Genetic investigations of Upper Palaeolithic Europe have revealed a complex and transformative history of human population movements and ancestries, with evidence of several instances of genetic change across the European continent in the period following the Last Glacial Maximum (LGM). Concurrent with these genetic shifts, the post-LGM period is characterized by a series of significant climatic changes, population expansions and cultural diversification. Britain lies at the extreme northwest corner of post-LGM expansion and its earliest Late Glacial human occupation remains unclear. Here we present genetic data from Palaeolithic human individuals in the United Kingdom and the oldest human DNA thus far obtained from Britain or Ireland. We determine that a Late Upper Palaeolithic individual from Gough’s Cave probably traced all its ancestry to Magdalenian-associated individuals closely related to those from sites such as El Mirón Cave, Spain, and Troisième Caverne in Goyet, Belgium. However, an individual from Kendrick’s Cave shows no evidence of having ancestry related to the Gough’s Cave individual. Instead, the Kendrick’s Cave individual traces its ancestry to groups who expanded across Europe during the Late Glacial and are represented at sites such as Villabruna, Italy. Furthermore, the individuals differ not only in their genetic ancestry profiles but also in their mortuary practices and their diets and ecologies, as evidenced through stable isotope analyses. This finding mirrors patterns of dual genetic ancestry and admixture previously detected in Iberia but may suggest a more drastic genetic turnover in northwestern Europe than in the southwest.

Cool paper that shows that the British can still get some things done. Basically, they found that genetically and culturally there were really two different populations in late Pleistocene Europe, and that the earlier post-Magdelenaian populations left some impact on the mostly Villabruna-descended populations like Cheddar Man.

Basque against the Romans

Posted on March 29, 2021March 29, 2021 by Razib Khan

Genetic origins, singularity, and heterogeneity of Basques:

Basques have historically lived along the Western Pyrenees, in the Franco-Cantabrian region, straddling the current Spanish and French territories. Over the last decades, they have been the focus of intense research due to their singular cultural and biological traits that, with high controversy, placed them as a heterogeneous, isolated, and unique population. Their non-Indo-European language, Euskara, is thought to be a major factor shaping the genetic landscape of the Basques. Yet there is still a lively debate about their history and assumed singularity due to the limitations of previous studies. Here, we analyze genome-wide data of Basque and surrounding groups that do not speak Euskara at a micro-geographical level. A total of ∼629,000 genome-wide variants were analyzed in 1,970 modern and ancient samples, including 190 new individuals from 18 sampling locations in the Basque area. For the first time, local- and wide-scale analyses from genome-wide data have been performed covering the whole Franco-Cantabrian region, combining allele frequency and haplotype-based methods. Our results show a clear differentiation of Basques from the surrounding populations, with the non-Euskara-speaking Franco-Cantabrians located in an intermediate position. Moreover, a sharp genetic heterogeneity within Basques is observed with significant correlation with geography. Finally, the detected Basque differentiation cannot be attributed to an external origin compared to other Iberian and surrounding populations. Instead, we show that such differentiation results from genetic continuity since the Iron Age, characterized by periods of isolation and lack of recent gene flow that might have been reinforced by the language barrier.

The main takeaway seems to be that Basque distinctiveness dates to the Roman period.

Demographic Reticulation!

Posted on September 21, 2020September 21, 2020 by Razib Khan

I made an offhand comment on Twitter that it would take a few decades for the public to understand the insight that demographic reticulation is ubiquitous in human prehistory. This is probably the biggest claim in Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past.

Well, it turns out that that is a new phrase. I confirmed it myself. If Google is perplexed, it must be a novelty.

As many of you know ‘populations’ are constructs. Really what’s happening is that we’re tracing the genealogies of genes, and those summary genealogies give us insights into population splits and mixture.

Reticulation is a pretty general concept in evolution and genetics. I didn’t really know what other terms to give the dynamic I was alluding to.

What other term describes the process you see in the admixture graph below?

Northeast Africans before and after the Bantus

Posted on September 6, 2020September 6, 2020 by Razib Khan

One of the major changes in our understanding of human genetic demographic patterning across the world over the last twenty years is that we no longer believe most of the extant patterns date to the Last Glacial Maximum ~20,000 years ago. If you do not think this was a view held a generation ago, read Stephen Oppenheimer’s The Real Eve: Modern Man’s Journey Out of Africa. In a broad sense, the current understanding is reflected in 2014’s Toward a new history and geography of human genes informed by ancient DNA. There has been a lot of change, and a lot of admixture.

One of the regions subject to this change has been Africa. The expansion of the Bantu-languages across much of the continent seems to have been accompanied by widespread genetic replacement. Groups like the Pygmies of the Congo rainforest, the Hadza of Tanzania, and the various Khoi and San people of southern Africa, reflect a broad underlying genetic diversity that was to a great extent erased. Though it is correct that African people as a whole exhibit more genetic diversity than all other people in the world, when looking at genetic variation across peoples (as opposed to within them), Bantu Kenyans and people in Malawai are quite similar.

The reason for this is straightforward. The differentiation of Bantu peoples dates to the last four to one thousand years. This is not a great deal of time for genetic drift to accumulate differences. Rather, a primary avenue for a variation to occur is through admixture with the local substrate, which as noted above did not occur in much of the range of the expansion.

All that being said, the landscape that the Bantus expanded into obviously wasn’t empty. And it wasn’t just hunter-gatherers. Ancient DNA and cultural anthropology make it obvious that Cushitic-origin pastoralists occupied large zones of East Africa. They interacted with the Khoi pastoralists of southern Africa at some point, transmitting pastoralism (and some genes!). But to the northeastern edge of the Bantu range, there were other people, who did not speak Afro-Asiatic languages. These are what we today call “Nilo-Saharan” populations.

As can be seen on any language map, the Nilo-Saharan people seem to be squeezed between two expanding cultural zones: the Afro-Asiatic and Bantu. But after the decline of Cushitic speakers in East Africa (due to absorption into expanding Bantus), Nilo-Saharan peoples like the Masai seem to have moved southward, and specialized in pastoralism. Nilo-Saharans are a diverse bunch, and I’m not sure how ethno-linguistically coherent they are. Some of them, like the Masai, seem to have assimilated Cushitic pastoralists. Others, such as the Dinka of Sudan, have almost no Eurasian ancestry.

A new paper in AJHG, High Levels of Genetic Diversity within Nilo-Saharan Populations: Implications for Human Adaptation, uses whole-genome sequencing to look at the relationships of these peoples to others in Africa. The abstract:

Africa contains more human genetic variation than any other continent, but the majority of the population-scale analyses of the African peoples have focused on just two of the four major linguistic groups, the Niger-Congo and Afro-Asiatic, leaving the Nilo-Saharan and Khoisan populations under-represented. In order to assess genetic variation and signatures of selection within a Nilo-Saharan population and between the Nilo-Saharan and Niger-Congo and Afro-Asiatic, we sequenced 50 genomes from the Nilo-Saharan Lugbara population of North-West Uganda and 250 genomes from 6 previously unsequenced Niger-Congo populations. We compared these data to data from a further 16 Eurasian and African populations including the Gumuz, another putative Nilo-Saharan population from Ethiopia. Of the 21 million variants identified in the Nilo-Saharan population, 3.57 million (17%) were not represented in dbSNP and included predicted non-synonymous mutations with possible phenotypic effects. We found greater genetic differentiation between the Nilo-Saharan Lugbara and Gumuz populations than between any two Afro-Asiatic or Niger-Congo populations. F3 tests showed that Gumuz contributed a genetic component to most Niger-Congo B populations whereas Lugabara did not. We scanned the genomes of the Lugbara for evidence of selective sweeps. We found selective sweeps at four loci (SLC24A5, SNX13, TYRP1, and UVRAG) associated with skin pigmentation, three of which already have been reported to be under selection. These selective sweeps point toward adaptations to the intense UV radiation of the Sahel.

There are few moving parts in this paper. The three primary ones involve genomics, phylogenetics, and evolutionary and population genetics. Basically, discovering new variation, genetic relationships, and the possibility of adaptation.

As noted in the paper these are under-sequenced populations. Using deep-sequencing they were able to discover variation particular to Nilo-Saharans, more than three million novel variants. This is unsurprising. As a stylized fact non-Africans diverged a bit over 50,000 years ago. If you look for new variants, you’ll find some, but there’s only so much new variation that could accumulate in 50,000 years. In contrast, African populations have lineages that are 100,000 to 300,000 years diverged. This means there is a lot of time for the variation to accumulate (also, less variation lost through bottlenecks due to larger effective population sizes).

Nilo-Saharans are particularly interesting because many people suspect that they are the closest Sub-Saharan Africans to the ancestors of the people who expanded outside of Africa 50-60,000 years ago. Part of this is geography, but there are suggestive signs in the genetics this is true (though perhaps the Hadza are the closet to non-Africans?). The main issue with these sorts of suppositions is that the real population history is always more complex than stylized models.

African genomics tells us about deep structure and history

Posted on April 27, 2019April 28, 2019 by Razib Khan

Two interesting papers in Genome Biology that are open access, Whole-genome sequence analysis of a Pan African set of samples reveals archaic gene flow from an extinct basal population of modern humans into sub-Saharan populations and African evolutionary history inferred from whole genome sequence data of 44 indigenous African populations. Since they are open access you should just read both of them.

I believe they are the first in a series of papers over the next few years using whole-genome analysis to understand the population structure within Africa, and how it relations to the people who branched off from Africans. Eventually, this will also lead to research focused on medical and population genomics, looking at characteristics and forces beyond phylogeny.

Let the genomic die fly!

Posted on April 19, 2019April 19, 2019 by Razib Khan

A new “polygenic risk score” (PRS) paper is making some waves, Polygenic Prediction of Weight and Obesity Trajectories from Birth to Adulthood. Since it is open access I suggest you read it.

But basically, they took ~2 million common variants (there are about ~100 million common variants in the world population) in ~300,000 individuals in 4 cohorts, and used it to predict weight. A genome-wide polygenic score statistic. The correlation with BMI of the score is 0.29. This is pretty modest. But it seems to me that the biggest and most important finding is that it seems to capture a lot of the people at the tails of the distribution.

I’m becoming more and more convinced that the best things these PRS scores can do in the near-term is to identify people who are possibly at these tails. In a complex trait context, the tails are where for diseases a lot of the people who are going to have issues later in life exist. People with BMI in the range 25-30 may have a modest increase in risks, but someone who is very obese, with BMI above 35, is at much greater risk. Over 40% of the people in the top decile here were obese. Only 10% of people in the bottom decile were.

This research comes out of the context of earlier work on the heritability of BMI. It’s around 0.75 or so. That means it runs in families. Combined with the fact that in the recent past, or in other nations, there is a great variation in median size and distribution, one can intuit that genetic dispositions and environmental context both help explain the variation we see around us. The modern American environment is clearly obesogenic. When most of the American population were involved in physical jobs on farms the environmental context was very different.

Over the next few years, there risk scores for BMI will get better, and expand to other populations. One thing that some people are pointing out is that we know it’s heritable, so why not just look at your family? As many of you know, Mendelian segregation means that siblings may have quite different risk profiles on the genomic level. Polygenic risk score prediction is I think going to be extremely interesting and informative in the case of traits which are known to be found within families across generations (e.g., autism), but don’t seem to impact everyone. Perhaps we’ll find for a given characteristic expression is random, due to some life event or cofactor such as infection. Or perhaps we’ll find that differences among siblings have some genetic basis in variants inherited from parents?

Addendum: One of the authors, Sek Kathiresan, has been answering questions on Twitter.

What did modern humans look like during the “Out of Africa” event?

Posted on April 16, 2018April 16, 2018 by Razib Khan

Recently I was having an email exchange with a friend (a prominent public intellectual who is not a scientist), and we were thinking about what “ancestral Africans” looked like. More precisely, the populations which were resident around ~100,000 to ~200,000 years before the present. These are the people who are depicted in paleoanthropology documentaries. Here were some of my major contentions:

1) We don’t know what they looked like
2) They probably were more likely to look like modern Africans than non-Africans
3) But modern Africans are diverse in their looks and we could expect that ancient Africans were too

The neighbor-joining tree above is generated with a naive model of successive bifurcation.

1) Khoisan split off 200,000 years ago
2) Mbuti split off 150,000 years ago
3) Mende split off 100,000 years ago
4) Japanese about 50,000 years ago
5) While Pathan and Basque only 15,000 years ago

The model is wrong in the details. Pathan and Basque have some ancestry is which recently diverged, and much that is deeply diverged. The 15,000 year value is just an average. Similarly, the Khoisan have some Eurasian ancestry. But in the broad sketch it illustrates that some African populations diverged a very long time ago from other groups.

Ancient Africans date to ~200,000 years before the present for all the modern populations. Khoisan to Japanese. You could probably use phylogenetic character reconstruction methods to attempt to infer what ancient Africans looked like…but I’m not sure that it would be useful since modern humans have spread over so many ecologies over such a short span of time.

Outside of Sub-Saharan Africa perhaps on the order of 95% of the ancestry derives from an expansion from a small founder group between 60 and 80 thousand years ago. Removing the “Basal Eurasian” component, groups as diverse as Native Americans, Oceanians and East Asians probably derive their ancestry from a common group which flourished between 50 and 60 thousand years ago (this pulse is the majority of the ancestry of Europeans and South and West Asians as well).

The point here is to illustrate that 50,000 years is definitely sufficient for a great deal of diversity to have emerged in human physical variation. And yet the Khoisan are ~200,000 years diverged from their ancestors within Africa. We actually know that indigenous southern Africans have been selected for lighter pigmentation. We also know that loci associated with pigmentation in modern humans exhibits a lot of variation in Africans, and this variation is likely an ancestral feature of our species.

In sum, the number of generations between ancestral Africans and all modern descendent populations is great enough that I’m not uncertain that we can predict what they look like in anything except their skeletal features. Additionally, most of the history of anatomically modern humans was likely highly structured within Africa. That’s another way of saying that ancient Africans themselves were probably physically diverse.

With all that being said, all things equal ancient Africans probably are more likely to look like modern Africans than modern non-Africans. The main reason is simply that modern Africans occupy the same broad ecological landscape as ancient Africans, and many of our features, from our build to our complexion seem dependent upon environmental pressures. There’s lot of evidence that very light skin is probably a derived characteristic of our species (there are consistent signatures of sweeps around pigmentation loci). And, there is also evidence that some of the archaic introgression into non-Africans may have consequences in our morphology and external physical characteristics. For example, Eurasians seem to have very high frequencies of Neanderthal variants of the keratin gene. This is implicated in hair, skin and nail development.

Addendum: Note that even if we have ancient genomes, polygenic characteristics are still hard to predict. Even today common SNPs only explain a minority of the variation in hair color in Europeans.

Who We Are and How We Got Here, a book worth reading

Posted on February 28, 2018February 28, 2018 by Razib Khan

Yesterday I talked to a friend who has a review copy of Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past. They gave me a preview (their overall assessment was positive).

I haven’t personally asked to get a copy because, to be honest, I thought there wouldn’t be anything new in it. If you “read the supplements” what more could there be in 368 pages? So I was waiting until the end of the month to buy the book and read it in my own sweet time as due diligence.

Well, this morning I asked a publicist to send me a copy. I will be getting it next week. The reason is that I’m told the latter portions of the book are quite challenging and candid as to what genetics may tell us in the 21st century. Who We Are and How We Got Here is a 21st-century revision and update of The History and Geography of Human Genes. But it’s apparently a lot more.

Also, I make a small cameo in the book, as does Eurogenes and Dienekes. I have always appreciated how the David Reich and Nick Patterson and their whole lab has taken people outside of the halls of the academy seriously. They didn’t need to as a matter of professional necessity but often engage as a matter of decency and seriousness.

Another great-great-great…great-uncle in Asia

Posted on October 15, 2017October 15, 2017 by Razib Khan

The paper which surveys the relationship of the 40,000 year old Tianyuan sample is finally out in Current Biology, 40,000-Year-Old Individual from Asia Provides Insight into Early Population Structure in Eurasia. There isn’t anything too surprising here. Here is the part of the abstract that presents new finding:

…we generated genome-wide data from a 40,000-year-old individual from Tianyuan Cave, China…We find that he is more related to present-day and ancient Asians than he is to Europeans, but he shares more alleles with a 35,000-year-old European individual than he shares with other ancient Europeans, indicating that the separation between early Europeans and early Asians was not a single population split. We also find that the Tianyuan individual shares more alleles with some Native American groups in South America than with Native Americans elsewhere, providing further support for population substructure in Asia [8] and suggesting that this persisted from 40,000 years ago until the colonization of the Americas. Our study of the Tianyuan individual highlights the complex migration and subdivision of early human populations in Eurasia.

The Tianyuan sample lived about ~40,000 years ago in China, and it does not seem to have been the direct ancestor of modern East Eurasians. It also seems to have had some relationship to the Australo-Melanesian affiliated population which contributed ancestry to the indigenous peoples of South America. Additionally, it also shares ancestry above what you’d expect with a 35,000 year old Paleolithic European, the GoyetQ116-1 sample, which is found in an Aurignacian context.

There are some direct conclusions that one can infer from this paper. First, as known beforehand the divergence between East Eurasians and West Eurasians has to predate 40,000 years before the present since this sample already shares drift with East Eurasians far more than West Eurasians. In the paper, the authors give an interval of 40,000 to 80,000 years before the present, which seems advised. Remember that “Basal Eurasians” separated before the divergence of East and West Eurasians.

Second, “ghost” populations were common. There are at minimum two ancient Eurasian populations, represented by the Oase1 sample in Romania from 40,000 years ago, and the 45,000 year old Ust’-Ishim from Siberia, who were not closely related to any populations which left descendants today.

Third, the human “family tree” looks more like a human “family bramble.” One of the interesting points in this paper is that Tianyuan shares drift with Goyet, but does not share drift with El-Miron, which seems to be descended in large from a population like Goyet. The key here is to note that Goyet is the closest proxy to some of the ancestors of El-Miron, but it may not be the ancestor at all. So if Goyet-like populations were heterogeneous in relation to East Eurasian, then El-Miron may descend from a group which never mixed with East Eurasians.

This is clear when you read many of these ancient DNA papers closely. The Mal’ta boy was representative of a population which contributed to both Northern Europeans (via Eastern Hunter-Gatherers) and Amerindians, but the deeper results also indicated that the common contributor to these populations was not the Mal’ta population, but related to them. That is, there is no expectation that the sparse sampling of ancient DNA in many regions and epochs will find the ancestral populations, as opposed to groups related to the ancestral populations.

This is a looking-through-the-glass-darkly situation. The true pattern of population relationships of the past needed to be inferred from a finite set of individuals randomly drawn from those populations. If most of those populations left no descendants due to common and repeated local extinction events, then it may be that most of the time we’re going to have to triangulate to the “true” ancestral groups, who left descendants simply due to luck.

Finally, this should really put the nail in the coffin of the idea that we can think of ancient populations are algebraic recombinations of modern populations. Modern groups almost certainly sample only a small part of the distribution of ancient populations.