Your ancestors aren't who you think they are | David Reich: Full Interview

0:05 When you send your DNA to a direct to consumer ancestry testing company,

0:08 you might get back an answer like you're 15% Irish

0:11 and your 30% Nigerian and you're 20% South Asian or something like this.

0:18 And what you're being told is an answer

0:20 that's related to a very specific moment in time.

0:23 But if you ask a question about another time,

0:26 80,000 years ago, you can ask another question.

0:29 What are your ancestral components?

0:31 And the thing that just comes clear from ancient DNA

0:34 is that the people who lived in any one place

0:36 a thousand years ago and 5000 years ago are almost never

0:39 directly ancestral to the people who live in that place today.

0:43 My name is David Reich.

0:44 I'm a geneticist at Harvard University,

0:48 and I work on learning about the past and the future,

0:53 using DNA extracted from ancient human remains.

1:01 Part one.

1:02 How ancient DNA re explains the world.

1:07 Chapter one.

1:08 The founding of a new field.

1:11 Ancient DNA is really a remarkable technology.

1:15 There was no data of any real meaning before 2010.

1:20 And at that time, we simply could not look at DNA from the bones of people

1:25 who lived in the past and compare them to each other and people who live today.

1:29 So really, all we could do was to speculate about how people who lived

1:34 at any one place in the past related to other people and to people today.

1:39 And there was a reasonable set of hypotheses that people had about

1:45 about the history of our species and how it spread around the world.

1:48 And that hypothesis was,

1:50 to a large extent based on the idea that people spread in a simple way,

1:55 moving from one place to the other out of an origin in Africa,

1:59 maybe 50 or 60 or 70,000 years ago, spread to different parts of Eurasia

2:04 and eventually to Australia and to the Americas.

2:08 And once people got to a place,

2:10 people pretty much stayed there without much change afterward.

2:14 So then after 2010, it became possible to lift

2:17 the veil on the data and see whether that was true.

2:21 And what was really amazing was every time we

2:23 got data from a place that we haven't seen, the results were surprising and new

2:27 and made and different from what we expected, different from what I expected.

2:31 The biases I had coming into this work in 2007 had just become

2:36 cheap enough to generate genome scale data from lots of people living today.

2:41 Today, this is routine, but at the time it was at just the price threshold

2:44 had just been crossed and it made it possible to do this.

2:47 So we were very excited to be also

2:50 just studying how people today related to each other.

2:53 I was working mostly on medical genetics

2:55 and trying to learn about risk factors for disease,

2:58 and especially in African-Americans and Latino populations

3:01 who are populations of recently mixed ancestry

3:04 from Native Americans and Europeans and West

3:07 Africans in the last handful of generations.

3:10 Then I got invited to participate in the analysis

3:13 of the DNA that was being produced from ancient Neanderthals,

3:16 ancient, archaic humans.

3:18 So we got brought onto that consortium,

3:20 which was this sort of amazing group of people

3:22 working together to analyze this very, very special data.

3:25 I thought, this is the best data in the world,

3:27 and I want to spend my own time analyzing it.

3:30 So I changed everything I did,

3:32 and it was completely clear that this data was revolutionary.

3:35 We would be able to see how people who lived,

3:38 70,000 years ago, 50,000 years ago related to people living today.

3:43 And these were very special people.

3:44 These were, people whose skeletons were

3:47 very different from many people living today.

3:49 And we could ask all sorts of questions

3:51 about when they separated from each other and maybe,

3:53 maybe did they even mix with people as modern

3:56 humans spread into the territory these people were living in?

3:58 So it was an incredibly exciting thing to do.

4:01 And the results were even more exciting than we could have anticipated,

4:03 because when we came into this work, we had this bias.

4:06 I've had so many biases entering the different projects I've worked on.

4:11 We had this bias that modern humans, as they spread out of Africa,

4:14 had not mixed with the archaic humans they encountered,

4:17 like the Neanderthals that had been

4:19 the approximate finding from analysis of modern DNA.

4:22 There wasn't really very strong evidence that there was

4:24 an archaic component in Non-africans little hints here or there,

4:28 but mostly it looked like a simple out of Africa spread.

4:31 And so nobody really thought that there was good evidence

4:34 for, archaic humans having mixed with non-africans during the spread.

4:38 But then when we looked at the data, there was very,

4:41 very clear evidence that Neanderthals that we

4:44 were sequencing from Europe were more

4:46 closely related to people outside of Africa

4:48 today than people inside Africa today.

4:51 And then we kept trying to make this result,

4:52 try to go away, and it just wouldn't go away.

4:54 And we looked at it in several different new ways,

4:56 with completely different types of ways of analysis.

4:59 And it just got stronger and stronger.

5:00 And it became absolutely clear that as modern humans moved out of Africa,

5:04 they mixed with the local archaic population,

5:07 and about 2% of the DNA from Non-africans today comes from these archaic humans.

5:13 And that was just the first of many,

5:15 many surprises that comes out of this ancient DNA revolution.

5:19 Really.

5:19 In 2010, the same year that, we published this paper on the Neanderthals,

5:24 we got DNA from another archaic human in Siberia

5:27 that we completely had not expected at all.

5:29 It was a pinky bone from a little girl,

5:31 and it had no other skull or big morphologically informative parts to it.

5:36 And we got a high quality genome sequence from it,

5:39 and we could see how it was related to Neanderthals and to modern humans.

5:42 And it was not either.

5:43 It was something else.

5:44 So we didn't even know there was another group of people living there.

5:47 But this was another group, and these people, amazingly,

5:50 had also mixed with modern humans, but different modern humans.

5:52 So people living in New Guinea, for example,

5:54 derive maybe 4% of their DNA from these, what we call Denisovans.

5:59 After the cave, they were discovered.

6:01 So again and again, when we get DNA

6:03 from a place we haven't haven't gotten DNA before,

6:05 we see these surprises because the world is so full

6:08 of of of history that we didn't know about before,

6:12 and it just shows us how little we knew before and how little we still know.

6:16 The truth is that the world's a mystery,

6:17 and that we've carved out areas of clarity.

6:24 Chapter two A new scientific instrument.

6:30 Sometimes the measure of when you have

6:32 a new seam of important information is when

6:35 when you apply this new technology to topics

6:39 that it's not been applied to before,

6:41 things are just consistently new and surprising.

6:44 We had the skeletons of the Neanderthals.

6:46 We had a lot of knowledge about the tools they made and where they were,

6:50 just where they lived.

6:51 And there were lots of theories about how they might relate to modern humans,

6:55 when we might share a common ancestor.

6:57 And you can test those.

6:58 Some of those theories are right.

6:59 Many, many of them are wrong.

7:01 And you can actually look at it once you have data like this.

7:04 Ancient DNA is not more important than

7:05 these other ways of learning about the past,

7:07 like based on the tools people make or in later times,

7:11 the languages people speak.

7:13 But it's a quite different type of information.

7:16 What's really happened is the introduction of a new scientific

7:19 instrument that is able to measure things that has not been,

7:22 that have not been measured before.

7:24 So there have been other scientific instruments

7:26 that have been revolutionary in their area,

7:28 arguably more important ones, like the telescope or the microscope.

7:32 So when the telescope was first invented and people peered through it,

7:35 they could see planets around Jupiter, right?

7:37 Or they could with the microscope,

7:39 you can see whole worlds of microorganisms you didn't

7:42 even know about before that were not even envisioned before.

7:44 All these tiny creatures of, like,

7:47 protests and fungi and all of these things with amazing

7:51 structures that were not even

7:52 envisioned in people's philosophies and imagination.

7:55 And in the same way, when you can take DNA from people who lived in the past

8:00 and compare it to each other and to people living today,

8:02 you see things that you just didn't even imagine before the step migrations,

8:07 the expansion of Lapita pottery culture,

8:09 people of East Asian ancestry to the Southwest Pacific.

8:13 Every time we get DNA from Africa from more than 3000 years ago,

8:17 it's a huge surprise.

8:18 You look at Malawi and south central Africa,

8:20 the people there 3000 years ago, nothing like the people there today.

8:24 They're a population that simply doesn't exist anymore.

8:27 You look at Cameroon,

8:28 which is supposed to be the homeland of the people who speak Bantu languages,

8:31 which are spoken in the Congo area or in much of East Africa,

8:35 or related to the languages that live in much of the West Africa.

8:38 You would think, oh, this is just going to look like Nigerians or something,

8:41 because that's what it should look like.

8:43 Given that Nigerians look like most Kenyans

8:45 and look like most people from Zimbabwe.

8:47 But no, it's like a completely different

8:49 population that we didn't know about before.

8:52 This amazing study just came out from Libya of people

8:55 who lived about 7000 years ago in the Sahara,

8:57 a completely different population that we didn't know about before.

9:01 So every time we get data, it's just surprising.

9:04 So we developed a whole set of statistical

9:07 techniques to try to interpret this data.

9:10 We applied these tests to look at the Neanderthals

9:12 and the modern humans and the Denisovans,

9:14 and to try to understand how they might have mixed to form people living today.

9:18 So it was a combination of the statistical techniques to get clear

9:22 interpretation from these patterns of genetic data and this amazing new data,

9:26 both from modern people and even more from ancient people.

9:29 So it became clear that it would be possible to obtain super high resolution

9:34 information for how people relate to each other and to people in the past,

9:38 and that it was only now becoming possible to do that.

9:41 And the results were not what was expected before.

9:44 What happened at the end was, I think I, became very,

9:49 very interested in trying to read these two areas,

9:52 studies of people living today and studying of ancient DNA,

9:55 and the people who were leading this work

9:57 in the world were not really applying these methodologies,

10:01 at all to people living in the last 20,000 or 15,000 years.

10:07 But the last ten or 15 or 5000 years is

10:10 when the world got to be the way it is today,

10:13 when all the languages diversified,

10:14 when the cultural variation that you see appears and so you can look at diverse

10:20 people around the world today and try to understand how that all comes together.

10:23 And so I completely retooled my laboratory.

10:26 We started with a mammoth that had been dredged up,

10:29 in hurricane Sandy on the ocean floor off Cape

10:33 Cod so that we didn't contaminate it with human

10:35 DNA and see if we could extract DNA

10:37 from a mammoth tooth that had been dredged up.

10:39 And then we started looking at our first human DNA,

10:42 and we started working with a colleague, David Anthony,

10:46 who is an archeologist and an amateur linguist who had

10:51 become interested in asking how it is that the languages

10:54 that are spoken across a wide swath of Eurasia come

10:57 to be so related to each other languages like Sanskrit,

11:00 Indic, Iranian languages, English, which is a Germanic language.

11:04 Italic like Latin.

11:06 Languages like Celtic, like Gaelic in Ireland, in Armenian in the Caucasus,

11:11 languages that are now no longer spoken like toe Korean.

11:14 That's was in eastern China,

11:16 or since young or Hittite and the and what's now Turkey and the Anatolian

11:21 how it is that these languages came to be so closely related.

11:25 So he had a theory which was that this might be related

11:29 to people who lived in the steppes north of the Caspian Sea,

11:32 a step culture called the Yamnaya.

11:34 Who are the first people to take wheeled vehicles which,

11:38 using wheels that had just been invented out onto the open steppe lands,

11:41 he had an idea that they might

11:42 be important in the spread of Indo-European languages,

11:45 so we sequenced some bones from the Yamnaya and from their successor cultures,

11:50 and also some bones from, Germany,

11:53 from, a series of samples that we had covering the same time period to see

11:58 whether the Yamnaya might have had an impact on the history of these people.

12:02 And we got this incredible result from this data.

12:05 So the incredible result was that the Yamnaya had

12:07 this very distinctive ancestry that is a cocktail of, very distinctive,

12:14 genetically recognizable components that are similar

12:16 to what you see in Europeans today,

12:18 but somehow more extreme and when you look at the farmers

12:22 who lived in Europe before or 5000 years ago,

12:24 for example, in Germany or Hungary.

12:26 They don't have this type of ancestry you see in the Yamnaya.

12:30 And then after 5000 years, it's everywhere.

12:32 So it's very clear that the Yamnaya spread,

12:35 powered by whatever technological innovations they have,

12:39 they spread and we see Archeologically,

12:40 they spread all the way from Hungary in the west to Mongolia in the east.

12:44 And then somehow, even though the Yamnaya themselves don't spread,

12:47 people who descend from them through

12:49 cultural mixing and genetic mixing spread further.

12:52 And we see after about 40, 840, 740 600 years ago,

12:56 this Yamnaya ancestry appearing in each place in Europe.

12:59 So we see it appearing in Germany at about this time,

13:02 and it's about three quarters of people's ancestry.

13:05 It gets to Great Britain a little bit.

13:06 About 40, 500 years ago,

13:08 the first big stones at Stonehenge go up and then within 50 or 100 years,

13:12 the previous population is almost gone and it's

13:14 displaced by people with ancestry from this east,

13:17 from the east, it's a huge genetic impact.

13:19 There was an idea that I think was common or prevalent

13:23 in archeology that once the first farmers came to Europe,

13:27 it would have been very difficult for anyone new to make an impact.

13:30 There were linguists who thought that maybe the steppe was a reasonable

13:34 place for the Indo-European languages to have

13:37 spread for reasons related to language,

13:39 and that sharing of words based on corresponding

13:43 to technologies that would have only been invented,

13:46 for example, 50, 500 or 6000 years ago.

13:48 But people mostly thought that there could not

13:50 have been a big spread from the steppe,

13:52 because how could pastoralists herding sheep and cattle make a huge

13:57 impact demographically on a densely settled farming area like Europe was,

14:01 after 5000 years ago?

14:03 But they did.

14:04 The genetic data shows they did it,

14:05 and it's the main ancestral population of Europeans throughout

14:08 northern Europe and even much of southern Europe today.

14:10 Something huge happened.

14:11 And again, it's the kind of surprise that this field yields.

14:15 So that's what happened in our work when

14:18 we first started collecting DNA from ancient humans,

14:21 we started seeing these transformative changes

14:25 that happened radical shifts in ancestry,

14:27 huge mixture proportions, often sex biased,

14:30 where the two sources of ancestry that would come together in any mixture event,

14:34 for example, European hunter gatherers and European farmers,

14:37 or Papuans and people of East Asian ancestry

14:40 in the Southwest Pacific would come together in very asymmetrical ways,

14:44 with men from one group participating

14:46 and women from another group participating.

14:48 And this is telling you about events in the past.

14:50 These were disruptive events.

14:52 These were events that completely changed the people

14:55 who lived in different places in the past,

14:57 and often were much more extreme or different

14:59 from, or completely unexpected compared to what people thought about before.

15:04 And so what this has produced is an amazing dialog between genetics,

15:08 which provides this unambiguous information about what happened

15:12 in the past in archeology and sometimes linguistics,

15:15 which provide other sorts of information.

15:17 And it forces a new way of talking

15:20 and discussing and thinking about these events in the past,

15:23 and also gives us some kind of humility.

15:26 I think about the extent to which we can know what happened in the past,

15:30 because if you expose our understanding of, for example,

15:33 what happened to the sun in the Southwest Pacific or what happens and, you know,

15:37 3000 years ago or 2500 years ago, what happened in Europe 5000 years ago,

15:40 or 4000 years ago, what happened in our camp

15:43 with archaic humans 40 or 50,000 years ago?

15:45 If you expose it to a new type of information and it's so,

15:48 so, so shocking, what does that mean?

15:51 If we use the new type of information that we don't have at all yet, right.

15:54 Like, what if we could get a new type of information,

15:57 I guess would be that it would also be very surprising.

16:04 Part to human history is not a family tree.

16:09 Chapter one A braided trellis.

16:13 The tree of human populations that I learned when I was like

16:19 learning about how human populations around the world are related to each other,

16:23 has a common ancestor, maybe somewhere in Africa.

16:26 That's based on early data from the 1980s, based on mitochondrial sequences,

16:31 which are inherited from mother to daughter splitting

16:34 into different African lineages that are represented today,

16:38 and then spreading out of Africa and splitting

16:40 into, like East Asians and Australians and ancestors of Europeans.

16:45 And then maybe Native Americans might split off

16:47 East Asians like a simple tree like that.

16:50 And that corresponds to an idea that as people move around, they split,

16:54 they don't really mix, and they settle in different places around the world.

16:58 But the truth is from genetic data that in every one of the places I mentioned,

17:03 all of those groups are profoundly mixed, of multiple, very divergent groups.

17:07 So the truth is not a simple splitting from a common ancestor,

17:11 but a kind of braided trellis going back into the past.

17:15 And that trellis doesn't stop 50,000 years ago.

17:19 It just keeps going.

17:21 So as we get Neanderthal sequences or Denisovan sequences

17:24 from these archaic humans and compare them to modern humans,

17:27 we see that they too are mixed with different groups of people,

17:30 and modern humans are mixed or have contributed to a mixture of other groups.

17:35 And so there's no one time that we've been able to reconstruct

17:38 where there's a single homogeneous population

17:41 that's ancestral to everybody, who comes after.

17:44 Rather it's some kind of constant change

17:48 of some changes over time with mixtures,

17:52 separations, structuring, coming back together.

17:55 Some groups disappear, other groups don't.

17:58 And it's just very complicated.

18:00 And that's sort of the fundamental nature of who we are.

18:02 It's not kind of an unusual thing.

18:05 We used to be that we had this null hypothesis that no mixture occurred,

18:09 and we would test that null hypothesis.

18:11 And that is almost a silly null hypothesis at this point,

18:14 because everybody is mixed.

18:15 Nobody is pure.

18:17 When you send your DNA to a direct to consumer ancestry testing company,

18:20 you might get back an answer like you're 15% Irish and you're

18:25 30% Nigerian and you're 20% South Asian or something like this.

18:30 And what you're being told is an answer

18:32 that's related to a very specific moment in time.

18:36 Basically, what you're being told is if you

18:38 roll back 500 years or a thousand years,

18:42 what fraction of your ancestors lived in Ireland?

18:44 What fraction of your ancestors lived in West Africa somewhere?

18:48 What fraction of your ancestors lived in South Asia somewhere?

18:51 And we're making our best guess of that.

18:53 But if you ask a question about another time 80,000 years ago,

18:57 you can ask another question what are your ancestral components?

19:00 And there it would not be in those particular

19:03 places because those people have moved so much.

19:05 But there the question will be, what fraction of your ancestors,

19:08 if you're a non-African person,

19:09 will live in Europe because they're Neanderthals?

19:12 What fraction of your ancestors will live in sub-Saharan Africa?

19:15 Because they're from the African population

19:17 that expands out of Africa around that time.

19:19 So it's a matter of what time, in the past you're talking about.

19:24 But the way I think about it is back in time.

19:26 And if you look at any one person today,

19:29 you can look at all your genealogical ancestors.

19:31 I have two parents.

19:33 They have two parents, my four great grandparents,

19:35 eight great great grandparents, 1632, 60, 428, 256, 512, etc.

19:41 it gets pretty big, pretty fast and at any time depth of number of generations.

19:46 Back in the past, you can ask what fraction of them lived here,

19:50 what fraction of them lived there.

19:52 And the thing that just comes clear from ancient DNA is

19:56 that the people who lived in any one place a thousand years ago,

19:59 or 5000 years ago are almost never directly ancestral

20:02 to the people who live in that place today.

20:04 So the strategy used by these direct to consumer ancestry and testing companies

20:08 of using modern day Irish as surrogates for who lived in Ireland in the past,

20:12 maybe that works in the last thousand years,

20:14 but it's not even going to really work very well.

20:17 2500 years ago or 3000 years ago,

20:20 because there's just too much mixing and migration.

20:22 And if you look at the genetic data, what you see is there are periods

20:26 of continuity where populations descend from the previous groups,

20:30 but it's interpolated and separate,

20:32 and every few hundred years or few thousand years by profound disruptions.

20:37 I'll give you one place where we know this particularly intensely,

20:40 which is the island of Great Britain.

20:42 In Britain, the whole island is cleared of people by the last glacial maximum.

20:47 So that's between 25,000 to 19,000 years ago that the glaciers begin to melt.

20:52 People walk across the lower North Sea,

20:55 which was just a land bridge, and get into Great Britain.

20:58 There's begins to be people 14 or 15,000 years ago.

21:01 There's hunter gatherers related to the people on the continent.

21:03 They hang out there for a very long time, until about 6000 years ago.

21:08 And then farmers stream in from the continent.

21:10 They are related to people who are

21:12 from what's now Turkey who brought farming to Europe.

21:15 They first got to Europe about 80, 500 years ago.

21:18 They spread across the continent.

21:19 They get to Britain 2500 years

21:21 later and they are completely transformative event.

21:23 So only 1% of the DNA, if that of the people in Britain after

21:28 this event comes from the hunter gatherers of Britain.

21:30 So if you want to ask, am I descended from, quote, man,

21:33 the sort of individual who has excavated in Cheddar

21:36 Gorge where cheddar cheese is made and we

21:39 have their DNA and their DNA is corresponds

21:42 to someone who you could genetically predict had very,

21:44 very dark skin, almost like what you

21:46 would expect from Nigeria or something like this.

21:48 And blue eyes, like a combination that's very rare today.

21:51 How much DNA do people in Great Britain have from this person?

21:54 No more than 1% because of what happened.

21:56 6000 years ago.

21:57 And then flash forward 1500 more years

22:00 and the last big stones at Stonehenge go up.

22:02 They are put up by the descendants of these first farmers.

22:06 There has been continuity for 1500 years in Great Britain,

22:09 and then within 100 years, those people are 90 to 100% displaced

22:13 by a new stream of people coming from the continent.

22:15 At this time, we now know from an area

22:17 like the Netherlands associated with this Bell Beaker event.

22:20 That's another 90 to 100% spread.

22:23 And then wait a bit more and about 3000 years ago,

22:26 there's a huge stream of people from what's now what's probably from France,

22:31 probably related to the spread of Celtic languages.

22:34 And that displaces half of the DNA

22:36 of the Britons and then associated with the Saxon period,

22:40 maybe 4 or 500 C.E., 1600 years ago, 1500 years ago.

22:45 There's another third ancestry,

22:46 and then there's continued flowed into Britain, after that.

22:50 So if you ask the question,

22:51 a white British person today and you say this person,

22:54 it they feel like they're pure descendants of people

22:57 who lived in Britain a long time ago.

22:59 Nothing could be further than from the truth.

23:01 They're like homeopathic,

23:02 diluted compared to the hunter gatherers or the first farmers.

23:07 And that's not unusual.

23:08 Britain's actually relatively protected from the churn of people

23:11 that has occurred in many other parts of the world.

23:13 If you look at Hungary,

23:14 it's much more extreme and even more frequent disruptions.

23:17 If you look at Italy, there's many disruptions.

23:19 It's not just Europe.

23:20 If you look at the Near East,

23:22 there's disruption after disruption after disruption.

23:24 Every place in Africa, we've had data to look at.

23:27 There's disruption after disruption after disruption.

23:29 There are interesting, exceptional areas.

23:32 There's actually seems to be an unusually high amount of continuity

23:35 over many thousands of years in many parts of the Americas,

23:38 which is really fascinating.

23:40 But there too, there are disruptions as well.

23:42 So the pattern seems to be one of for the most part,

23:46 there are periodic profound movements of people coming from somewhere else,

23:52 either completely displacing or mixing with the people who were there before.

23:56 And this idea that any one people is

24:00 somehow purely descended from the people who lived

24:02 in the past and was a common thing

24:05 that people thought prior to the genetic revolution,

24:09 are being confronted with this data.

24:11 It's really an untenable picture.

24:13 It's just it's every time we look, it's not true.

24:16 So the thing that I keep coming back to is that everybody is mixed,

24:22 nobody is pure mixing is happening again and again in the past,

24:26 and it's not something to deny or to fight against,

24:29 but it's it's instead something to embrace as part of our history.

24:37 Chapter two the pattern that wouldn't go away.

24:42 The discovery that modern human ancestors,

24:44 people coming out of Africa after 50,000 years ago or so,

24:48 interbred with these archaic humans of Europe and East Asia,

24:51 the Neanderthals in Europe and western Eurasia,

24:54 and the Denisovans in eastern and southern Asia was really a shock.

24:58 So it was a shock because I was part of the group of geneticists as a graduate

25:04 student and in the community I grew up

25:06 in that had established this out of Africa orthodoxy,

25:09 that all people outside of Africa descend from a small

25:12 group of non-africans who then spread around the world,

25:16 didn't mix with the archaic humans are there,

25:18 but displace them and spread to all the different

25:20 corners of Eurasia and the Americas and Australia.

25:24 So then I came into this project when I was analyzing

25:27 the Neanderthal DNA of being very much biased toward finding support for that.

25:32 But of course, we tested whether there was mixture

25:35 and it was just the data forced our hand.

25:38 It was just very clear that the Neanderthal

25:41 sequences were closer to Non-africans than the Africans.

25:43 And we kept trying to make the results go away.

25:45 I think the results got a lot better than they otherwise would have,

25:49 because we looked at it from every angle saying,

25:51 well, maybe this could be because of this other theory or this other hypothesis.

25:55 Maybe it's not mixture, maybe it's some kind of deeply structured

25:59 modern human population that expands out of Africa, and that's explaining it.

26:03 But we ruled that out, and we looked at it another way,

26:05 and we ruled that out one thing after the other.

26:08 And then like, we got this real sucker punch with the Denisovans,

26:12 and you look at it and like, bang!

26:14 It's related in a special way to people

26:16 from New Guinea on the other side of the world.

26:18 Right.

26:18 A very tropical area, even though this bone is found in Siberia.

26:21 And what's that about?

26:22 And you sort of are confronted,

26:24 this data and you say, I don't know anything anymore, right?

26:26 Like the world is full of secrets about

26:29 the past and how people are related to each other.

26:31 And all my presuppositions and all my biases are probably wrong,

26:35 because every time we get data from somewhere

26:38 that we didn't have data from before, we turn out to be shown to be wrong.

26:42 And then the subsequent data just kept showing example after example.

26:46 After that, Malawi and Cameroon and the Southwest Pacific

26:51 and multiple examples in Europe and different groups in Russia,

26:55 multiple groups in the Middle East, you know,

26:58 the different groups in North Africa, you know, multiple groups in the Americas.

27:02 Just example after example, after example, after example.

27:05 It just seems to be the pattern you really just

27:08 get sort of clubbed over the head with the data.

27:10 And you just have to change your approach to the data,

27:13 the pattern that we found in each case,

27:16 when we get a new ancient genome sequence is that you get

27:21 a DNA sequence and you position it relative to the other sequences.

27:25 You have other ancient sequences, but also modern sequences.

27:29 And again and again you find that this individual

27:33 descends from a mixture of at least two, sometimes more than two,

27:36 ancestral populations that are related differentially

27:39 to other groups you have data from.

27:42 So when it's the case that every time you collect new data,

27:45 you find evidence that this group is mixed,

27:47 the inevitable conclusion from induction is that everybody is mixed.

27:52 You know, at every time that, you know, today,

27:55 you might think that people living in the Americas, North America,

27:59 African-Americans and Latinos are the product of mixtures of continental

28:03 divergent populations due to disruptions in the last 500 years.

28:08 And I'd be the last person to say

28:09 that the last 500 years have not been disruptive.

28:12 But actually there's been disruptive periods in the past

28:14 which brought together equally different groups of people.

28:17 And this has occurred several thousand years ago in the history of Europe,

28:21 and several thousand years ago in the history of South Asia.

28:24 And at each place a very profoundly disruptive at work.

28:28 It's not so unusual from that perspective.

28:31 So once you see example after example, example after mixture,

28:35 you realize that at some time depth that's just occurring all the time.

28:42 So part three, Rethinking human evolution

28:48 models chapter one Ancient Genes today.

28:53 One of the puzzles, the great puzzles, in my opinion,

28:57 that has been thrown up by the ancient DNA revolution,

29:00 has been the one of how modern humans

29:03 and archaic humans are related to each other.

29:05 So we have sequences now from three

29:09 very different groups of humans, from modern humans.

29:12 They like people from sub-Saharan Africa, for example,

29:15 and their non-African descendants, and also Denisovans and Neanderthals.

29:22 We have high quality genome sequences

29:23 from multiple individuals from each of these groups.

29:26 As a fruit of the amazing research that's been done,

29:29 and through a lot of struggle, we know about how they're related to each other.

29:33 We know when these lineages split from each other, mostly,

29:36 and we've identified, in addition to those split, some important mixture events.

29:40 When these lineages interacted with each other at different periods of time.

29:44 But the standard model we have right now is really almost hard to believe.

29:51 So the standard model right now is that modern

29:54 humans are a distant cousin of Neanderthals and Denisovans,

29:59 who are most closely related.

30:01 Neanderthals and Denisovans stemmed from a common

30:03 ancestor maybe half a million years ago,

30:05 which separated earlier from the ancestors of modern humans.

30:08 But actually, maybe there's some alternative way of thinking

30:11 about what happened that can really change our vision

30:15 of the relationships amongst these groups and make

30:18 it sort of the Copernican readjustment where we say,

30:21 well, maybe, maybe the maybe the solar system revolves around the sun,

30:24 and the Earth is a satellite of that.

30:26 And maybe, maybe that's how it looks.

30:28 And so I kind of feel that we're

30:30 missing a trick and that there's something there,

30:32 and that maybe if we come up with a different model,

30:34 it will be much more plausible.

30:36 And so that's one of the exciting, frontiers that I think that we're like,

30:42 hopefully that maybe in the coming years,

30:44 we, we as a community would make progress on Arduino is chock full of function.

30:50 That's the instruction set for our bodies.

30:53 We are composed of our ancestors.

30:55 And if you are a non-African person,

30:57 about 2% of your genes come from Neanderthals.

31:00 Or in some cases, Deniece events, and they're contributing.

31:05 You know, you have 20,000 genes,

31:06 and if it's 2% of them, that's like 400 of them.

31:09 Copies of those 20,000 genes, are going to come from these archaic humans.

31:16 They do stuff.

31:17 And, some of them may be adaptive and important.

31:21 The context for the work that I've been a lot

31:24 of my research has been focused on in the last few years,

31:28 and I'm tremendously excited about right now,

31:30 is to try to use genetics from ancient people to understand how

31:34 people got to way that that be the way they are today biologically.

31:37 So what we've done in our research group in the last few

31:40 years is we've generated data on an absolutely huge number of people.

31:45 So the paper that we're currently publishing has data from more

31:48 than 10,000 ancient people from Europe, in the Middle East.

31:51 And these individuals are from all over

31:53 an all time depths up to 15,000 years ago.

31:56 So the last 10,000 years, this is the time period where in Europe,

32:00 in the Middle East agriculture comes in.

32:02 It's a very eventful period.

32:03 Culturally, it's a period of extreme population

32:05 growth where people began to live differently.

32:07 And we can compare the rate of evolution at the beginning

32:11 of the time period to the end of the time period, and it's gotten faster.

32:14 And so not only has it gotten faster,

32:17 but genetic variants that were under positive natural

32:20 selection before in some cases switch to negative selection.

32:23 So it's a period of oscillation of natural selection.

32:26 So what seems to have happened is that there's cultural change

32:30 that occurs due to differences in the way people live with agriculture,

32:34 living more densely, being close to animals,

32:37 urbanization, other things that are changing.

32:41 And they are prompting changes in the genome that are resulting in, on average,

32:45 changes in frequency of the genetic variants.

32:48 That might be useful for that, perhaps for people

32:50 living in these new and differently challenging environments.

32:54 But if you look at the individual positions that are significant in Europe,

32:58 they tend to move in the same direction in East Asia,

33:00 even though the histories are completely independent in this time period.

33:03 And if you look at the genetic variants that are significantly Asia,

33:07 they tend to move in the same direction in Europe.

33:09 So it's as if these two independent experiments of nature,

33:12 people who are genetically unrelated for the last 30 or 40,000 years,

33:15 when pushed under similar cultural transformations,

33:18 the movement to agriculture increase in population density,

33:22 eventual movement to urbanism.

33:24 The genome responds in the same way and accelerates in its evolution.

33:28 So this is sort of a very striking pattern.

33:31 It's also striking in light of the fact

33:33 that a lot of the genetic literature has suggested

33:36 that evolution has been relatively quiescent over a longer

33:40 time scale of hundreds of thousands of years.

33:42 So it seems like we've entered a period

33:44 of rapid adaptation in response to profound, lifestyle changes.

33:54 Chapter two The Genome in Motion.

33:58 One really striking pattern is a genetic variant at the tick two gene.

34:04 This is a the strongest known risk factor for clinically severe tuberculosis.

34:10 It occurs at a few percent frequency in Europeans today.

34:14 But if you look at the genetic data,

34:15 maybe 6000 or 8000 years ago, it was nearly 0%.

34:19 It almost wasn't there.

34:20 And then it rockets up to 9% or so

34:23 in our data set about 3000 years ago under clear positive selection.

34:27 And then it does a U-turn and it

34:30 goes rockets down to 3% clearly under negative selection.

34:33 So something maybe exposure to tuberculosis as it became

34:38 and Demick might have caused it to decrease in frequency.

34:41 And then something before that some other

34:43 potentially immune related trait may increase in frequency.

34:47 So this is an example of an individual

34:50 variant whose function we know a lot about.

34:52 There's also traits that we can actually that we know about because there

34:57 have been genome wide association studies where

35:00 tens or hundreds of thousands of people,

35:02 or sometimes even millions of people, have been measured for a trait like,

35:08 for example, risk for bipolar disease, manic depressive disorder.

35:14 And people know what genetic variants make

35:16 people more or less susceptible to these traits.

35:19 There might be hundreds or thousands of individual genetic

35:22 changes that measured in hundreds of thousands of people.

35:25 We now have a little incremental effect on the risk for this trait.

35:29 And if you put them all together,

35:31 you get a moderately good predictor for this trait.

35:33 Genetically better, for example,

35:35 that knowing whether your sibling has the disease, for example,

35:38 or better, for example,

35:39 than knowing your socioeconomic sort of situation in life.

35:43 So what we can do is we can

35:44 take this genetic predictor of risk for bipolar disease

35:47 and ask whether it's shifted over time in a way

35:50 that's too systematic to be due to natural selection.

35:54 So that's the test we implemented.

35:56 And for bipolar disease, we see there has been a downshift over time

36:00 in the clinically predicted risk for bipolar disease by a huge amount,

36:05 about one standard deviation almost on the scale

36:08 of modern genetic variation, which is a lot.

36:11 And clearly what manifests itself today as risk

36:15 for bipolar disease was negative in the past,

36:18 but it wasn't negative very far back in the past

36:20 because otherwise it would not been at the high,

36:22 high, high frequency it was at that time.

36:25 Now what was being selected is not obvious,

36:28 but some genetic propensity to this has been under changed,

36:32 selection over time and has shifted radically over time.

36:35 So that's one of about 100 complex traits

36:38 that we know about through genome wide association studies whose,

36:41 change over time is very statistically significant

36:45 and that we can watch evolve in this way.

36:47 We're changing all the time.

36:48 It's more evidence that we're changing.

36:55 Part for what genetics does and doesn't say about difference.

37:01 Chapter one archeology, ancestry and Political myth.

37:07 What do people fear about what genetics might

37:10 reveal about nature of differences across human populations?

37:14 I think that part of this is ideological.

37:17 So when we were working on our work and on on our studies in Europe,

37:21 for example, between 5000 and 3000 years ago,

37:24 and found this huge ancestry transmission associated with the Yamnaya spread

37:29 in this and the cultures locally known as the Corded Wear or the Bell Beaker.

37:33 And we showed that there was a huge disruption in the ancestry.

37:37 I think it was very surprising in light of the discussion that had

37:41 happened within the discipline of archeology

37:44 for the last 80 years prior to that.

37:47 And so a potted history of this is something like as follows.

37:51 At the beginning of the 20th century,

37:53 several archeologists had begun to dig across

37:57 large areas of Europe and had done it, begun to do it scientifically,

38:00 and noticed that the same types of pots appeared at what

38:03 looked like at about the same time in different places.

38:06 So the corded, where is a particular way of making

38:09 pots and burying people with certain types of artifacts.

38:12 And it appears, oh, first,

38:13 in northeastern Europe and then spreads over a pretty wide area,

38:17 eventually getting even to places like the Netherlands,

38:20 and almost at the same time,

38:22 maybe a little later, around the same time, another burial style,

38:25 another way of making pots known as Bell beaker spreads,

38:28 with starting in Iberia in Spain and moving northward and eastward.

38:33 And together these spread over Europe.

38:35 They're often called horizons as, transformative archeological change.

38:40 And several people, including the German archeologist Gustav Cassano,

38:44 thought that what you're seeing here is the spread of people.

38:47 So there's a new people moving.

38:49 They make pots in a certain way.

38:50 They make other things in a certain way.

38:52 They have a religion or a culture that they bring with them.

38:55 And you can recognize the spread of that people

38:58 through the, types of pots that they're making.

39:01 And they connected this to the spread of Indo-European languages,

39:04 which people had been wondering about since the late 1700s,

39:08 why they were all connected to each other.

39:09 And they said, maybe these are the Indo Europeans.

39:11 And they argued then, the, the school,

39:16 the German school argued that actually these are the end of Europeans

39:19 and the homeland of the Indo Europeans is in northern Europe,

39:23 the place where the Aryans are using

39:25 the language of the Sanskrit and erotic epics.

39:29 And this is the ancient homeland of the Germans,

39:32 and the Nazi state should expand into the area

39:34 that was previously occupied by the Aryan homeland and so on.

39:37 So this archeology was co-opted for political purposes

39:42 by the Nazis after the Second World War.

39:44 There was a strong reaction to the misuse of archeology for political purposes.

39:49 The justification of national claims to land, to territory,

39:53 to heritage based on, reconstructions of the past

39:58 and archeologists challenged the logic that had

40:01 been used to argue that these horizons

40:04 and similarities of parts corresponded to movements of people.

40:07 They pointed out, reasonably, that people could copy from each other.

40:11 For example, if you have a cell phone and you're in Korea,

40:14 doesn't mean that you descended genetically or closely related genetically

40:18 to the people in California who first started developing smartphones.

40:23 The truth is, is that genetics is an important

40:27 predictor of how people differ from each other.

40:29 I think that it's something that scientists are in general scared to talk about.

40:33 They want to just not talk about the nature

40:36 of differences across groups of people or groups of individuals,

40:39 because it's such an anxiety provoking topic.

40:42 But in the species that we live in, there's

40:45 so much variability across individuals that there is no sameness.

40:49 I wish we were all the same.

40:50 But we just have to, as experts on this topic,

40:55 have a conversation about this, because if we don't do so,

40:58 we're going to leave a vacuum that will get filled with pseudoscience.

41:01 I think the effect of studying the nature of human history and the history

41:09 and have mixture and finding who lived in the past in one place,

41:14 is consistently showing that the history is not what people thought it was.

41:20 That myths of purity are, almost always wrong.

41:24 And I think it's generally argued against racist

41:28 and nationalistic narratives and, sort of supported better,

41:33 or the idea of a more,

41:35 pluralistic and tolerant world certainly gives us a lot of perspective.

41:40 So, for example, I think that it should give

41:44 us humility about whether our story about ourselves is,

41:47 is the correct story ancient DNA has disrupted

41:51 a number of assumptions about who we are.

41:54 Simple migration out of Africa, displacing the archaic humans who were there.

41:58 That's out the door now.

42:00 And it has also changed our understanding of the pattern of human evolution,

42:05 as one of simple branching to one.

42:08 That's definitely a trellis of a mixing.

42:11 And mixing is again, over time.

42:13 It's provided some little insight about the nature

42:17 of how natural selection has shaped our species over time.

42:20 It's showed that, for example, our genome is malleable,

42:24 in response to environmental changes like

42:27 the ones that have occurred in the last, ten or 15 or 5000 years.

42:31 And so I think it adds understand nuance to our understanding of how

42:36 our species is changing and makes it

42:39 clear that it's continuing to change going forward.

42:42 It sort of shows how we're all connected in myriad ways we never imagined.

42:47 I think the things that have been amazing about

42:49 this field has been the discovery of DNA from archaic humans,

42:54 including from archaic humans we didn't even know existed before,

42:57 like Denisovans, who mixed with each other and with modern

43:00 humans in ways we didn't know about before,

43:03 which really changes our understanding

43:04 of the relationship of modern to archaic humans.

43:07 So that's been, like,

43:08 revolutionary in terms of our understanding of our origins.

43:11 Another thing has been the ability to, learn about how the hunter

43:18 gatherers of each region in the world today relate to people today,

43:22 and to realize that it's almost always

43:23 the case that people who are hunter gatherers

43:26 in each region are not the primary ancestors

43:29 of the people who live in the same region.

43:31 And even the first farmers are rarely the direct

43:33 ancestors of the people who live in each place today.

43:36 That is the past is a really, as long as the cliche goes,

43:40 an undiscovered country and is profoundly different

43:43 from the one we live in today,

43:44 and we can't guess at it without actually having real data about it.

43:48 From the perspective of how those people relate to people today.

43:52 So I think that's been another big picture result.

43:55 Another big picture result has been that sex biased mixture,

44:01 with men and women from different groups participating radically differently.

44:06 And some of these mixture events is pretty common in the past.

44:09 What this means we don't know.

44:11 In some cases it's evidence of probably extreme unhappiness and displacement.

44:17 So one of the examples that we know from pre-history is in Spain,

44:22 about 4000 to 4500 years ago.

44:25 There's a movement of these steppe pastoralists to drive people into Spain.

44:29 It's about 40% of the ancestry gets displaced by people coming from the east,

44:34 but 100% of the Y chromosomes come from east.

44:36 These are translate so which means

44:38 that the local man completely failed to reproduce.

44:40 So this must be very disruptive event for the local

44:43 man in Spain for a 5004 to 4500 years ago.

44:47 What happened?

44:48 We don't know.

44:48 But you can actually see from these processes

44:52 a lot about inequality in the past and about,

44:55 the different participation of different groups

44:58 of people and events in the past.

45:00 So that's been another thing that we learn about.

45:03 We learn about sort of whole patterns of, of, of living in the past

45:08 that were not vivid before and that we really couldn't appreciate before.

45:15 Chapter two.

45:16 How is this useful?

45:18 I had, experience, I think, in the 2000 when I was giving a talk,

45:23 I took to a class for undergraduates at MIT and I gave this talk about,

45:30 I think, human chimpanzee evolution.

45:32 So I had published this paper,

45:34 which remains my favorite paper that I've ever published to date.

45:37 In 2006, that compared human and chimpanzee and gorilla genome sequences.

45:42 So I had just become available.

45:43 These are my closest relatives and also to the orangutan sequence,

45:48 which is the next closest and found evidence that the human

45:52 and chimpanzee lineages initially separated maybe 5 to 7 million years ago.

45:57 And then remained separate for a long

45:59 time and then remixed before finally separating again.

46:03 So and that either humans or chimpanzees

46:05 are the result of this major hybridization event.

46:08 And I remain I continue to think that the evidence for this is very powerful.

46:11 In fact, it's even more powerful now than it was at the past.

46:15 So I was sort of excitedly telling this class at MIT,

46:17 this story and the evidence for it,

46:20 and trying to convince them that they should be convinced by the evidence.

46:23 And this one in undergraduate at the front,

46:26 just looked me in the face at the end and say,

46:28 how is this useful to me in the future?

46:31 And I was like, I was like, I just didn't have a good answer.

46:37 And like, I think the answer I would have liked to give is that, you know,

46:44 not everything has to be justified for medicine

46:49 or how it's going to improve our health or how it's going to be used to further

46:56 industrial or military technology or mining or whatever,

47:01 that the pursuit of knowledge for its own sake,

47:04 whether it's in philosophy or in poetry or in literature

47:09 or in theology or in science or in understanding cosmology,

47:14 is sort of like a jewel in our crown as a species, I say.

47:19 I wish I had said something like that, but I didn't say it fast enough.

47:22 You know, sometimes there's value to these things

47:25 and something that you do is useful.

47:26 Probably this work we're doing,

47:28 probably the work that my research group will have done,

47:31 is doing will pay for itself in terms

47:33 of its medical application because of this natural selection work,

47:37 because actually, as you learn over time

47:39 about how genetic change variants change over time,

47:42 it's telling you about how the risk factors strategy

47:45 has got to be the way they are today.

47:47 And it's already clear this has relevance for understanding that.

47:50 So that by itself that probably justifies that.

47:52 But that's kind of like saying that the reason we should go

47:55 to the moon is to like have space ice cream or whatever it is,

47:58 and freeze drying or something like this, like,

48:01 you know, like like there should be some value

48:03 in the world to just learning about things for their own sake,

48:07 without trying to have some kind of utilitarian justification for it.

48:11 I think that in the end,

48:13 this pursuit of knowledge for its own sake has spinoffs.

48:16 And, you know, and probably my own training is that I

48:20 trained in a medical genetics laboratory for part of my training.

48:24 And I think it's a reflex to try to make things relevant to medicine.

48:27 So I don't even think I'll be able to get away

48:28 with not somehow making the connections

48:31 to medical genetics and medical research.

48:33 But I don't think I should need to do that.

48:35 I think that some of this work is just interesting in its own right,

48:38 and it's important, and it certainly has social resonance.

48:42 I think it forces people to think in new ways and rethink their prejudices.

48:45 And maybe that's a fundamentally good thing.