speakers. Last Sunday, as my husband
and I were up on the roof watching the lunar
eclipse, we recalled the thrill of Apollo
11 landing there and the moment when Neil
Armstrong stepped out of that module onto
the moon's surface. I was seven years old. We had just had
sheer wonder that man was stepping on the moon. My husband, looking
up at that blood moon, said, it's still pretty
amazing to think about. And here's what I thought. Yeah, it is. But it ain't nothing
compared to what you're going to hear about on Friday. I'm that excited
about today's program. For as marvelous as it
is to go to the moon, it's equally
marvelous and perhaps even more pretentious
for the planet's welfare to be able to go into
the mitochondria. As you'll hear today,
the science of DNA is not only telling
us about our past. It's already improving human
health and planetary ecology. And its potential for the
future seems limitless. But this technology
is new, very new. And like any sharp tool,
can be used for good or ill. And so we need to
develop it thoughtfully and with safeguards. And so in addition
to the science today, we'll be discussing
the ethical aspects of this new world. There we go. When I was a kid, DNA
was pretty simple. You had your nucleotides, the
base pairs, A, T, G, and C. And you knew that the DNA
in the nuclei of your cells was transcribed into RNA, which
left the nucleus as messenger RNA bearing the code to
guide the translation and assemblage of
proteins from the building blocks of amino acids. Your genes on your DNA
are the sections of DNA that actually code for these
proteins that make up your cell walls, your antibodies,
your enzymes, your hormones, your muscles, basically you. As you probably know,
there are long stretches of DNA between the genes,
sections that don't actually code for proteins. We've learned that this
so-called junk DNA contains instructions for
gene regulation, like switches that turn
things on and off invisibly. I would say, unlike a Volkswagen
diesel exhaust system. The various combinations
of the four nucleotides form-- the four nucleotides form
64 distinct codes, or codons, that match up with 20
different amino acids. These amino acids,
sorry, form the proteins that form your body and
shape your behavior. But think about it. Four nucleotides,
20 amino acids, there are hundreds of naturally
occurring amino acids out there that are just left on the table. What if we could recruit
some of those amino acids to make new proteins? What if we could make more
effective medicines with them? What if we could make new fuels? If we had a larger
vocabulary of nucleotides, could we deploy DNA
to make new things? But the first question is, can
we even add nucleotides to DNA and get it to work? And that's exactly what Floyd
Romesberg's lab at the Scripps Research Institute has done. As you'll hear
today, he succeeded in creating a synthetic DNA with
two new nucleotides, x and y. With six letters
in your alphabet, there's a possibility now
of 216 codons that can now code up to 172 amino acids. And you can see
that the potential for creating new proteins
then explodes from here. It's as if the original
four letters limited us to the vocabulary of Hop on Pop. With all due respect to Dr.
Seuss who I love, with two more letters, we begin to
be able to write words that get us to Harry Potter. And who knows, with more
nucleotides, suddenly we can write Infinite Jest. And then there's your
mitochondrial DNA. If this is a cell,
here's your nuclear DNA right there in the red. These little bodies here
are your mitochondria. They are your generating
bodies in every cell. The mitochondria exist
outside of the nucleus. And they have their own DNA,
a primitive circular DNA that we probably inherited
from bacteria long, long ago. When I was in school, we use
to learn about mitochondria at the very end of the section. It was an afterthought. You inherit your nuclear
DNA from both your parents. Half your code comes from the
nucleus of your father's sperm and half from the nucleus
of your mother's egg. When they merge, your
nuclear DNA is formed. But the egg also carries
mitochondria in a cytoplasm. If you're a woman and
you have children, you'll pass the same
mitochondria along to your children. In fact, the mitochondria
in your cells, whether you're a woman or a
man, are the same mitochondria in your mother's mother's
mother's mother's mother's line, all the way on back. So while your nuclear
DNA is inherited from all your ancestors,
your mitochondrial DNA comes from your
maternal lineage. This fact offers up
both promise and peril. Mitochondria are rather
prone to mutation. They lack some of the
genetic repair mechanisms that we have in nuclear DNA. And over time, over
millennia, the mutations allow us to track who
our ancestors were by looking at these mutation
patterns across populations. If you're a man
or a woman, we can trace your maternal
genetic lineage through your mitochondria. Now, men have a second option. We can also trace a
man's father's father's father's father's line
through his paternal lineage through his y chromosome. So I was curious about this. Back in 2009, I ordered
a gene testing kit from the National Genographic
Project, which we'll hear about from Spencer Wells today. You can see here that it's
just a simple cheek swab. I was curious about my own
lineage, but even more curious about my husband's lineage. Mark is African-American. He's a dead ringer
for Barack Obama. He is. You couldn't believe
how many people ask him if he is Barack Obama. Or even "are you
Osama?" has happened. What do you say? Like many Americans,
the records that would allow us to learn
about his family heritage just don't exist. So we were very curious
to look at his DNA and trace back to
his African roots. So first we did my
mitochondrial DNA. And here's what we found. Like all humans alive
today, my maternal line can be traced back to
Ethiopia 150,000 years ago. My maternal
ancestors left Africa in the second wave
of people who did through the Saudi
peninsula, probably following better game after
the Ice Age retreated. We came up through Central Asia. Then banged a left and into
Europe, where we probably would have bumped
into the Neanderthals. And John Hawks
will be telling us more about that
mixture in our lineage. And then I discovered,
not to my great surprise, that most of my genes hail
from Britain and Ireland. So then we ran Mark's DNA. Smart scientist and
feminist that I am, given the choice between
running his mitochondrial DNA or his y chromosome DNA, I said,
oh run your y chromosome DNA. We were so excited to see
where in Africa his family had come from. So when we got his results,
there was something of a shock. Mark's paternal lineage
is more Irish than my own. Now, I have to say we
were both dismayed. Because confronting
us in Mark's genes was a story we
weren't sure we liked, was the evidence of history. We were looking at
a legacy that spoke of slavery, of human
ownership probably, of race and gender dominance. It gave us some pause. Of course, if we thought
about it for half a second, Mark's paternal line
was not the place to look for his
African heritage. We should have checked
his maternal line. So that's what we did next. Mark's mitochondrial DNA, of
course, told a different story. His mitochondrial
mutations lead us to West Africa, probably
a Fulbe or Fulani tribe. If we reran those results
today, what is it, six years later,
we would probably get even greater specificity. Because the more people
who join this project, the better that it is able
to map human migration. But our experience finding
out about our heritage was, I would say, not
an unmixed blessing. And I think there are real
ethical quandaries when the genetic story
line contradicts our own personal or
our group's or tribe's identity and our beliefs
about our origins. But bringing up
the question of who has the right to determine
your group's heritage. There's also a therapeutic
side to mitochondria as well. Mitochondria can be diseased. They get flaws in
their own DNA, which can be devastating to health. We'll hear this afternoon
from Alison Murdoch of Newcastle University
about the brave new way to prevent mitochondrial
disease that is passed from mother to child. I'll give you a brief outline. There are a few ways to do this. This is one of them. If a woman has
defective mitochondria and she goes to have
a child, there's a strong likelihood
she'll pass it on. But once she and her
partner conceive, it's possible now to extract
their healthy nuclear DNA from the unhealthy egg. If another woman with healthy
mitochondria donates an egg, it's possible to strip out
the nuclear DNA from that one, and replace it with the
prospective parent's nuclear DNA. This gives the patient couple a
chance to have a healthy child. This has been done
successfully in humans. But as you can
imagine, the process of creating life from
three parents effectively, instead of the usual two, has
raised some ethical issues. Dr. Murdoch will tell
us about the technique and about the new
legislation that has made this legal in the UK. We are also joined today by
Jacob Corn from UCal Berkeley. He will talk about
the technique you may have heard of called CRISPR,
which makes it possible to, if you will, copy edit
the genome, specific genes or their transcription
factors, those on off switches in the junk DNA. This genome editing
has the potential not to just prevent as in
the mitochondrial scenario you just saw, but
actually to cure many diseases, such as cystic
fibrosis or sickle cell anemia. This of course has enormous
ethical implications. This April, scientists in
China reported the first CRISPR manipulation of a
non-viable human embryo. If you follow the news, you'll
know that many scientists have urged a moratorium
on applying CRISPR to human germ cells,
that's egg cells and sperm cells, until the full
implications can be discussed among scientific and
government organizations. But genetic engineering is
already being used to create transgenic animals,
sometimes known as Chimera, , as for the mythical beast that
was a part lion, part eagle, part goat. In this case, a
human hormone gene is inserted into
a sheep's genome. The resulting transgenic sheep,
otherwise perfectly normal, then excretes the human
hormone into its milk, where it can be harvested to
create inexpensive medicines and potential vaccines. ZMapp, the experimental
Ebola drug, is a product of one of these
transgenic molecular pharmacy experiments. But it's not just medicines. You may have heard of the
spider goat, a regular goat who has received some DNA from
the golden orb spider. This cartoon from
Modern Farmer Magazine-- believe it or not, wow it's
really modern-- is facetious. This goat does not
actually fling webs. But it does secrete
proteins into its milk that can be made into an elastic
fiber that is 10 times stronger than steel. Imagine that, elastic
and stronger than steel. This has the
potential to be made into body armor, parachutes,
surgical sutures, and other fantastic stuff. Other applications
of DNA science will allow us to
protect the planet. Many of you may have heard a
recent NPR broadcast discussing the technique of sampling
ocean water to measure the free floating
DNA that is cast off, just shed, by fish,
sharks, plants, whales, and other things as an
inexpensive way of creating a real time ocean census. This would allow us to follow
migrations of hard to track animals. Or take leeches, or more
specifically the leech's lunch. They turn out to
be quite useful. Because they suck the blood
from just about anything. And they provide a good
tool for censusing a forest. In the Yunnan
province of Vietnam, rangers collect leeches off
trees, note their locations with a GPS, bring
them back to the lab, and put them in the blender. If you then use a process
called metabarcoding, which amplifies sections
of DNA, you can tell what species are in the forest. And using this,
they've discovered that the shy and elusive saola,
aka the Asian unicorn, which has not been seen in the wild
since 1999, is not extinct. This information could be
used to protect its habitat. But could we go further? Could we, for example,
use DNA from one species, even an extinct species like the
mammoth, to help another one? Beth Shapiro of UCal Santa
Cruz will, for example, talk about how to
clone a mammoth and perhaps how
mammoth DNA might help to save modern day
elephants by increasing their range. We will also explore with Greg
Hampikian of the Innocence Project the use of DNA for
better or worse to solve crimes. While the intention is
to free the innocent and convict the
guilty, there's also the potential of flawed science
and bad forensic technique to convict the innocent. He'll be talking
to us about that. And throughout the
day, and with your help to bring up questions,
and explicitly with the talk this afternoon
of Art Caplan, the bioethicist from the New York University
School of Medicine, we'll consider the ethics
of this fast moving science. I think this promises to
be a really exciting today. Thank you for joining us, and
I'd like to get us started. If our first panel on
Mammoths, Neanderthals, and Your Ancestors
might come up, I'll introduce George Church
who can then introduce them. So George Church
is one of our own, a former professor of genetics
here at Harvard Medical School and the principal investigator
of the Personal Genome Project. He's the director of the
Center for Causal Consequences of Variation and a member
of the collaborating group for the Molecular and Genomic
Imaging Center, both of which are NIH centers of excellence
in genomic studies. He's written Regenesis,
How Synthetic Biology Will Reinvent Nature and Ourselves. And you can get
his full bio, which is really quite impressive,
in your program too. So I won't belabor it. But thank you, George,
for moderating the panel. -Thank you. I'm just going to
do it from here. This is really one of
my favorite topics. Every aspect of it, everything
that we just heard just resonates. And it's very exciting
to be on this panel. And I'm just going
to moderate it. I'm going to try to get out of
the way as quickly as possible. I don't want to repeat
what's in your booklet. I hope you look it over. But I will say a few
things that maybe aren't so obvious in there. And I'll go in the order
that the speakers will speak. John Hawks has a
massive online presence. This is his web blog. That's not a picture
of John there. This is what he actually
looks like here. But he has been involved in
the Rising Star workshop. And for those of you who follow
the Rising Star as a reality singing program,
this is different. This is named after
the rising star cave, which has a
record for discoveries of our ancient human remains. He'll tell you about
that, very exciting. And he has a-- he
runs one of most massive the massive open
online courses, or MOOCs. He got his PhD From
University of Michigan. What's not in your
thing, in your pamphlet, is he has a degree in
English and French, in addition to anthropology. And he has a distinguished
achievement professor of anthropology
at the University of Wisconsin in Madison. He has some affection
for Neanderthals. And I think many
people have speculated about my relationship
with Neanderthals. Full disclosure, I
am 3% Neanderthal. And he's made a
huge contribution to that level of understanding. Beth Shapiro, and I met I
think because around 2007 I was mouthing off
about ancient DNA, which she was a
pioneer in producing progress in ancient DNA. And I was simply
asked by journalists whether we could go from
reading it to writing it. And I didn't have the common
sense to dodge that question. Anyway, that led to
us holding a meeting that Beth, a very tiny one at
Harvard, really was the star and really made
a big impression. And she has worked not
just with passenger pigeon, but with mammoths, giant
bears, camels, and horses, all the extinct large animals
that stimulates us when we were young kids and ever since. I think this kind of field
work is the stuff of Hollywood crusaders and so on. And she has this lovely book. I love the title, but
I also like the cover. Because it not only has
a mammoth and the words how to clone a mammoth,
but a mirror mammoth and how to clone
a mirror mammoth. So that's an inside
joke, but anyway. And Spencer Wells has a really
special place in my heart. In 2005, he was in
the process without, I think as he would say
without knowing it, founding the incredible field of
direct to consumer DNA analysis in the form of
the Genographic Project. And he's been an explorer
in residence in the National Geographic, which I think all
of us admire that sort of work. And he's led that
project for 10 years. And we've already seen
some slides from that. But about that
time, I was involved in, a little after
that, with 23 and Me and with the Personal
Genome Project. And he helped us out
in those early days of the Personal Genome Project. And he's written many
books, such as-- here's one of the early ones
I read before I met him 2002, The Journey Of Man. He's written two, the
start, Pandora's Seed. He began his PhD when he was
19 years old here at Harvard. And then both he and Beth spent
a considerable amount of time at Oxford. So without further ado,
we will begin with John. Thank you. -All right, so here we are. I've just come from
Gibraltar, where I was at a meeting with--
Gibraltar's a cool place for Neanderthals
for lots of reasons. Most notably because of
Gorham's and Vanguard Caves, which are some of the most
important sites representing later Neanderthals,
and really even up to the end of their existence. Last year at Gorham's Cave was
announced the first known case of Neanderthals engraving
on the walls of a cave. And so we now have the
Neanderthal hash tag as a legacy of their
social media savvy. It's a really exciting place,
and I love going there. And the exciting part
about this latest trip was that we had a
meeting that involved some of the really big
names in understanding Neanderthals and their
contribution to our evolution. People like Milford Wolpoff,
Chris Stringer, Erik Trinkaus, people that in the '90s were
at each other's throats. Because they disagreed about
almost everything fundamental about the way that modern
humans originated and how Neanderthals did or didn't
contribute to our populations. And the thing that was
exciting about this last week for me is I put
together, as a summary to the conference,
a list of facts that now everyone involved in
understanding human evolution universally agrees on. And the thing about
it is that these facts are facts that
have largely arisen through the application
of ancient DNA technology to ancient fossil remains. It is hard to
overstate the extent to which our present
understanding of human evolution has emerged. Especially the later
phases of human evolution has emerged within
the last 10 years. And in large part due to
our systematic investigation of the microscopic and
submicroscopic aspects of fossils, including
ancient DNA. So I thought that one
way to talk about the way that we've changed our
understanding of evolution by looking at ancient remains
and the DNA in ancient remains is to go through this list. And I can tell you that the
leading people in the field who have been leaders for
30 years last week saw this list and
all signed off on it. So I tell you that
this is really stuff that we now understand. So one is that
living people today have Neanderthal ancestors. And many of you have
heard that, of course, living Europeans,
living East Asians, living Native American
peoples have around 3% or so of Neanderthal ancestry. And peoples in
sub-Saharan Africa don't have that kind of level. But nevertheless, because humans
are all geologically hugely connected-- if you count
back your ancestors, it doubles every
generation back in time until it encompasses nearly
the entire human population. Living sub-Saharan Africans
have Neanderthal ancestors, even though their genetic
component from Neanderthals is much lower. Living people today come in some
small part from Neanderthals. But this contribution
is regionally distributed among living
people in surprising ways. So this is a chart that
shows shared derived alleles with Neanderthals,
so mutations people have in common with Neanderthals. And there's a lot of
them in our genomes. But the difference between
different populations is the signature that
some of these populations have more Neanderthal
than others. That's where we get
this 3% figure from. And the higher you
are in this chart, the more Neanderthal you are. East Asians, are
more Neanderthal derived than Europeans. They have a little
more on average. This is counterintuitive,
because Neanderthals are known for being a
Western Eur-Asian population. They're known best from Europe. So we've discovered that
this population has a legacy. We've discovered that this
legacy is not distributed as we might have anticipated. This is super interesting. DNA evidence has come from
Neanderthals, not only across their European range. And these are all
sites that have generated whole mitochondrial
genomes of Neanderthals. And there's a few more now
than when I made the chart. But also surprisingly
across a Central Asian range that we did not
formerly recognize as being Neanderthal central. We now think that this is
the hotbed of Neanderthals. And that the European
Neanderthal population was continually
being replenished from this eastern source. But the other thing
that we've begun to appreciate with
the investigation of ancient remains that are
less morphologically useful, the skulls and whole
skeletons tell us immediately what something looks like. But most of the fossil
record is fragments. And in particular,
those fragments now are generating super
interesting evidence about ancient populations. So this is Denisova Cave. I've had the privilege
of going there are a number of times
at the invitation of the Russian
Academy of Sciences. It's a beautiful place
with important excavations that cover the
last 120,000 years of archaeological history. And this cave has produced
in one little pinky an entire genome at what is
now high coverage that tells us about the prehistory
of a population that archaeologists did
not suspect existed, a population that is as
distinct from known Neanderthals as the most different living
people are from each other. So there was an ancient
population represented at this place, Denisova Cave. This aging population we've
begun to call the Denisovans. And the Denisovans
have contributed their genetic heritage to some
human populations, largely the populations of Australia
and Highland New Guinea, but in a very small fraction
across most of Eastern Asia and the new world. So we've got not one, but
multiple ancient populations that we've now sampled
with ancient DNA. And they co-existed. We didn't expect this at all. And the idea that you would
have bone fragments that were basically
useless for any kind of morphological
understanding create what is denser evidence of the
heritage of an ancient group than we have gotten
from most living people is just astounding. This is not only true of the
Neanderthals and Denisovans. We can now look inside
of human genomes and make some
estimation of what we call ghost lineages,
ancient populations that contributed to each other. If we look at my field guide
to Pleistocene hookups, you will see that we understand
quite an enormous amount about the contributions
of different populations to each other in the past. The Denisovan population,
which is known only from a pinky and
two teeth, we know from looking at that genome has
contributions from a mystery population, a ghost lineage of
ancient humans, that separated from the main line of our
human evolutionary history more than a million years
ago, much more differentiated than Neanderthals,
and Denisovans, and modern humans
are from each other. We know from looking within
the genomes of living Africans, using the same signatures that
we recognize from ancient genes to recognize the contribution
of ancient populations. We can now look for those
same signatures in cases where the ancient
genes don't exist. And we can find ancient African
populations as differentiated as they are from living
people, as Neanderthals are from living people,
that have contributed to living African populations. Our evolutionary
history involves many archaic and
modern groups of humans that in every instance
we have discovered interbred with each
other and contributed to later populations,
mostly to a minor degree. This is a very different
picture of our evolution than we had in the year 2005. No one would have
said that this is what our evolutionary
history looked like. It's looking much more like
a braided stream in which you have these populations that
are separated for quite a long time, maybe with
some minimal interbreeding between them. But populations that have become
quite inbred, quite separate from each other. And sometimes later in time
they come back into contact to contribute to each other. Sometimes of course, they
must have gone extinct. This goes now back
as far in time as 400,000 years
where the earliest hominin nuclear and
mitochondrial DNA has been recovered from
a site called Sima de los Huesos, which is at the moment
the most productive hominin fossil site in the
world for producing skeletal remains of what
we now recognize from DNA are fundamentally the ancestors
of European Neanderthals. So there is so much to learn
from these ancient specimens that we're now
through technology being able to recover. All right, when we look at
these ancient populations, I like this tree. It's a cluster. And it doesn't really recognize
all of the complexities of this relationship. But it's a good measure
of genetic difference. Because when we look at the
known Neanderthal nuclear sequences, we now have a
substantial additional sequence in this from the Altai. They are really homogeneous. There's very little genetic
difference among them compared to the genetic
differences among living peoples and compared to the
genetic difference between them and the Denisovans. So Neanderthals are inbred. And when we look at particular
Neanderthal specimens, this is a graph from a
paper describing the Altai Neanderthal specimen, in
some cases the inbreeding is profound. This individual was
actually inbred. His or-- I believe it's a
female-- her parents were in fact relatives and second
degree relatives, something like uncle niece or
grandfather granddaughter. This is highly inbred. And when we look generally
at ancient human populations, including-- this is
a chart that shows population size estimates over
time, going back in the past. And the Neanderthals,
the Denisovans, and in fact, the ancestors
of most modern human groups have passed through
very small stages where they were highly
endogamous, highly inbred. It's something that we've
learned from ancient DNA that every population that
we know about that existed in humans before
40,000 years ago was much more inbred
than living people are. We come from a heritage that
is many small groups that were dynamic, that were
continually limited in their genetic
variability by the expansion and probably replacement
of many small populations and the rejoining of these
populations later in time. This point goes to address that. I mentioned earlier, we've
got Neanderthal sequences from Central Asia
and from Europe. What we now know is
that the European ones seem to differentiate
between those that existed before
50,000 years ago and those that existed
after 50,000 years ago. And the ones later seem
to have been replenished from probably a Central Asian
source, as far as we understand the data. In other words,
Neanderthals were not a static population that was
sitting in Europe well adapted to the Ice Ages. It was a dynamic population. It was a population that ranged
over large areas of space and continually expanded
across them with groups replacing other groups. We're looking at human
evolutionary history that is much more dynamic
than we anticipated just by looking at fossil remains. Looking at the fossils, we
tended to connect the dots and say as we go back
in the past, what we're doing is we're
linking ancestors and their descendants. What we now understand is
that we cannot link ancestors and descendants in
any direct sense. What we're linking is points
along this very complex reticulate structure,
some of which are separated for a while. And so maybe maintain a
morphological contiguity with each other. But many of which are not
particularly closely connected. Many of the things that we
think in anatomical terms were probably important might be
epiphenomenon of selection maintaining them
over time within these geographic regions. When we go to more recent
phases of our human evolutionary history, ancient DNA has
been equal in its import. Although to my colleagues
who are anthropologists working in human evolution,
they haven't yet quite come to as good a grip of
the importance of this. This is the earliest known
modern European skull in Europe. It's from a Romanian site called
Oase, which is deep in a cave that you have to get
under water dive through to get to where the
cranial remains are. When we look at Oase
compared to-- this is a chart that shows its
genetic similarity to Europeans at the top and Asians and
Native American populations the bottom. You can see that
this Oase individual is closer to the Asians
than it is to Europeans. The first modern Europeans
were not European. When you look at
this skull compared to other ancient
specimens, these are all ancient DNA specimens
from slightly later phases of evolution, you can see that
it's about equally related to them as it is to Asians
and Native Americans. Our archaeological
record of modern humans, the one that we thought
was sampling the ancestors of today's Europeans, today's
Asians, and so on, doesn't. It samples ancient populations
that by and large have not contributed to great degree
to any living people. This is a chart of the
mitochondrial haplotypes found in the hunter
gatherers that lived before the advent of
farming in Europe. 83% of them have a
haplogroup mitochondrial called haplogroup U. This is the
same proportions of haplogroups in the first farmers in Europe. And as you can see,
that haplogroup U, which was 83% of this
population, is only 12% of this one. There are new haplotypes,
including mine, haplogroup H, that arrived in Europe
after the advent of farming. So Europe is a dynamic place. And we now have through ancient
DNA time slices of this. And we can show that the
ancestors that we once were thinking we're
tracing our descent from these first
Europeans actually have almost nothing to do with
today's European populations. And Europe has undergone a
succession of colonizations from other places. This is almost certainly
not unique to Europe. This is almost certainly the
history of recent humans, that our population
has been dynamic. And the ancient
archaeological remains that we have are
by and large not the ancestors of living people. So of course, the
other side of that is that once we look
at ancient DNA remains, we can get a better
resolution on how those ancient DNA remains are
connected to living people. And through that
process, reconstruct the ancestral populations. So this is some of the
work from David Reich's lab here at Harvard. And you can see that there
is a reconstruction here of where Europeans came from
based on what we've learned from ancient genomes from
Europe and from other parts of the world. And you can start to say
that there were populations in the past that we
haven't necessarily sampled well
archaeologically that are source populations that have
contributed to today's people through mixture. And we can start
to get a picture of how those mixtures happened. This is all happening
through ancient DNA. As a consequence, when we look
at recent prehistoric humans, the kinds of cultural
changes that we see-- I'm going to use the example
of language shifts. But things like the
advent of farming, things like the advent
of the Bronze Age, the things that
archaeologists have been trying to track for more
than 150 years, those things we now recognize are in
large part mediated by the growth and
dispersal of relatively consanguineous populations. These are people that actually
expanded and carried with them stuff, and the ways
that they made things, and the languages they used. And of course, we've had
this idea for a long time. Of course, many of you
know the Proto-European as an ancestor of today's
Indo-European languages is reconstructed to have
originated somewhere in Central Asia. And we've studied
the language shifts that have happened through
probably the spread of these language groups. For example, the Celtic
languages into Western Europe, the way that people must-- the
way that languages must have moved and the Iranian languages
of Central Asia and into India. We now can recognize that
these sorts of phenomenon were actually
mediated by people. The people were moving. And that phenomenon that we can
see in recent prehistoric times was almost certainly the mode
of dispersal of ancient people long before we have these kinds
of archaeological records. Well, we've learned
not only about the way that populations
move and interact. We've learned about the way that
these populations have changed through natural selection. One of the realizations
that we have had, and I'm using slide of Tibet
here to remind myself that, is that genes that we've gotten
from these ancient groups sometimes are useful
to us and actually make us able to do things
that we wouldn't do otherwise. In Tibet, people are highly
able to live at high altitude and to have more
normal birth weight children at high altitude
by function of changes to their circulatory system,
changes to their blood. One gene involved in
this, a gene called EPAS1, has an allele which is common
in Tibetan peoples, which is very rare in nearby
peoples in lowland China. And that this allele's
most close relatives that we know of in anywhere,
any archaeological or living population, are the
Denisova genome. So we have here the
origination of what is today an effective
adaptation to a very unique extreme environment. We don't know the
Denisovans were living in the highland
plateau of Tibet. We don't have an archaeological
record put in there. But we can say that the
raw material of variation that came from this population
was made use of in later humans to adapt to this
extreme environment. We're in other words, changing
as a function of interbreeding with these ancient people. It's useful to us in some cases. And of course, I am sorry
I put up this table. But they did have
a graph for this. And of course, we've
been changing recently. And this is a great
chart that shows it. These are three genes that
are involved in pigmentation in human populations. And the one that I want
to draw your attention to is all the way over
here on the right side. This is the frequency of
the light pigmented version of these genes today
in the Ukraine. So this one HERC2, which
is eye color related, has frequency today of 65%. This one SLC45A2, which is
pigmentation related, is 92%. These are the frequencies
in Ukrainian archaeological samples from around
5,000 years ago. You can see that these
genes that make people light pigmented today are new. They haven't been
around that long, and they've been
growing in frequency during archaeological time. In other words, when
we look at the past, the recent past, the
past of a few generations ago, we're looking at
people that in many cases are different from the people
that are their descendants. And of course, there have
been population movements that also contribute
to those differences. But when we look over
time, some things are becoming more common. Because they're useful,
things like milk drinking, lactase persistence, things like
genes related to pigmentation. We've discovered
through ancient DNA the beginnings states of
some populations in these. And you can see how the
genes grow over time. Humans are changing. And it's a really
exciting thing to think that we can see that
happen now through looking at the ancient genes. Finally, ancient DNA
is changing the way that we actually conduct
archaeological excavations. And this is a great one. This is from a site
called El Sidron. And this is my
friend Antonio Rosas at the bottom, who's one of the
main anthropologists working at the site. And you can see the guy in
the bunny suit who is getting ready to go into this cave. This cave is a really
constrained place. And you've got to work in this
little place wearing this bunny suit. And he's got his samples
that he's carrying out. We're actually
really careful now about how we treat
ancient remains knowing that some of the most
valuable evidence that can come from them is DNA that, as
I'm sure Beth will tell you, is highly fragmented,
hard to get, and demands that we tried
to eliminate, to the extent that we can,
contamination sources. I've been recently involved
in the Rising Star Project. And last month, we
were fortunately able to announce the fossil
remains that we've discovered, which now at more
than 1,500 specimens are the most abundant fossil
hominin site ever discovered in Africa. We named on this basis a
new species, Homo naledi. There was some news about this. And so it's a really
exciting thing. We're working in a very
constrained cave, a cave where our team has to access the
chamber with the fossils through a very narrow 18
centimeter wide crack that is a 12 meter free climb
down to the bottom, that they have to wedge
themselves down through. To do this kind of work, we
recruited through Facebook a number of, where else
would you find them, extremely skilled
archaeologists who also have this kind of climbing
and caving ability and also have the dimensions to fit
through this very narrow chasm. And I'll tell you
that there's all kinds of exciting things about this
and the way that we work. But we're working in this fossil
chamber that, at the moment, is a bed of fossil hominin bone. And our team is on
bare foot protocol to make sure that they're
not stepping on something. It gives them that
sensory ability. Now, when I told my
ancient DNA colleagues that I've got barefoot people
walking through the cave to make sure that we
don't damage things they started to scream. I mean, it was like, oh
my god, the contamination. In fact, we do
quite a lot to try to avoid contamination
and collect samples that are
collected in relatively sterile circumstances. And of course,
technology has developed to the point where we can
differentiate in a lot of cases the kind of modern
contaminants from ancient DNA, actual endogenous ancient DNA. So we're pretty-- even
our excavation conducted under relatively
extreme circumstances is one in which we're
changing practices to make it possible to get
evidence from this ancient DNA source. You start seeing that take hold
across a field, where people recognize the importance of it. They recognize universally
what you're learning from it. And they recognize that this
is contributing new information that they could not have
gotten by any other means. And what it tells
you is that we now have a field that's been
transformed by the application of DNA technology. And it is hard to
imagine where this is going to take us in the future. But if you told me 10 years
ago that it would take us to the point where some of
these real bold elephants of anthropology were all
sitting at a table agreeing with each other about
the basic facts, I would say that you were
talking about science fiction. So ancient DNA has done
more than contribute to our knowledge
of human evolution. It's actually advanced
the process of science by making us able to look
at facts where previously we were working only with
theoretical preconceptions. -Beth Shapiro. -Very impressive timing. -Yeah, that was amazing. -Make sure this
is going to work, because I am too short
to stand behind a podium. I've come to accept that. It's life, right? All right, so this is the past,
present, and future of stuff. So I thought I would start by
just talking about the past, present, and future of
this field that I work in, that John so very nicely
gave a overwhelmingly supportive introduction to. Let's see. I will get to cloning
mammoths, but it will be toward the end
in the future section. Because we haven't done it yet. Although, George is
trying very hard. So ancient DNA, it all
started way, way back in the early 1980s when
a team of researchers in Allan Wilson's
lab at Berkeley, they called themselves
the extinct species study group, extracted in sequence
the tiny little fragment of mitochondrial
DNA from the skin, museum preserved
skin, of this guy. This is a quagga. They were able to sequence
a little bit of DNA from this thing. And wait for it, it turns out
it's related to the zebra. So it wasn't exactly a
overwhelming exciting scientific result.
But the idea that DNA was preserved in things
after they died was new. And this was pretty exciting. It actually set off a
whirlwind of research. People started to extract DNA
from pretty much everything they could find. There were mummies that
had DNA in them, probably. There were Miocene aged leaves. And before we knew it, there
were dinosaur DNA sequences being published from things
that were entombed in amber. The problem with all of this
stuff is that it wasn't real. We come to know now
despite being published in very high profile
journals there were quite a few papers that
should have been retracted. I don't think ever
have been retracted. But everything about dinosaur
DNA, for example, is not true. And the reason is because
DNA does not last forever. As soon as an
organism dies, the DNA that's within all
of its cells starts to break down into smaller
and smaller pieces. One of the things that causes
this DNA decay is the sun. So when we're alive
and we go outside, the sun, UV radiation from
the sun, hits our cells and actually breaks our DNA. But we have proofreading
enzymes that will go along and fix this damage. So we don't get cancer
every time we walk outside. Of course, these are all
energy requiring processes. And after you're
dead, you no longer have any energy, so UV
radiation, oxygen, water, enzymes that are in microbes
that are in the gut. If the gut bursts during
decay, all those microbes go circulating around
the body, starting breaking down that DNA into
smaller and smaller fragments. Also tons of
microorganisms in the soil. When these fossils are
sitting in the soil, these microorganisms
will colonize the bone, break down the DNA into
smaller and smaller pieces until eventually DNA,
which when you're alive you can think of as these
massive long strands of party streamers, look
more like confetti. But not really attractive
confetti like this, like the confetti from
the New Year's parade that emerges after the last
snow melts in the Kmart parking lot in April, or May if
you're in Edmonton, as many of my friends are. So yes, DNA degrades. And this process will
continue at a rate that depends on where the bone
is actually preserved in cold and dry places. Or cold places that are pretty
much temperature stable, the DNA lasts for
longer than it tends to last in warm and hot places. And today, the oldest
DNA that is known comes from this place where I
spent a lot of my time outside of Dawson City in
Canada's Yukon territory, where we find all of this
frozen sediment that's up there. And this oldest DNA is
about 700,000 years old. The reason that we
know it's that old is because in places like
this part of the world, we find these
volcanic ash layers. This white layer
here is actually from an eruption, a volcanic
eruption, that happened. This particular eruption
was about 80,000 years ago. And we know that by dating
it using different methods like thermal
luminescence dating. So we found this particular
bone, this 700,000 year old bone, in association
with an ash layer from an eruption that happened
somewhere around 680,000 or 700,000 years. So we know how
old this bone was. And the DNA in this bone
was in terrible condition. There was very little of it. It was highly damaged
and highly fragmented. But we were still able to
generate enough data from it to sequence a complete
genome of a horse-- it's just a regular horse, the kind of
horse that still lives all over the place today,
Equus caballus-- and start to learn about how
horses had evolved and changed through time. Now, 700,000 years is a lot
more recent than 65 million, or 85 million, or even
100 million years, which is how far back we have
to go to get dinosaur DNA. So for the time being
at least, Jurassic Park, at least as it was in
the mind of Michael Crichton and the many
writers, and producers, and directors who continued
this chain of fantastic blockbuster scientific
films, will remain a thing of science fiction. Sorry. But it is not all sadness. And there are some
incredibly well preserved remains
that are found. And that we now know have
quite a lot of DNA in them. That mummified mammoth that
you see up here in the corner, this is one that was found up
in the New Siberian islands a couple of summers
ago that was thought to be associated with blood. It was of a thick red,
black, viscous substance. That I'm told they've
discovered now was not blood, but instead what is officially
known as corpse juice. Nonetheless, whether or not
it's blood or corpse juice, this is an incredibly
well preserved specimen. And we are able to extract and
sequence DNA from these things. So the field is called ancient
DNA for obvious reasons. And as I explain
it now, I'm just going to go through a
little bit, a short history, of different parts of it, my
involvement with ancient DNA really in the past
and the present and then get to the future. So the past of ancient
DNA is not very long. I started working in
ancient DNA in 1999. I was at Oxford University
in Allan Cooper's lab. We were one of the only
labs at the time that were really focusing
on extracting DNA from ancient remains. So it was pretty lucky that
we got in there at this time. I work mostly in a part of
the world called Beringia. This is Beringia here. It spans Alaska,
the Yukon territory here, across the Bering
Strait and into-- that's the part of the world that
Sarah Palin, remember her, can see from her backyard. It's too bad she's not involved
in this particular race yet. Isn't it? No it's not. Do you see that
the sea level here? This light coloration
here indicates the sea level is a lot--
it's shallower there. And so during the
Ice Ages when a lot of the water on the
planet was taken up into forming massive
glaciers that it sat on top of the
continents, the sea was lower than it is today. And this was actually exposed. It was a very
important land bridge for the movement of plants and
animals between the continents. And we had lots of movements. Horses and camels moved from
North America into Asia. Bison, things like that,
humans moved from Asia into North America. It was a really very
important conduit. Today, Beringia looks
a lot like this. I'm in this-- I'm
taking this picture from this lovely helicopter. This is in the Taimyr Peninsula
in North Central Siberia. I'll show you the
helicopter in a minute. Because it's awesome. But in the past it
looked more like this, where there was a rich
diversity of plants and animals and included things like
mammoths, and mastodons, and woolly rhinos. They never made it
into North America, but they were pretty
abundant in Siberia. There were a couple
different species of horses in North America. There were crazy
things like giant bears that if stood up like this
would be 15 16 feet tall. We killed them. My favorite extinct animal, the
giant beaver, my size beaver. It's just the funniest
extinct animal. And so we go out
into these places. And we fly in these
stellar machineries. I told you I was going to
show you a picture of this. You'll notice that
there is actually some glass missing from
some of these windows, which is really good. Because these gas
tanks here that take up most of the inside,
that's where we sat. And after we took off
and celebrated the fact that we had gotten
off on the air, some of the people who were
sitting on these gas tanks decided that smoking was
a good way to celebrate. So at least we had the windows
to let the cigarette smoke out. We stay in five
star accommodation. That's my tent. I took this picture by backing
up and unfocusing my camera. So you could see the depth
of field of mosquitoes that we would deal with there. And we wander along
collecting bones. This is actually from near
Dawson City in the Yukon. There is active
gold mining going on there called placer mining
where these miners are washing away this frozen dirt to try
to get to the gold bearing gravels that are beneath. But as they're doing
that, thousands of bones, really well preserved
bones that look like they could be a couple
years old or even modern, come out of this. And we walk around
and collect them. In an average day
of working up there, we can collect something like 30
or 40 bags that look like this. Those are mostly horse, bison,
reindeer, and mammoth bones. But when we get lucky,
we get a carnivore. They had smaller
populations, but we can still find them, sometimes wolves
and bears, giant bears. We take a chunk out
of each of these bones and take them back to the lab. And we can do this over time. So we know how old all of
these different bones are. And we measure the amount
of genetic diversity that's in these
populations over time by extracting DNA and
sequencing a bit of that. And then we can use those
measures of genetic diversity and a statistical approach
called coalescent theory to translate diversity
into population size. And so we could come up with
these plots of how populations changed in size,
a big population sometime in the past
going to a smaller population in the present day. And we've generated
population level data sets, hundreds of individuals,
for about a dozen species so far. And here's just four
of them to give you a taste of what we can learn. The top graphs with
the colored background show a reconstruction
of how much habitat was available to
these animals over time. And this time scale goes
through the last 50,000 years. The tick marks here are
the fossils, the dates, that we use to make
these reconstructions. So the top line is the amount
of habitat that they had. And then this lower line with
these confidence intervals around it are how big
their populations are. So you can see from
this that as the amount of habitat available to each
of these species increases, so do their populations. And as the amount of
habitat decreases, so do their populations. So these animals are really
responding opportunistically, dynamically to the
amount of stuff that's there for them to eat. You notice that
none of the species that I've put up on
here-- and you'll notice with these
stuffed toys here, you can clearly see that this
is horses, bison, caribou, and musk ox. None of these are extinct. That's because when
species are extinct, it's very hard to reconstruct
their population dynamics. What we found in doing this work
is that as populations declined toward extinction, we didn't see
a decline in genetic diversity. What instead we saw
was that we were seeing an increase in the
amount of genetic distance between individuals. So what it looks like is
happening as species approach extinction is the habitats are
becoming more and more patchy. And individuals are getting
stuck in these patchy habitats. And then all of
these patchy habitats are declining simultaneously. So this approach of looking
at one genetic locus or two genetic locus and
trying to find out when extinction is
happening in this way doesn't work in the case
of these types of animals. Fortunately, we have moved
on to the present day where we don't have to use
these one locus or two locus analyses anymore. Because instead, we
have this crazy type of new sequencing
technology that lets us see every little bit of
surviving DNA in these bones. We call this next
generation sequencing. This is one of
the fancy machines that you can buy for
around $1 million and go out and sequence
a whole bunch of DNA. And this has been absolutely
fantastic for ancient DNA. Will people working
with modern DNA complain a little bit
about these machines, because they can only sequence
short fragments of DNA at a time? We don't care at
all, because we only have short fragments of DNA. And for this
reason, Alumina like to parade me around when
they're doing their sales pitches as the person
who goes up and says this is a great machine. It does everything
I need it to do. Absolutely everything, nothing
wrong with it whatsoever. So the kind of things
that we can do-- and here is just a plot. This is a paper that
we published recently, a couple weeks ago. This is a plot from
some bones that we collected in the Arctic. They're about 120,000 years old. And in the early 2000s, we
tried to extract DNA from them and amplify them up using
the traditional way. And we found absolutely nothing. We couldn't get any DNA out
of these bones whatsoever. And they're cool
bones, because they belong to a really
interesting species that I'll show you a
picture of in a minute. So we decided we would go
back to these bones now using this new next generation
sequencing technology and see if there
is DNA in there, just two short
fragments of DNA to be able to amplify using the
traditional technologies. And it's true. This is the fragment
length distribution for these three bones. You see that the mean
fragment length of them is around 30 base pairs long. That's not just confetti. That's like the punch holes
from the pieces of paper that you would get to
throw up in the air. Normally we get millions of
bases at a time, 35, again, and somewhere around 30. These are terribly preserved. The bottom plots also
show that the ends of each of these molecules
have typical types of damage that you see in ancient DNA. And this is a really cool
thing that this next generation sequencing technology
lets us see too. And gets back to
what John was saying about how we can tell the
difference between modern DNA contaminating DNA and
the real ancient stuff. Modern DNA doesn't
have this damage. But ancient DNA does. So if we have a
bone, a human bone, that we know has human DNA in
it, the first thing we can do is only look at
the molecules that have this signature of damage. And that will exclude, for the
most part, the contamination that's in there. So this particular
individual, this species that we were interested in,
is the North American camel, Camelops hesternus. And it was a long
paleontological mystery, which kind of camel this thing
was most closely related to. New world camels, which
are lamas, and alpacas, and vicunas, things like
that, or old world camels, the dromedary and
Bactrian camels. Most people thought
it was new world. Turns out we have this
paleontological information from aging DNA. That in fact, they are
more closely related to old world camels
than new world camels. So it's not as
exciting as humans have-- there's a new human. But camels are interesting. Other things we can do
with ancient DNA-- I'm just going to do one more
thing to try to give you just a taste of what
we can do after we have these ancient genome--
is study the consequences of the evolutionary
consequences, genomic consequences,
of different processes. And one of these
is domestication. We had a paper
last year where we looked at the process
of horse domestication. So ancient DNA on these orange
things are ancient horses. The blue ones are modern horses. These are the Przewalskis,
which is a wild modern horse. You see that according
to mitochondrial DNA there is a ton of
diversity in horses. A lot of horse females
were presumably used in the domestication
process, but very few males. We have nuclear genomes now
from a couple of different horse genomes. And because there seems
to be a lot of-- there seems to be a mixture
between these wild horses and modern horses. We can't learn about the process
of domestication looking only at modern DNA. So we sequenced complete
genomes from a couple of horses that we know were not
domestic, because they lived before domestication happened. And were able to then
scan through their genomes using a variety of
approaches and identify parts of the genome
that look like they have been under selection
since domestication. These are the genes
that were the product of the domestication process. And we found a lot
of different things that we expected to find. We found genes that
have been associated with different behavior,
being nicer, being calmer, being tamer. Genes associated
with pigmentation, we knew that these genes
were under selection and domestication. And genes associated
with their agility, ataxia genes, et cetera. We also found a
whole bunch of things that we don't know what they do. So the next step
in this is to try to use different
types of experiments and make these genes
express in tissue culture and figure out what's going on. So ancient DNA can
show us where to look to better understand different
evolutionary processes. And now on to the future. So what does the future of
ancient DNA have for us? So we study a lot of
different things in my lab. We study all of these
large mammals that used to live during the Ice Age. We study domestication. We study adaptation
to climate change. We try to better understand
how populations and species responded to changes to their
habitat that were associated with the last Ice Age. Because we want to be able to
make more informed decisions about how to use what
limited resources we have today to
preserve species in the face of predicted
climate change. And often when we
publish these papers, because they're
about big mammals that people think are fun, we
get some popular press coverage for these things. And people call. And I'm always really
excited to talk to people about the kind of
cool and interesting science that we're doing. But they really only ever want
me to answer one question. And so to answer this question
I wrote this book, which basically says no, not yet. And I'm not going
to go through a lot of the details in this book
anyway in this talk right now. But in the book I do go through
the technical challenges associated with bringing
a mammoth back to life, as well as talk about some
of the ethical and ecological questions that are
going to come about. Of course, it would
be remiss of me since I'm standing
here not to highlight some of the really cool stuff
that is being done there, in particular by this
motley crew of-- are you wearing the same thing? These guys have
been doing-- and I'm sure we're going to get a chance
to talk about it when we have the discussion section here. But they have been doing
some fantastic work, at least to set us off on the
right foot to figure out what we would need to do
if we were going to bring a mammoth back to life. I think one of the
most important things to remember when we think about
bringing a mammoth back to life is that the Asian elephant
and the woolly mammoth, Asian elephants are known to
be the closest living relative of woolly mammoths,
are separated from each other by somewhere around 5 to 7
million years of evolution. And this is not very much. This is not very much change. So when you look at
an Asian elephant, you're actually looking at
something that is about 99% mammoth already anyway. So if you're going to
take an Asian elephant and turn it into a
mammoth, all you have to do is identify that last
1% and change that. 1% of 4 billion or so
base pairs is still like a million and
a half changes. But he's a pretty smart guy. I'm sure we can figure
out how to do this. We're also having--
we're getting a much better understanding
of the types of differences, the types of genetic changes
we might need to make. One of the first discoveries
that differentiated mammoths and Asian elephants came
from Kevin Campbell's group in Manitoba. They were looking at
hemoglobin, a protein that is a part of the red blood cells. And they discovered by
sequencing hemoglobin from elephants and mammoths
that part of this gene-- it's made up of several genes. But part of it
was different only by three different changes. Three nucleotide changes led
to three different amino acid changes. So the protein sequences in
hemoglobin were really similar. They wanted to
know though, what's up with these three changes. Why are these three
changes actually important? And so they
expressed these genes and measured their ability to
carry oxygen around the body, this is what hemoglobin
do, in different types of environments. And they found that the woolly
mammoth version of hemoglobin was much more efficient
than the Asian elephant version at carrying oxygen
around the body when it's cold. Pretty good. So this is obviously one of the
adaptations that occurred along the lineage to woolly
mammoth since the divergence between woolly mammoth
and the Asian elephant. So George and his team are
looking through these genomes now. There are a whole
bunch of genomes that have been published
recently for woolly mammoths, as there are a lot
of genomes that are becoming available
for Asian elephants, and making lists of the
genes that are different, and trying to figure out whether
they're important or not. And then using CRISPR
technology, which we'll hear about later, to try
to swap these genes out, cut and paste your way
from an Asian elephant into a woolly mammoth. So why might we want to
do this, just a little end on why we might want to go
through all of these problems considering there are
major technical, ethical, and ecological hurdles that
we will encounter along the way to doing this? In the purpose-- for
the case of mammoths, there are really two
reasons that are discussed. And the first is ecological. There is a place in
Northeastern Siberia that's led by a guy in the
Russian Academy of Sciences called Sergey Zimov. And he calls this
Pleistocene Park. Pleistocene was much more
recent than the Jurassic, so well within the realm
of possibility here. Pleistocene Park is a
great and exciting place. And Sergey already has
introduced into Pleistocene Park a couple of
different big animals that are reminiscent of
the Pleistocene, the Ice Ages, where we had things
like mammoths and mastodons. He has bison from North America,
about five different species of deer. There are some horses that
he's introduced to this area. And he's shown that over the
course of just a few summers that just having these large
mammals on this landscape, just their presence
alone, has been enough to reestablish this rich
grassland that used to be part of this landscape in Siberia. Just having these animals
there churning over the soil, distributing nutrients,
distributing seeds makes the grass come back. In essence, they provide
their own habitat. And he's also seen
that other species that are in danger of extinction,
because their habitat is declining, like
the saiga antelope, have started visiting the park. This is a picture of his park. Over here on this side,
he has the animals. And he has them
fenced out over here. You see that this is actually--
you can't see it very well, the colors here. But there is some green
little patches of green all around here. And they're actually
very different. It's much more speciose,
more biodiversity, than you get on
this particular side where you don't have animals. This is also early spring. And this means that there
has been enough grass that some of it is still
surviving this harsh Siberian winter. So these guys really
are ecosystem engineers and bringing these animals
back, this large herbivore. I mean, elephants play
important fundamental roles in their ecosystem in Africa. There's no reason
to assume mammoths didn't do the same in Siberia. So he argues,
Georgia argues, and I would argue that bringing
back these large herbivores will help to
reestablish interactions in this ecosystem
that have been missing since their disappearance. And the same argument
could be made for many other extinct species. Not to bring them back
just for the purpose of looking at an extinct
species and seeing it in a zoo, but to reestablish
interactions in ecosystems that have been missing since the
time of extinction that perhaps can be used to
reinvigorate that ecosystem and save living species, which
is actually the second reason. This research is
not only teaching us about what Asian
elephants might have done or might have looked like. But it's teaching
us about elephants. And Asian elephants are in
danger of going extinct. What if we could
use this technology to take some of these genes that
have been shaped by evolution in past to reinvigorate
the diversity, the genomes, the survivability of
species that are in danger of going extinct today. If we could take an elephant
and give it the capacity to live in colder climates,
we could expand the range that an elephant could live in. Saving elephants, rather than
bringing mammoths back to life. So I'm going to leave it there. But we can talk
about the ethics, and ecological challenges, and
technical challenges later. But for now, thank
you very much. -Wow, what an amazing
time to be a scientist. If you had told me 10 years
ago that ancient DNA labs would routinely be producing the
data they're producing today, I would have laughed
you out of the room. So it's really incredible to see
how the field has progressed. But let's step back. Let's think about
modern humans, which is what I'm going to talk about. And let's think about this
whole issue of human migration patterns, and in
particular, why humans are dispersed so widely
around the earth, and why we're so diverse. Put yourself back in time,
European age of exploration. You're cruising
around the Pacific with Captain Cook at the
end of the 18th century. And every time you crash ashore
on one of these remote islands, you find people there. And they're somewhat like
people you've seen before, but different in some ways. The language is
slightly different. They may look a
little bit different. Why are people so
widely dispersed, and how do we account for
the diversity that we see? We see a dizzying
amount of phenotypic diversity, facial
appearance and so on. That's really the
overarching problem in the field of human
population genetics. And like any big overarching
problem or theme, you can break it down into
sub themes or questions that we can start to chip away
at using the tools of science, gathering data, formulating
hypotheses, coming down on one side or the other. First question we can
ask is one of origins. Are we in fact all
related to each other? And if so, how closely? And the second is
one of journey. If we do spring from a
common source as a species, how do we come to occupy
every corner of the globe in the process of generating
these patterns of diversity that we see today? Well, the question of
origins, as with so many other big questions
in biology, seems have been answered over a
century ago by Charles Darwin. He wrote in his second most
famous book, The Descent of Man, "In each great
region of the world the living mammals are
closely related to the extinct species of the same region. It's therefore probable
that Africa was formerly inhabited by extinct
apes closely allied to the gorilla and chimpanzee. And as these two species are
now man's nearest allies, it's somewhat more probable
that our early progenitors lived on the African
continent than elsewhere." OK, so Darwin solved
the origins problem. Except he didn't really. He of course, is talking about
our shared ancestry with apes. And he was absolutely
right, even though he didn't have any data
to support that at the time. We now know from
the fossil record that apes do appear
first in Africa around 23 million years
ago, the great apes. At a time when
Africa was actually disconnected from the
rest of the world's land masses due to plate tectonics. It bumps into the
Arabian Peninsula between 16 and 18
million years ago. And at that point, we have the
first so-called African exodus. The apes that left at that time
and ended up in Southeast Asia ultimately evolved into the
gibbons and the orangutans. And the ones that
stayed on in Africa ultimately became gorillas,
and chimps, and yes, us. So yeah, Darwin's
absolutely right. If you go far enough back
in time 23 million years, you find that humans are an
African species in common with all other great apes. But that's not really
the question I'm asking. Rather I want to know about
us as a species, Homo sapiens, individuals that we would
recognize as being like us if they were sitting out
here in the seventh row. That question has been
approached historically through the study of
paleoanthropology, going out and digging up
bones and saying typically on the basis of morphology,
often skull shape, this looks a little
bit more like my cousin Frank than that does. This is the missing link. This is where we all came from. What I would like
to suggest though as a geneticist who values
large amounts of data is that while the field
of paleoanthropology gives us lots of
fascinating possibilities about human origins
and dispersal patterns, it doesn't give us
the probabilities about direct lines of
descent that we really want unless we're lucky enough
to get ancient DNA out of them. Possibilities about our
past and our origins, but not the probabilities
about direct connections to those individuals. This is a great example. What you're looking at here
are three extinct species of hominids. From left to right, Homo
habilis, Homo erectus, and paranthropus boisei,
robust australopithecine, all uncovered from the same
location in northern Kenya and all dating to
roughly the same time, about 2 million years ago. So which one of these guys am
I actually directly related to, descended from? I don't know. Possibilities about our past
and our origins, but now the probabilities
we really want. Well we as
geneticists of course, take a slightly
different approach. We turn the problem on its head. Instead of digging
things up out of the past and guessing at how they may or
may not connect to the present and to us, we start
in the present. And we work our
way back in time. Because we know it's
an absolute certainty that everybody alive
today had parents. And those parents had
parents and so on. So it should be possible
to build a family tree for everyone alive. And I'm sure that a
few people in this room have done that, built
a genealogical tree. It's the second
most popular hobby in the country after gardening. Told it's poised to overtake
gardening very soon. But no matter how far back
you can trace your family line and ultimately it all ends at
Charlemagne for some reason if you're from Europe. The point is that you
eventually run out of the written record
about your ancestors. And we simply enter this
dark and mysterious realm we call history, and
ultimately prehistory. But it turns out, of
course, that we're all carrying a written
historical document inside of ourselves,
inside of nearly every cell in our body in our DNA. And that allows us to see
back beyond that brick wall, as the genealogists call the
place where the records stop. Back to the very earliest
days of our species. And we've already had a
bit of a primer on DNA. But just to reiterate,
long linear molecule, the famous double helix
described by Watson and Crick in 1953 composed of four
subunits-- A, C, G, and T. And it's the sequence of
these A's, C's, G's, and T's that basically
provides a blueprint to make another
version of yourself. And in every
generation, you have to copy all of this information
to pass it on to your kids. And it's a lot of information. There are billions of
A's C's G's and T's in the human genome. And so it's like copying
a very long text. Imagine a long book. Say, War and Peace. But it's like copying
1,000 volumes of that. And you've got to do it in
eight hours, which is roughly what it takes with yourselves
to replicate your genome. And this is really
important information. So what are you going to do? Well, you're going to
get your cup of coffee, or whatever it is that
keeps you focused. And you're going to
check and double-check very carefully as you're
copying all of this information, because it's so important. But inevitably, what is going
to happen as you're copying? You're going to make a
typo, a spelling mistake. Those happen at the
DNA level as well during this copying process, at
a rate of around 100 mutations, as we call them, per
genome per generation, as the DNA is being
copied to be passed on. And as these mutations
get passed down to children, grandchildren,
great-grandchildren, they become what we
call markers of descent. So if you share one
of these changes, one of these markers in your
DNA, with another person, that means you share an
ancestor-- the person in the past who first had
that mutation in their DNA and passed it on
to the two of you. This is actual
DNA sequence data. Old-fashioned Sanger
sequencing, which is still the gold standard
for clinical diagnostics. Five individuals-- one,
two, three, four, five-- have had the same region
of their genome sequenced. They've been lined up. The first thing you'll
notice as you read down through these sequences is that
they're basically identical. In fact, humans are 99.9%
identical at the DNA level. We only differ,
on average, at one in every thousand nucleotides
positions from someone we're not even closely related to. So it's actually quite difficult
to find these genetic markers. But if you look carefully
enough, down here in this region, GGGG GAG. A single letter change from
a G to an A, that's a marker. If you share that
with another person, you share an
ancestor-- the person who first had that change and
passed it on to the two of you. And by looking at the pattern
of these markers in people from all over the
world, focusing in particular on what you heard
about earlier-- mitochondrial DNA, the Y
chromosome-- we've been able to construct family trees
for everybody alive today. Everybody in this room-- in
fact, everybody walking around on planet Earth all
7.2 billion of us falls somewhere onto one of
the branches of these trees. Now, these are very,
very simplified versions of the assays we
actually when we're designed in the essays
in the laboratory. But they still look
complicated, sitting out there in the audience in the morning. So let's simplify them, combine
them, turn them on the sides so the roots at the
bottom and the branches are at the top like a real tree. What's the take-home
message here? Well, it's that the longest
branches in the human family tree, the branches with the
most mutational changes in them, if you will, are found only
in African populations. And what that means
is that because of this molecular clock-like
process, the mutations ticking off at 100 per genome
per generation, Africans have been accumulating
genetic variation for longer than any other group. And, therefore, our species
originated in Africa. And at some point in
the past, a small group carrying a subset
of African diversity set out to leave
Africa and colonized the rest of the world. OK. So we're back where
Darwin started. Except we're not. We're not talking about
23 million years ago. We're talking
about an event that happened within the last 200,000
years, which is in agreement with the fossil record. Modern humans first appear
at Omo Kibish in Ethiopia, 195,000 years ago. That's roughly the coalescence
point for mitochondrial DNA. And it's only in the
last 60,000 years-- 2,000 human
generations-- that we've set out to colonize
the rest of the planet. An early coastal migration
leading very rapidly to Australia. Later inland migrations
moving to Europe, crossing the Bering land
bridge into the Americas-- the great paleolithic
wanderings of our species. And it's really detailing
these migration patterns. That's been the goal of
the Genographic project over the course of
the last decade. Now, what is
Genographic all about? You referred to it earlier. Thank you very
much for the intro. At its core, it's
a research effort. It's an effort to make
sense of the human journey, the human story. And this is field
research that's being carried out by a
consortium of scientists around the world,
working primarily with indigenous people, world's
traditional and indigenous groups. Now, why are indigenous people
so important to this effort? Well, think about
your own ancestry. I'll think about mine out loud. I have ancestors from all over
northern and western Europe. I lived until very
recently in Washington, DC. Just recently moved
to Austin, Texas. What does my DNA tell you
about the ancient patterns in any of those places? It's hard to say,
because I'm a mutt. And I've moved around a lot
recently, or my ancestors have. Ideally, we want
people who have lived in the same place for a long
period of time, thousands, tens of thousands, even,
in the case of Africa, hundreds of thousands of years. The world's indigenous
and traditional people. And they give us that glimpse
of the ancient genetic patterns linked to a particular
geographic location that allow us to make sense of
these human migrations. But when we were
designing the project, I felt very strongly
that it should not just be the story of the world's
100, maybe 200, million indigenous people. It's the story of all of us. All 7.2 billion. So we wanted to open
it up to members of the general public who
were interested in testing their own DNA, getting
the results back, but also joining the project, increasing
the size of the database, giving us more power to
detail these migratory routes. So the public participation
side, a very important component. By selling the kits
we help to fund the research we're
doing out in the field with the indigenous
populations, who, obviously, can't afford to buy a kit. And also, the third
component, the legacy fund. This is a way to give something
tangible back to the world's indigenous and traditional
groups, many of whom are leading ways of life
that are endangered today. They're often forced
to leave behind their ancient villages,
their homelands, typically moving to a growing megacity. They become part
of the melting pot. Their kids stop speaking
the original language. They lose the culture
within a generation or two. We're actually going
through a period of cultural mass extinction
at the moment that parallels the biodiversity crisis. Linguists tell us that of the
6,000-some-odd languages spoken on Earth today, by the
end of this century, between half and 90% will be
extinct, no longer spoken. We're losing a language
every two weeks. So through the legacy
grants, we want to do something to raise
awareness about this, and if we can, to slow
it, or even to halt it. Well, how are we doing? How have we done in the project? Well, an easy way is
to look at numbers. On the indigenous side, we've
sampled roughly 72,000 people representing over 1,000
different populations from every inhabited continent. Not Antarctica, obviously. Published more than
50 scientific papers. Have several more
in the pipeline. So that's gone very well. The big surprise,
though, for me, has been the excitement on
the part of the general public in getting their own DNA tested. When we were
designing the project, I had a bet with the then-CEO
of National Geographic, John Fahey, about how many of these
kits we were going to sell. He said, Spencer,
if you sell 1,000 of these things in the next
few years, you'll be lucky. Nobody's going to pay $100
to get their DNA tested. I said, I don't know, John. I don't know. I think there's
something to this. The day we launched the project,
10,000 people ordered kits. And we hit 100,000 eight months
later at the end of 2005. And we're up to
around 700,000, now. More than 140
different countries. We've even sold two
to Vatican City. Very curious about those two. And it's raised a fair amount of
money, obviously, for the field research, and also the legacy
fund, which we've given away $2.5 million so far. Just some examples of
projects that we've funded. Project to save the Yaghnobi
language, the last remnant of ancient Sogdian, once the
lingua franca of the Silk Road. If you had gone to Bukhara
or Samarkand or Kashgar 1,500 years ago, you would have
been speaking to the merchants with this language, now spoken
by around 1,500 people who, historically, have lived in
the remote Zeravshan River Valley of northern Tajikistan. Most have now been resettled
to the capital, Duchanbe, and their kids learn to
speak Russian and Tajik. And so the language is literally
on the verge of extinction. We're helping to fund school
curricula in the Yaghnobi language. Project to preserve a
traditional dance pattern among the aborigines in
the northern territories. Their song lines, their stories
of where they came from. A project along the Yukon River,
an intertribal collaboration with tribes from Northwest
Territories in Alaska to raise awareness about
the environmental issues that are affecting, in
particular, the salmon population there, which is so
important their way of life. And this is a great
example for people who need a practical reason
for saving this information and these cultures. The Shuar people
trying to preserve their ethnobotanical knowledge. Now, how many people have
been prescribed a medication by their physician recently? Probably most of us. And about 30% to 40%
of those ultimately trace back to plant sources. We often know about
these plant sources because of accumulated
traditional knowledge. Tens of thousands of years
by thousands of populations experimenting with
their environment, learning which plants do
certain things and which don't. When we lose that
body of knowledge, how many potential treatments
for cancer or Ebola or HIV might we be missing out on? We don't know. But I think it's
worth preserving. So I'm often asked, what's
your biggest surprise? What's the most surprising thing
that's come out of the project? And there are
scientific discoveries, ranging from populations
fissioning off in Africa long before we started to leave
and headed toward becoming separate species, or the
genetic impact of Ghengis Khan, whatever it might be. But the big surprise for
me has been the excitement on the part of
the general public in getting involved
in this project. In particular, the power of
what we call citizen science-- non-specialists getting
actively involved in the research project. And I've got a great example. This came to light by
accident a few years ago. A woman writes into
the project and says, love what you're doing. Lots of members of my extended
family have taken part. We've told lots of
people about it. It's very, very exciting. So cool. However, it seems like
you got my results wrong. And I'd like you to retest me. Because what you've
told me is I'm carrying a mitochondrial type
that's common in Central Asia and Siberia. And I know for a fact from
the genealogical records that my family came
from a little village just outside of
Budapest, which you're looking at here in Hungary. So clearly, I've
got to be European. I can show you records going
back to the 16th century. Please retest me. Thank you very much. Now, when I heard about
this, I got really excited, not because I enjoy
asking the lab to cherry-pick one sample
out of 3/4 of a million. They tend to get a little
annoyed about that. But rather, because
the Hungarians have a fascinating history. Linguistically, they're very
different from other groups living in Europe. Most of Europe's languages fall
into the Indo-European language family that we
heard about before. A Western Eurasian collection
of languages that includes the one I'm speaking now. Part of the Germanic branch
of Indo-European languages. Romance languages--
French, Italian, and Spanish-- the
Celtic languages, the Slavic languages,
and so on, all parts of the single language family
that probably originated somewhere in the southern
Russian or Ukrainian steppe 6,000 to 8,000 years ago. And the languages have
diversified over time. Except there are a couple
of languages in Europe that don't fall into that
category, into that family. One is Basque, which is what
we call a linguistic isolate. It's unrelated to
anything else on Earth, as far as we can tell. It could've been
brought here from Mars. Might be distantly related
to ancient Sumerian and some of the languages
spoken in the north Caucasus, like Daghastani, but
it's very, very tenuous. OK. That's off on its own. Hungarian is a member of a
different language family. It's related to languages
spoken in Finland, in Lapland, the Saami languages
of northern Scandinavia. All part of what we call
the Uralic language family. And as you can see, most of
the diversity in that family lies over in Siberia. And this makes
sense, because it's believed that the ancestors
of the Laplanders, the Saami people, migrated into
northern Scandinavia from Siberia within
the last 5,000 years, bringing with them
their language that belong to this Siberian group. We also know from
historical sources that people from somewhere
further east-- the Magyar people-- around 1,000 AD
migrated into central Europe, settled there and created
present-day Hungary, bringing with them their culture, their
love of chicken paprikash-- that actually came later,
because peppers, of course, come from the new world. But anyway, the
Hungarian language. So they had a massive cultural
impact on central Europe. And so geneticists,
using this as a guide-- and we often use
linguistic patterns as a guide when we're thinking
about the sampling-- went in and sample the number
of people you typically study in one of these projects. 50, 75 people. And what they found is that
the Hungarians are identical, essentially, genetically to
the surrounding non-Hungarian speakers. You don't see any
genetic evidence for this huge cultural shift. So when this woman
wrote in to us, I said, well, here's an opportunity to
see if this public database is actually useful. We don't have to limit ourselves
to the scientific sample sizes. How many samples do we have
in the database from people with Hungarian ancestry? Well, we had 2,300 at that time. And lo and behold, we
found that 2% to 3% of the lineages on both
the male and female side were coming out of Asia. So that makes sense. Power of large numbers. We knew that was going to be
something cool about creating this large database. But the only reason we knew
to look for this pattern is because this
woman wrote in to us. And that is the power
of citizen science. Instead of having a few
scientists sit back and say, these are the important
things we should be studying, we dispersed that out
into the community. And you've got, effectively,
hundreds of thousands of scientists, now, who
are looking independently at their own data, trying
to make sense of it and finding patterns like this. And building that kind of
citizen science functionality into the project has been a very
important part of the relaunch that we did a
couple of years ago in the creation of Geno 2.0. I'm going to end by talking
a little bit about consumer genomics. And I despise the term
direct-to-consumer genomics, because we don't talk about
direct-to-consumer bookselling. Direct-to-consumer
implies that there should be a mediator that's
making decisions for you. And I feel very strongly that
genomic information should be available to
anybody who wants it. I want to talk a
little bit about how the industry has progressed. And it's really
reflected in this graph. These are number of people who
have bought kits, DNA testing kits. So the industry traces
back to around 2002, when Bennett Greenspan founded
Family Tree DNA down in Houston, primarily
because he was interested in looking
at his Y chromosome to see if he is a
descendant of Aaron, carrying the Cohen
modal haplotype. And I'm not going to go into
the details on that story. But there is very,
very good evidence that you can find a
genetic pattern that links Ashkenazi Jews living
today to this historical figure within the last
3,000 to 4,000 years. So he founded this
company, primarily driven by personal curiosity. It was very much a
cottage industry. Literally, a few hundred
people would hear about this and decide to get themselves
tested every year. And then we went live
with Geographic in 2005. And the numbers started
to tick up significantly. 23 and Me and Navigenics
launched in 2007, in part off of the
back of our success showing that there was
a viable business there. But they launched with
products that were priced at $1,000 and $2,500. So not a lot of
people buying those. They dropped the
price in 2008, 2009. And you see this uptick here. We continue up until 2013. End of 2012 beginning of 2013. The millionth person
purchases a test kit. So it took us 11 years to
go from zero to a million. The second millionth person
bought a kit a year later. And we're over four
million now, or will be by the end of this year. No signs of it asymptoting. And this is a really,
really interesting curve from a business point of view. So there are lots
of companies that have piled into this space. My buddy Ken Shaheen
at Ancestry tells me that they're probably going
to sell between 500,000 and 700,000 kits
next year alone. So why is this happening now? Or in the last couple of years. Well, there are a few reasons. I think DNA has become part
of the national consciousness. Big corporations
talk about whatever it is being in their DNA. People are less afraid of
Jurassic Park scenarios, because they know that
it's very unlikely. They're less concerned
about being cloned. All of these were major issues
in 2005 when we launched. People are also more
comfortable now sharing their most private information. When we launched in 2005,
there was no Twitter. There was no Facebook. And people dump stuff
out onto the internet for public consumption now
that a generation ago, people would have been appalled at. And now it's standard. And I think part of that
is trusting companies with their genetic information. And I think this is a
really important part of it. We've hit enough people having
tested themselves now that it's very likely when you
go to a cocktail party with 30 or 40
people in the room, at least one or two of them
will have tested their DNA. And they may be
talking about it. And so we've reached this
threshold for viral spread. Word of mouth, so important. A great parallel would
be the film industry. This is what the film industry
typically wants to see. You pump a ton of money upfront
before the film comes out into promotion,
marketing, advertising. And you make all your
money in the first weekend, or the first couple of weeks. So ideally, you're going to see
in the film industry something like that in terms of revenue. But occasionally, a
film will come out. And it's released
in a few theaters. And it generates some buzz and
starts to get good reviews. And the filmmakers and the
distributors say, well, maybe we should release it
in more theatres, and so on. And it continues to grow. And I think that's
what we're seeing in the genomics industry. My Big Fat Greek Wedding being
one of my favorite films, a really good example. And you'll notice that it
actually made more money in the end than Twilight did. So anyway, I'm
going to end there. And if you want more information
about the Genographic project, there's the url. Thank you. [APPLAUSE] -OK. So what we'll do now is
have a little discussion among ourselves
for a few minutes as people collect their
thoughts in the audience. And I'll ask when we
transition to the audience in a few minutes that you
stand in front of the mic, and that will be my signal
to make that transition. And put your comments in
the form of a question. Just like Jeopardy. A short question, ideally. And we'll ask you to
identify yourself, as well. So I'll start this off by
asking John a question. So you had a tree that included
Neanderthals and Denisovans and modern. And the Neanderthal had all this
flat, non-diverse branching. But the Denisovan
had just one branch. But you described all
kinds of modern DNA. Is there a difficulty there in
characterizing the diversity? Tell us about that. Sure. The Denisovan we only know about
from one genome, fundamentally. We have a couple
of other specimens that have mitochondrial
genomes that we know that it's not just a one-off. But we only a very limited
amount about its diversity. We have two ways of
looking at diversity with genomes in this sense. We could look at between
the samples, in a sense. And we can look
at within samples, because a genome
encompasses variation from the entire population
that produced it, in a sense. With Denisovans, we have
an additional option, which is to look at the
present-day people that carry some fraction
of Denisovan genes, and characterize the diversity
between that fraction in living people and that same
fraction in the ancient genome. Sort of interesting
with Denisovans, because when you do
that, the Denisovans-- I love the fact that we
have a name for something that we can't see. But the Denisovans that
contributed to living people are different from the Denisovan
that we have in this cave. That may be not too
terribly unexpected, considering that the
living people that carry it are in Australia and in
the Sahul ancient land mass, and the cave where we
find them is in the Altai. But this may have been a
population that, itself, was diverse, maybe as diverse
as living peoples' populations. Yeah, there's always a problem
comparing variation that way. But we can try to parse it out. And you look within
those genomes, and you're looking at what
are homogeneous ancestral populations of them. And where we see
divisions between them, they really stick out to us. With living people,
you're looking at-- I like to emphasize to people--
you're looking at a population structure where it's actually
very difficult to create a tree like that, because
it creates the impression that people are
quite separate, when in reality, every different
component of our genome has different geographic
distributions. In fact, we're sort of a smear. In the distant past, it looks
like we were less of a smear. It looks like we really were,
in some sense, subspecies once upon a time. And today's humans are not. Today's humans are really
clinal in their variation, and are exceptionally
similar compared to these ancient people. -So just to follow
a tiny bit more. If you did that same
exercise of comparing modern Denisovans with
ancient Denisovans, and it looks diverse. And the Neanderthals, your
whole tree was ancient. What if you look at modern
Neanderthals, like you and me? And are we non-diverse, as well? -It's a super question,
because it's the obvious one. And people haven't done it yet. We are looking at, in you
and me, 3% of our genome coming from Neanderthals. And that 3% in you
and that 3% in me doesn't overlap a whole lot. They're fundamentally
rare genes. And we are mostly
not overlapping. -So that means there is 6%
between the two of us, right? -We don't know what
the asymptote is. But if you start
adding up people, you start to get up
to-- we, today, sample, among probably everyone in
this room, more than half of the Neanderthal genome. But it's distributed
as rare segments in different ones of us. -So somebody might
have all of it, huh? -If they do, I want to write
a comic book about them. -They'll be my superhero. So quick one for Beth. By the way, any of
you can chime in here. You are making the argument
that as the habitat degraded or disappeared or went
up, the populations would track for your
stuffed animals, the horse and the caribou and so on. But for the caribou, it
looked like the population is almost perfectly flat, while
the habitat was going really south fast. So did I misinterpret that plot? -No, it's right. Caribou was
non-significant as far as tracking the habitat changes. And caribou, they've lived
in the very high Arctic, and they need this. They have a very good strategy
for surviving long-term. And that is that they hang out
in places where people don't. So they still live in
the very high Arctic. They eat lichens and
things like this. And their populations have
actually done really well, as other things that would
have been competing with them for resources have collapsed. So what you see there
is a second-order effect of the horses and
bison and mammoths disappearing from
the part of the range that they would have
overlapped with caribou. And even though the
habitat is declining, caribou have hung on. Something about their
ecology makes them different. And because there's no
longer any competition, their populations have
actually increased in size, even though the amount of
habitat available to them is declining. This is just further
proof-- like we needed any-- that things don't
act in isolation. Species and organisms
are part of an ecosystem. And the ecosystem is
dynamic and always changing. And what we observe in
these different species is a combination of
effects from everything that's going on
in that ecosystem. -Spencer, you concluded on
a very provocative note. I thought it was
really interesting. Your plot is going
up quite steeply. When do we get-- so
we're like 2 million now. When do we get to 7.2 billion? -When the price drops. -That's meant seriously. -I don't think it's
going to anytime soon. I think there are huge
markets outside of the US. Most of the sales at the
moment are in the US or Canada. I think what's going to
drive that is utility. And the primary utility people
gain from the testing right now is genealogical. So they're finding second and
third cousins in the database. As the database
increases in size, you increase that utility. You're more likely to have
cousins that are in there. So it's price and utility. Sequencing is becoming
very, very cheap. And I'm on the
scientific advisory board of a startup called Helix,
which is a spin-off of Illumina. And one of the things that's
underpinning what they're doing is an exome-- so all
the coding regions of your genome-- for
significantly less than $100. And that is simply
unattainable outside of that organization right now. And as that price
approaches zero, it's going to become
much more widespread. -So once we have a
family tree of everybody, we've had instances where people
have been convicted or found based on getting DNA from their
brother or sister or something like that. So basically, it will
be nowhere to hide. -In essence, yeah. Yeah, that's what it means. I think, again, showing utility
is going to be important. But I do think that
in the next 10 years, it will be routine to
have your child tested either at birth or in utero with
non-invasive prenatal testing. And you will be seen
as an unfit parent if you choose to
ignore that information and not get yourself tested. And in fact, there was a lawsuit
a few years ago in France. A child born with Down syndrome
sued his parents successfully for having had him,
knowing that he was going to have Down Syndrome. Now, there was a technical
reason he did that. It was so he could access
certain financial benefits as a result of the ruling. But you have a couple
more lawsuits like that, and there's going to be a
strong incentive for people to do this testing. -Wow. That's impressive. Beth, I noticed on
your horse slides you had GRIN2B as one of
the genes on their behavior. Sorry. This has-- -Pass. -This has been associated
with actual experiments have been done on
mice showing an impact on their intelligence. -Does it make them
smarter or not? -Depending on what
you do with it. -Depends on the mutation. So I wonder if the horses
have the smart version or the dumber version? -Just wondering whether
you have done anything. -I think I have
the dumber version. -You had it under behavior. What did that mean? -I don't know. Those were lists that
came out of my paper. I'm sorry. Pass. -Have you been to
Sergei's Pleistocene Park? It's not that far
from the Yukon. It's right next door, according
to your favorite politician. -I haven't been. Have you? -I haven't been there, either. -Do you want to go? -Yeah, sure. -Let's go. -Let's do it. OK. -This is a field trip. -A National Geographic
trip, right. -Do you have that
power any more? -I do, actually. I'm still leading jet trips. -There might be enough
people in the audience who could underwrite it for us. -Who is interested? -Who wants to come too? Do you know what the
square kilometers is? -He's always buying more space. He just bought another big plot
of land just south of Moscow and some farmland
that he bought up. He's calling it
Pleistocene Park 2. -It's a little further south. -Yeah. But it's closer. -To a population center, yeah. -It's easier to
access by a highway, so that he can
get more visitors. -This is a real real estate
opportunity for all of you out there. We're going to need a lot of
land for the 80,000 mammoths we're going to have
in a few years. 80,000? -Well, the bison made
this huge resurgence from just a couple
hundred bison. Basically extinct
to 500,000 now. And, of course, there has to
be ranchers or public lands to support all these bison. And the reason it did-- I'm
sure everybody can guess-- is there are some
financial incentives. And it was the bison
burgers and other meat that's about 10 times lower
in fat and cholesterol. -So you think mammoth
burgers are going to be the-- -Well, we'll make
low-cholesterol mammoths. Yes. -The question is,
so when we think of the past in
prehistoric creatures, whether they be people or
mammoths or other things that used to walk, so
much of, I think, our understanding of them
is through the hard tissue remains that we have of them. So the experience of
going into a museum and seeing a mounted skeleton,
or thinking about physically what they appeared like. How much do you think
the further progression of ancient genetic
technologies are going to allow us to
conceive of the soft tissue and the physiological processes
that made them different, as opposed to simply the fact
that they look different. -Please state your name. -Adam [INAUDIBLE]. -Well, from the
human perspective, I would say we're
talking about things like Neanderthal immunity
and Denisovan altitude and metabolism. And that the genes
that we have today from Neanderthals in
northern Europeans seem to be associated
with lipid metabolism. It's stuff that you
never would have seen in the archaeological
fossil record. And that's really exciting. But the other side
of that excitement is that our inability--
when we see this gene is there, what does it do? Our inability to
understand that is not based on our lack of
fossil evidence anymore. It's based on what we don't
know about human biology. And so what for me
the promise of this is that by studying the
evolution of these things, we actually get a
different avenue on understanding their
biological significance. So it's actually not just
about figuring out what the Neanderthals were like. It's about using the
fact that Neanderthals have an evolutionary history
and a relationship to us to figure out how
our bodies work. And that's pretty cool. -Do you think that the majority
of Neanderthal DNA that stuck around in the human
genome has been selected for? -I think that the
majority is neutral. I think the majority
is just, here it is. It's a marker of interactions. And we know that there are
windows where nobody today has the Neanderthal version. So we know that there are
gaps that represent things that they had that didn't work. There are some things that
look like they're useful, that they've really grown in
frequency beyond what you'd expect. -MHC alleles. -Exactly. -I think this question
actually brings up something that's important,
both to the work that you're doing with mammoths and to
a lot of work that we're doing with other animals. And that is that we
now have more letters, more A's, G's, C's, and
T's than we understand. And for humans,
it's a special case. We know a lot more
about human biology than we do about
elephant biology or bison biology or horse biology. So we have the capacity to
make these lists of genes that were identified, mostly in
models-- model organisms, not wild animals-- and guess
what they might have done. But we're very much
at the point right now of trying to figure it all out. We have a lot of massive phone
books filled with numbers, but no names associated
with those numbers. -Which is a huge
shift in genetics. When I was getting my PhD
in Dick Lewontin's lab, Lewontin was famous
for saying, we've created this huge apparatus in
terms of theoretical population genetics. And we're ready to crank
the handle on the machine and get it analyzing data. Where is the data? Generating the data was the
hard part back in the '80s. And now, it's just ubiquitous. -This has been a really
wonderful morning. Thank you, all of you. My name is Abigail. I run a nanotechnology lab
at the Koch Institute at MIT. And this question is to all four
of you, but John in particular. You mentioned at the
beginning of your talk that 10, 15 years
ago, there were really big questions that divided the
communities that have really now had a chance
to come together, which is an amazing thing. What kinds of questions
do you see now as being really divisive
that DNA has yet to be able to answer? Maybe it could in the future. But what are some of those
big, dividing questions? And this could be to everyone. -I'll take it first, because
the other part of talk that I gave in Gibraltar
was this side-- what does the future look like? And from my point of view,
I'm interested in things like how did it come to be
that humans had this population structure in the past? What does that mean
relative to the way that we understand humans
to be distributed today as a widespread species with
great genetic commonalities among us? What can we make of rare
evidence in the past? I think of
archaeological evidence, and the fact that
we're now discovering that there's significant
plant use by Neanderthals. For instance, we used to think
of them as purely meat eaters, but we're finding the
evidence of the plant foods that they ate in
dental calculus. And now, we're
able to actually do some sequencing of oral
bacteria to try to understand their oral microbiomes. And those sorts of things
which suddenly give us a window on something that
we couldn't look at before, but the window is small,
and we know that it represents a larger world. How do we use rare evidence
to talk about things that were probably important,
but we can't see very much of? I don't know that
we're now polarized. I get a lot of happy feelings. I don't think that
we're divided by things. But I think that we're in a
phase where we haven't thought of the questions that
will dominate what we're doing for the next 15 years. -Heather [INAUDIBLE]. I'm interested in human
evolution, particularly the evolution of language. And I think that that
may be significantly tied to the difference between
humans and non-human primates being concealed ovulation. The question that
I'm curious about, has anyone identified
where on the genome the genes for concealed
ovulation might be located? Could that be figured
out by comparing it to non-human primates who do
not have concealed ovulation? And once you have identified
the genes associated with concealed
ovulation, I'm wildly curious to know if
the fossil records, if any of these
ancient fossils-- I think if we can find an
indication that they had concealed ovulation, then I'm
really willing to speculate that they might have language
skills for a whole lot of other reasons. But I'm just wondering if
anybody's thought about that, and if that could be identified,
and what the significance of it might be. -That's a good question. --[INAUDIBLE], presumably, is
looking into stuff like that. -Yeah. There are a number
of people looking into these sorts of things. A significant problem. This falls into the category
of we don't really understand how it works in humans. And comparisons
with other primates where it's not
concealed ovulation are a window of looking at that. But also, experimental
models become a possibility. And people make
transgenic mice and try to figure out what's going on. But at the moment, we
don't know the answer. -Good morning. I'm [INAUDIBLE]. I'm researcher at UMass Boston. I have a question about a very
widespread kind of notion. While both John and Spencer
pointed out that notion again, it is that I want
to see how truthful the fact that we all
originated from Africa. How about ancient people
in the Middle East? I'm Originally from
Iraq, Babylonian Assyria. And I have my own
family tree over there. And it goes back to, let's
say, just before Islam. And that's all. There's known lineage. We cannot be sure about that. So I want to know why is that? How far is that? How truthful that
kind of notion? On what basis scientists
based their notion that we all share African genes originally? -It's based on the
fact that Africans have far more genetic
diversity than people living outside of Africa. It's just absolutely clear. If I could show you a tree of
what the mitochondrial lineages really look like,
98% of the variation is found in African populations. The non-Africans
are a tiny subset. Now, we've given
them more letters, because scientists started off
looking at Native Americans and Europeans. And so they just
added new letters. So everybody in
Africa falls into what they call the L branch. But L is huge, and
very, very diverse. This is mitochondrial DNA. Same thing for nuclear DNA. You see far more
diversity within Africa than outside that. And the only way
to explain that is that people have been
living in Africa for longer, accumulating more diversity. It's had a larger
effective population size. It hasn't gone through
the sequential bottlenecks that you see in
non-African populations as they move into the Middle
East, then East Asia, then the Americas, and so on. It's just it's very clear
from the genetic information. -Yeah. We do have ancestry from
some non-African peoples. But the fraction is quite
small from Neanderthals and other people. One thing to keep in mind is
that the time scale that we're talking about for the
colonization of the world, initially, by these
African people is on the time scale of
50,000 to 70,000 years or longer, which, in
historic terms, of course, is 10 times the length
of all recorded history. So we are talking about events
that are shaded in the past. But they were events
that fundamentally shaped what today's genetic
variation throughout the world is. -Now, there are some
paleoanthropologists who feel that Homo
erectus originated in Asia and migrated back into
Africa, and ultimately that's the source of modern humans. In a long circuitous
way, ultimately, erectus does come out of Africa as well. But they went out into Asia,
and possibly came back, and then went back out. It's very complicated. -The earliest good erectus stuff
we have is outside of Africa. -Dmanisi. -Yeah, Dmanisi. So it's quite possible that
there were multiple foci that led to, ultimately, the success
and proliferation of what is today a fundamentally
African origin population. -Leonard Katz, MIT. There's been debate
about what role lethal raiding
between groups may have played in human evolution. Besides some evidence from
hunter gatherer groups, there's that of
chimpanzees, which practice lethal intergroup
raiding, which can lead to local extinctions of groups. On the other hand,
there were the bonobos, which seem to be nicer. In something that
NPR ran this summer, they seem to be suggesting
that that interbreeding shows that our ancestors
were like bonobos. On the other hand,
there is genetic data that seem to show differences
between mitochondrial DNA and Y chromosome DNA in
populations, which may show selective
wiping out of local males by conquering males. Now, how do you see
the data that you know, which is much greater,
connecting with this debate? -This is one of the big
shifts in our understanding. Archaeology has been through
a very interesting history over the last century. So a century ago,
it was a question of looking for homelands and
the wandering of peoples, and as they spread, they
spread their culture. And there was a lot of
talk of conquest and so on. And that fell out of favor in
the middle of the 20th century. What seems pretty clear there,
though, from the genetic data, is that that's probably
the way things happened. So as part of the
Genographic project we had an ancient DNA team led
by Alan Cooper and Wolfgang Haak. And they, as part of their
contribution to the project, did these things they call
transects through time. Very well-studied archaeological
sites-- Central Germany, there is one in Spain, there's
one in northeastern Europe as well-- where they go back to
about 6,000 years of history. It's well-documented. We have carbon dates. We know exactly what
the cultures are. When you see a cultural shift
at a certain point in time, you see an abrupt
genetic shift, as well. So maybe the people wandered
away and another group came in. But I think what's more
likely is that you literally had conquering groups of people
who came in and killed them off. Like it or not. My name is Irwin Shapiro. And I'm an astrophysicist
by profession. I have an interest
in biology, just as an avocation, if you will. I have a question in
regard to the DNA. I hear statements like
99.6% of the DNA of humans is the same as
chimpanzees or whatever. And I also hear
that we only know what something like
few percent of our DNA is good for-- making proteins. There are 23,000 genes,
or whatever they are. I was wondering if the
differences between individuals in a given species and between
species, whether anyone has looked at the
percent differences in the stuff we know what it
does versus the stuff we don't know what it does, and
whether this could give us any insights into possible
testing with lower animals or whatever as to what the
stuff we don't know about might be good for? If that's a coherent question. Protein-coding genes-- the
things that we know, at least, that make a protein--
tend to be conserved, which means that they don't vary
as much as other genetic parts of our genome, because if they
change, it often breaks it. Conserved means that
different species are more similar than they are
for the rest of the genome. So in that sense,
the parts that we think we know that, at
least, they make a product, and here's what it is,
tend to be less variable. But, of course, the variations
that occur in those parts tend to be pretty significant,
because they are variations that change the function. In part, we use
evidence about how much different parts
of the genome vary to try to understand how
that evolution happened. And that's true not
only, of course, within humans and
other primates, but across huge
swaths of species. -I'm Catherine [INAUDIBLE]. I work with students
at Regis College. And I wanted to ask a question
related to probabilities. In other words,
when we say there's been a genetic change
discovered here, and we see that
same change here, we assume there's relationship. But what's the mathematical
probability that these two changes happened independently? -So there are
regions of the genome that change very rapidly. They're hyper-variable. Microsatellites would
be an example of that. So repeats of, say, 10
or 12 or 14 CAG CAG CAG. And they tend to
add and subtract repeats very, very rapidly
in an evolutionary sense. So in that case, you
do have convergence. We do have homoplasy, as
it's known phylogenetically. But for the typical
point mutation, unless you happen to
be in a region that's prone to a high
mutation rate, it's exceedingly rare
that you're going to see exactly the same
mutation in the same location. Just do the math. You've got 3 billion
some-odd nucleotides in the human genome. Depending on the
population size, yes, if the human population
size were infinite, you would see the same mutations
occurring independently. But the human population
size has actually been relatively small
throughout most of our history. And we went through a
near-extinction event, probably around 70,000
to 75,000 years ago, where we think the
total number of humans alive dropped down
to fewer than 10,000. Possibly as few as a
couple of thousand. We came back from that. And that's why we have
a remarkably low level of genetic variation
as a species. So it really depends on
the region of the genome you're looking at. The ones that we tend to use
to reconstruct these trees don't change that often. And because of the tree
structure and the context and the fact that
they're not recombining, and so you're dealing with
a single intact chunk of DNA for the Y chromosome
and mitochondrial DNA, when you do see
recurrent mutations, you can actually notice them. They stand out very clearly. And you can exclude
that as a marker. -In fact, we use the rates of
the genome that mutate really quickly, that change
a lot, when we're interested in learning something
that happened recently in time, and the regions of the genome
that are more conserved when we want to understand
deeper evolutionary history. So if we're trying
to reconstruct the relationship between
all great ape species, we're going to focus on
parts of the genome that mutate more slowly
than if we were going to focus on reconstructing the
population history of the bison from Wood Buffalo National Park. We might use something
that's really fast-changing. And there, you really do have
to incorporate the possibility that you see the same
mutation happening more than once in your models. And that's done
probabilistically. -We're getting close to the
last two questions, I think. -My name is Charlotte Seed. And I manage a
genetic biorepository at Northeastern University. And my question is
primarily for Beth. In the field of
ancient DNA, what are considered
the best practices for preserving these
incredibly precious samples? Do you, say, treat
them any different from the other
samples in your lab? And has your work
with these DNA samples changed, maybe, your approach
to archiving and storing current DNA? -Yes. So the biggest problem
with ancient DNA is contaminating the sample
that you already have. So we have a separate
lab that everybody wears the bunny suit in. It looks like a lab
that you might work in if you're working with
really terrible diseases. But rather than having
negative air pressure, we have positive air pressure
to push stuff out of the labs that we don't want to get in. Everybody covers up. And we never do any
sort of amplification of DNA in the ancient DNA lab. And there's only
one-directional workflow. There's all sorts of
things that we implement. Sterilize everything
that we're using. Try to eliminate DNA. And it's still there. If you do deep
sequencing of nothing, you will get all sorts of
domestic animals and human DNA. It just happens. And microbial DNA, obviously. We do know that
short-term storage affects the preservation of DNA. In 2000, long time ago,
we were using PCR just to amplify different
length fragments from different samples. In the '50s, there
was gold mining taking place in Fairbanks, Alaska. And the samples that
were discovered, the bones that were discovered
as part of this, were split up. And some of them stayed
in Alaska and some of them went to the American Museum of
Natural History in New York, where they stored
them in the attic. And in Alaska at
the museum, there was a constant temperature. Little bit colder than
room temperature room where they were stored. So we took samples that were the
same age-- about 20,000 years old-- and amplified them. They were all taken out
of the ground in Fairbanks in the '50s. And the samples that were in
New York, the average fragment length that we could
get out of them was about 70 or 80 base pairs. But we could get 600 to
700 base pair fragments out of the Alaskan ones. In 50 years of being in the
attic, where in the summer it was 40 degrees above
0, in the winter it was 40 degrees below 0
Celsius, this had broken all of these
fragments of DNA down in just that short
of an amount of time. So the preservation of samples
after they have been taken out of the ground is,
actually, critically important for
preserving their DNA. And yeah, we try to keep
things in cool and constant temperature environments, with
as little exposure to changes in ambient moisture as possible. I think the water, actually,
expansion and contraction of water molecules, may
physically break down the DNA. But yeah, it has changed the
way we think about things. Hi. My name is Beth Thomas. I'm with Harvard management. And I have a
question for Spencer. And it has to do with the
future utility of the data that you may be
collecting, or they are collecting in the
genographics project that you're doing. And it's been interesting
watching Facebook. Just in a few years, they're
well over a billion members, albeit it's free. But as the pricing for
these kits declines, and maybe more momentum and
just the word of mouth gets out, it's going to be interesting
to see just how many people do use those kits. And I'm just curious
what you think the maybe future of that data is? If there's been a
progressive reveal or progressive
realization about what might be what the
information from Facebook might be used for it
in the future-- ads, whatever-- is there
a similar trajectory with the data at that project? -So, again, what's driving a
lot of the interest in consumer genomics right now is
genealogy and ancestry. Some people literally know
nothing about their ancestry. They're just
curious to see, am I part European, part East
Asian, whatever it might be? Other people--
African Americans-- are looking for a connection
to a tribal group, perhaps, in West
Africa, because they've lost that written history about
that side of their family. A lot of people are interested
in finding their cousins. They've got some issue
in their genealogy that can't be solved
with written records. And the only way
to resolve it is to test disparate parts
of the family tree and see if they
actually do connect. So ancestry is the
primary driver right now. That's for a couple of reasons. One is that the FDA shut down
that aspect of 23 and Me, which is one of the big players
in the field, the medical side. Part of the reason for doing
that is because of the way 23 and Me chose to
deal with the FDA. I don't think if they had
handled things differently they would have necessarily
had that part shut down. Part of it, though, is
because we haven't really shown the utility of
medical genetic testing for most healthy people. We still don't know enough
about the underlying biology and what all of these millions
of SNPs-- single nucleotide polymorphisms-- really do. Do they have any effect? Is the effect significant? Is it actionable? If somebody tells you you've
got a 1.2% higher risk of type 2 diabetes because
you've got a marker at whatever the locus is, is
that really meaningful to you? Those are the bigger
questions in human biology. I'm going to have to take strong
exception to both of those. When the FDA shut
down 23 and Me, it did not shut down 500 other
medical genetics centers. And they were providing
very high-quality data at many stages of
the human cycle. So there are very
valuable things that you can do pre-conception,
while you're still deciding who you're going to mate with. Noninvasive prenatal
testing you can now get from mother's blood. -Absolutely. -This is done in millions
of women around the world. -But I said the utility for
healthy adults hasn't really-- -But you still have the market
includes healthy adults that are planning on procreating. And then, for healthy adults,
there is a long list of things we cannot do. But there is also a significant
list of things we can do, including various cancers. Angelina Jolie was the-- -BRCA. ---the person who really changed
consciousness a little bit on that. So I think there's been
a tendency in the media to focus on the half
empty part of the glass. But I think we've
reached the tipping point that Helix and others
will exploit and make it available to all of us. So it's not 20 years
away and always will be. It's happened a few years ago. -Oh, I totally agree. -It became valuable for
adults whether or not they're procreating. -Absolutely. The problem with the 23
and Me data, of course, is that it's all
based on chip results. The markers on chips are
relatively high frequency. -They're at a relative
minority of the medical record. -And we're not
discovering anything new. The real utility is going
to come from sequencing a lot of people, as you know. -And that's exactly
what the 500 others do. All the high-quality medical
data comes from them. And some of them are in
the $100 to $200 range. -I want to thank this panel. This has been fascinating.
