Razib Khan One-stop-shopping for all of my content

June 6, 2017

Origin of modern humanity pushed back 260,000 years BP (?)

Filed under: Ancient DNA,Genetics,Khosian,South Africa — Razib Khan @ 12:45 am

The above figure is from a preprint, Ancient genomes from southern Africa pushes modern human divergence beyond 260,000 years ago. The title and abstract are pretty clear:

Southern Africa is consistently placed as one of the potential regions for the evolution of Homo sapiens. To examine the region’s human prehistory prior to the arrival of migrants from East and West Africa or Eurasia in the last 1,700 years, we generated and analyzed genome sequence data from seven ancient individuals from KwaZulu-Natal, South Africa. Three Stone Age hunter-gatherers date to ~2,000 years ago, and we show that they were related to current-day southern San groups such as the Karretjie People. Four Iron Age farmers (300-500 years old) have genetic signatures similar to present day Bantu-speakers. The genome sequence (13x coverage) of a juvenile boy from Ballito Bay, who lived ~2,000 years ago, demonstrates that southern African Stone Age hunter-gatherers were not impacted by recent admixture; however, we estimate that all modern-day Khoekhoe and San groups have been influenced by 9-22% genetic admixture from East African/Eurasian pastoralist groups arriving >1,000 years ago, including the Ju|’hoansi San, previously thought to have very low levels of admixture. Using traditional and new approaches, we estimate the population divergence time between the Ballito Bay boy and other groups to beyond 260,000 years ago. These estimates dramatically increases the deepest divergence amongst modern humans, coincide with the onset of the Middle Stone Age in sub-Saharan Africa, and coincide with anatomical developments of archaic humans into modern humans as represented in the local fossil record. Cumulatively, cross-disciplinary records increasingly point to southern Africa as a potential (not necessarily exclusive) ‘hot spot’ for the evolution of our species.

These results in the outlines were actually presented at a conference. I saw it on Twitter and don’t remember which conference anymore. But this is not entirely surprising.

First, much respect to Mattias Jakobsson’s group for breaking through the Reich-Willerslev duopoly. Hopefully this presages some democratization of the ancient DNA field as expenses are going down.

Second, notice how in most cases ancient DNA shows that modern reference populations turn out to be admixed. This was the problem with much of Eurasia, and why using modern genetic variation to make inferences about the past totally failed.

I am entirely convinced that the genome from Ballito Bay dating to ~2,000 years does not carry the Eurasian inflected East African admixture. The Mota genome implies that Eurasian admixture did not come to eastern Africa much before 4,500 years ago. There needs to be a much deeper big picture analysis of the archaeology of Africa and the genetic information we have to get a sense of what happened back then…but, it seems likely that the Bantu migration has over-written much of the earlier genetic variation.

The fact that ancient genomes always show that our current populations are admixed makes me wonder if the Ballito Bay sample itself is admixed from more ancient populations. That is, if we found a genome from 20,000 years ago, would it be very different from the Ballito Bay samples? The relatively thick time transect from Europe indicates that turnover happens every 10,000 years or so. Australian Aborigines seem to have been resident in their current locations for ~50,000 years, but this seems the exception, not the rule. Do we really think that the ancestors of the Bushmen were living in southern Africa for five times as long as Australian Aborigines?

Another curious aspect of this paper is that it suggests the effective population size of Bushmen is smaller than we might have thought, and they’re somewhat less diverse than we’d thought. That’s because East African (with Eurasian ancestry) gene flow increased heterozygosity, as well as inferred effective population sizes. I’ve mentioned this effect on statistics before. Unless you have a true model of population history (or close to it) your assumptions might distort the numbers you get.

There is another aspect to this preprint mentioned glancingly in the text, and a bit more in the supplements: they seem to only be able to model Yoruba well if you assume that they themselves are a mix of “Basal Humans” (BH) and other African population which gave rise to East Africans and “Out of Africa” populations. Note that the BH seem to diverge from other human populations before the ancestors of Southern Africans like the Ballito Bay sample. That is, BH could push the diversification of the ancestors of modern humans considerably before 260,000 years before the present.

The possibility of deep structure in the Yoruba is pretty notable because they’ve been the gold standard in many human population genetic data sets as a reference population. But this is not result of deep structure is not entirely surprising. For years researchers have been hinting at confusing results in relation to the possibility of Eurasian back-migration. Perhaps the deep structure was confounding inferences?

The authors themselves are quite cautious about their dating of the divergence. It’s sensitive to many assumptions, and in particular the mutation rate being known and constant over time. But I think it’s hard to deny that this is pushing back the emergence of modern humans beyond what we know today. The earliest anatomically modern humans are found in Ethiopia 195,000 years ago from what I know. As I said, I’m convinced that the ancient genome has shown that modern “pristine” populations have some serious admixture. But I’m not as convinced about any specific point estimate, because that’s sensitive to a lot of assumptions which might not hold.

Finally, first a quick shout out to the blogger Dienekes. As early as ten years ago he anticipated the basic outlines of these sorts of results in the generality, if not the details. We really have come a long way from popular science declaring that all humans descend from a small group of East Africans who lived 50,000 to 100,000 years ago. The real picture was much more complex.

Also, I have to admit I considered titling this blogspot “Wolpoff’s revenge.” As in Milford Wolpoff. The reason being that we’re getting quite close to territory familiar to the much maligned multi-regionalist model of modern human origins.

Note: These findings should make us less surprised perhaps by a “modern” human migration before the primary one out of Africa.

June 15, 2011

The Cape Coloureds are a mix of everything

A Cape Coloured family

I’ve mentioned the Cape Coloureds of South Africa on this weblog before. Culturally they’re Afrikaans in language and Dutch Reformed in religion (the possibly related Cape Malay group is Muslim, though also Afrikaans speaking traditionally). But racially they’re a very diverse lot. In this way they can be analogized to black Americans, who are about ~75% West African and ~25% Northern European, with the variance in ancestral proportions being such that ~10% are ~50% or more European in ancestry. The Cape Coloureds though are much more complex. Some of their ancestry is almost certainly Bantu African. This element is related to the West African affinities of black Americans. And, they have a Northern European element, which likely came in via the Dutch, German, and Huguenot settlers (mostly males). But the Cape Coloureds also have other contributions to their genetic heritage. Firstly, they have Khoisan ancestry, whether from Bushmen or Khoi. This is well known in their oral memory. The the hinterlands of the Cape of Good Hope are beyond the ecological range of the Bantu agricultural toolkit, so the region was still dominated ...

April 27, 2010

The ancestry of one Afrikaner

Filed under: Afrikaner,Anthroplogy,Genetics,Pedigree,science,South Africa — Razib Khan @ 5:41 am

A few weeks ago I reviewed a paper on the the genetics of the Cape Coloured population. Within it there was a refrence to another paper, Deconstructing Jaco: genetic heritage of an Afrikaner. The title refers to the author himself. It was an analysis of his own pedigree going back to the 17th century, along with his mtDNA, his father’s mtDNA, and his Y lineage. The genetics is a bit thin, but the pedigree information is of Scandinavian quality from what I can tell. Praised the records of the Reformed Church!

The author’s utilizes an inversion of the typical method whereby a survey of a population may give some insight into individuals within that population. Rather, he leverages the thorough church records of his Afrikaner community, and his local roots, to paint a picture of his own ancestry. Then he compares the results to those of the community as a whole. Though an N of 1 certainly has limits it seems that the author concludes that he is relatively representative because some of the statistics that emerge out of pedigree analysis seem to fall in line with what genealogists working with the whole community have found. Additionally, it is clearly that he has deep roots within the historic Afrikaner nation, so assuming random mating and little population substructure, inferences from his pedigree may have some general utility.

Afrikaners apparently have some peculiarities genetically which has made them of some interest to scientists. It turns out that they seem to exhibit high frequencies of classical Mendelian diseases, a hallmark of inbreeding or population bottlenecks. This aligns well with the thesis that Afrikaners are the descendants of a small group of founders who arrived in the 17th century and entered into a long phase of demographic expansion, which culminated with their long Trek into the veld to escape English domination as well as perpetuate their practice of slavery (James Michner’s The Covenant is a fictionalization of this). As I have observed before the primacy of the “first settler” seems to loom large in the minds of demographers.

J. M. Greef, the author of the above paper, seems to refute this simple story in his own genealogy, though not the core aspect of the importance of the first founders. First the abstract:

It is often assumed that Afrikaners stem from a small number of Dutch immigrants. As a result they should be genetically homogeneous, show founder effects and be rather inbred. By disentangling my own South African pedigree, that is on average 12 generations deep, I try to quantify the genetic heritage of an Afrikaner. As much as 6% of my genes have been contributed by slaves from Africa, Madagascar and India, and a woman from China. This figure compares well to other genetic and genealogical estimates. Seventy three percent of my lineages coalesce into common founders, and I am related in excess of 10 times to 20 founder ancestors (30 times to Willem Schalk van der Merwe). Significant founder effects are thus possible. The overrepresentation of certain founder ancestors is in part explained by the fact that they had more children. This is remarkable given that they lived more than 300 years (or 12 generations) ago. DECONSTRUCT, a new program for pedigree analysis, identified 125 common ancestors in my pedigree. However, these common ancestors are so distant from myself, paths of between 16 and 25 steps in length, that my inbreeding coefficient is not unusually high (f approximately 0.0019).

Inbreeding coefficient is the probability that one’s two alleles are identical by descent. That is, they come from the same individual. For example, in the case of Elisabeth Fritzl her children have many genes where the alleles are identical by descent because half of her own genes are from her father, some many of his alleles will come back to reside within the same individual as part of a diploid pair. J. M. Greef notes that his inbreeding coefficient is about twice as high as is the norm for the typical European. Europe is a region of relatively low consanguinity, so this is a stringent reference. In some populations the inbreeding coefficient can be as high as 0.01. In short, he’s not too inbred.

That being said, the data within his pedigree do seem to show disproportionate contribution by some ancestors. This makes sense for two primary reasons. First, some component of reproductive variance is random (often modeled as a poisson distribution). Second, some component of reproductive variance is due to innate fitness (e.g., the Genghis Khan Y haplotype may be a case of this). Equality of contribution just isn’t in the cards.

Figure 2 shows the distribution of relationships within the pedigree:

Panel a illustrates that one individual is an ancestor of the author 30 times over! Many individuals are ancestors only once. Panel b shows relatedness, and again, some individuals are much closer to the author than others, with a skewed distribution. Panel c shows the number of generations between the ancestor and the author. The median number is well above ten generations, so the author has deep roots in South Africa. Finally, panel d shows the number of steps between his parents for any given ancestor. Because the author’s parents are both Afrikaners they share many common ancestors, but the steps between seem relatively large, and confirms that the author is not particularly inbred (if the parents were first cousins naturally there would be much shorter steps to common ancestors). It is clear disproportionate amount of J. M. Greef’s genes come from early settlers. This makes sense insofar as demographic expansion was likely front loaded, with later settlers having less of a chance to make an impact on an already large population.

The following table shows the contribution by various European and non-European groups to the author’s ancestry, as well as estimates for the total Afrikaner population in earlier studies on the right.


Note one point: only a minority of the ancestry of the author and Afrikaners are ethnically Dutch. This is important, because it shows how culture can spread and overwhelm ancestry. The Dutch imposed their language upon the French Huguenots, and their religion upon the Germans (who I presume were mostly Lutheran if they were from northern Germany, though a minority were Reformed or Catholic surely). Obviously the Reformed Calvinist religion and Afrikaans language both have a unique stamp in South Africa, but the connection of the Afrikaners to the Netherlands remained profound rather late in history. The Prime Minister of South Africa from 1958-1966 was born in the Netherlands. And yet another fact hard to deny is that the Huguenot French component seems to have persevered to a greater extent culturally than the German. The last Afrikaner President was named F. W. de Klerk, his surname being a form of Le Clerc. Another prominent South African head of state was Daniel Francois Malan. The author observes:

It is not clear if my higher estimate of French contribution is because of a systematic mistake in Heese’s (1970) estimate, or if it is because of a quirkiness in my own ancestry. It seemed to be the case that when a lineage hit the French Huguenots it stayed in this group. It will be interesting to compare the degree of inbreeding of the early generations of Huguenots to the other early immigrants. In the light of the calculations of Heyer et al. (2005) there is an interesting possibility that the cultural inheritance of fitness may have led to a systematic bias in Afrikaners, since Huguenots tended to be more educated and trained than German emigrants who tended to be soldiers. We are currently investigating this hypothesis.

There is a joke that the Baltic possessions of the Swedish monarchy were conquered with Finnish soldiers. Similarly, the Dutch overseas colonial possessions were staffed, especially at a lower level, by the rural male population surplus of northern Germany. A great many of these, likely the vast majority, never returned home and died abroad. These men contributed greatly to the census size of the Afrikaner population during much of its history, but it seems plausible that their fitness was far lower than the established Dutch and Huguenot groups because they lacked the resources and capital to flourish in a world which was much closer to the Malthusian edge than today. Many people don’t leave descendants, and it seems plausible that these Germans were fated not to do so to a far greater extent than the Dutch and Huguenots whom they were employed to protect and serve. Because of the genetic closeness of the north German and Dutch populations (in reality, Dutch are really simply another group of north Germans who transformed their regional identity into a national one for various reasons) I doubt that more thorough genetic testing will resolve this, rather, more pedigree analysis needs to be done on other individuals. But it’s an insight into the fact that social parameters have often been crucial to fitness in the human past.

As for the non-white component, the author’s results match those of previous researchers. He confirmed the likely probability of these results by the fact that his father carries mtDNA group M, which is most diverse in India. And in fact his father’s maternal lineage does trace back to a woman who was likely an Indian slave (slave women had particular surnames indicating their origin). My previous posts on the Coloureds highlighted the large Asiatic component to their ancestry, and it looks like previous researchers ignored this and focused on the Khoisan and Bantu. They also attempted to calculate ancestry based on classical markers which were found in African populations, and are present in low frequencies in Afrikaners, but that might ignore Asian signature markers (additionally, I assume that there was some natural selection for G6PD alleles). A survey of the total genomes of Afrikaners should be able to resolve the details of their ancestry, but it seems that the Afrikaners are far more colored than white Americans, by a factor of 5, but far less than white Latin Americans like Argentineans, probably by a factor of 5.

Finally, the author was also able to assess whether his ancestors exhibited a trade between quantity and quality in terms of their optimal number of offspring. In other words, did those who favored an extreme r or K selected strategy suffer vis-a-vis those who produced a more moderate number of offspring, not too low, and not too high? The author did not find any evidence of a tradeoff, and an optimal fitness. He was careful not to generalize too much, especially in light of the fact that Dutch colonial South Africa was an atypical society in many ways. I assume that living on the frontier means not having to say you’re sorry if you breed too much or too little.

Citation: Greeff, J. (2007). Deconstructing Jaco: Genetic Heritage of an Afrikaner Annals of Human Genetics, 71 (5), 674-688 DOI: 10.1111/j.1469-1809.2007.00363.x

April 19, 2010

Cape Coloureds: an instance of a generality

cape1Several months ago I put up a post which reviewed the geographical connections within the total genome content of the Cape Coloureds of South Africa. These peoples (plural because distinctive ethnic groups such as the Griqua were subsumed into this category in the 20th century) are of diverse origin, though generally their African and European ancestry has been highlighted. To the left I’ve reedited a plot which illustrates the inferred proportion of ancestry from various groups in modern Cape Coloured populations. Note that there is a substantial proportion of Asian ancestry, both South and East Asian. This makes historical sense as during the period of the founding of the Cape Colony a substantial number of Southeast and South Asian slaves were transferred from the Dutch East Indies, as well as from Madagascar, which itself has a Southeast Asian component in its population. Additionally, observe that the Bushmen & Khoikhoi element has been separated from the Bantu element. Archaeologists assume that the former are indigenous to South Africa, while the latter arrived within the last 2,000 years as the edge of the Bantu expansion which swept out of Nigeria east and south. These two populations are obviously both African, but their common ancestry is very deep. In some phylogenies Bushmen may be represented as the outgroup to all other human lineages, implying that one has to go very far back indeed for a common ancestor. In other words, the Bushmen are not the “oldest” human population, but have the oldest point of common ancestry with other human populations (e.g., the last common ancestor between a European and an East Asian may be ~30,000 years ago, but that between a Bushmen and a European may be ~80,000 years ago).

But these studies do not tell us everything about the demographic history behind the ethnogenesis of the Cape Coloureds. In this case uniparental lineages, mtDNA which traces the matriline and and nonrecombinant Y chromosomes (NRY) which trace the patriline may offer some value. Unfortunately too often because of methodological considerations we have looked at the uniparental lineages first, and then the total genome content, which I think inverts the optimal order in terms of putting genetic findings in context. A new study focuses on the Cape Coloured mtDNA and NRY lineages, with the previous findings in mind, Strong maternal Khoisan contribution to the South African coloured population: a case of gender-biased admixture:

The study of recently admixed populations provides unique tools for understanding recent population dynamics, socio-cultural factors associated with the founding of emerging populations, and the genetic basis of disease by means of admixture mapping. Historical records and recent autosomal data indicate that the South African Coloured population forms a unique highly admixed population, resulting from the encounter of different peoples from Africa, Europe, and Asia. However, little is known about the mode by which this admixed population was recently founded. Here we show, through detailed phylogeographic analyses of mitochondrial DNA and Y-chromosome variation in a large sample of South African Coloured individuals, that this population derives from at least five different parental populations (Khoisan, Bantus, Europeans, Indians, and Southeast Asians), who have differently contributed to the foundation of the South African Coloured. In addition, our analyses reveal extraordinarily unbalanced gender-specific contributions of the various population genetic components, the most striking being the massive maternal contribution of Khoisan peoples (more than 60%) and the almost negligible maternal contribution of Europeans with respect to their paternal counterparts. The overall picture of gender-biased admixture depicted in this study indicates that the modern South African Coloured population results mainly from the early encounter of European and African males with autochthonous Khoisan females of the Cape of Good Hope around 350 years ago.

The main results are in figure 2 & 3. The top left panel shows the mtDNA variation on an MDS chart in relation to other populations, “SAC” = South African Coloureds. The bottom left panel shows NRY variation. And the right panel shows the estimated admixture for mtDNA and NRY by population.


The results are rather clear, excepting the difference between the MDS and admixture estimates which seem to place less weight on the Bantu component in the second than the former. The authors chalk this up to difficulties distinguishing the Khoisan from the “pan-African” component. Contemporary Khoisan show substantial overlap with Bantu groups (just as some Bantu groups in South Africa such as the Xhosa show a great deal of Khoisan ancestry), so there are some ambiguities in assigning a haplogroup to one population or the other (the overlap seems a product of recent admixture).

But be as that may be, it is clear that a major dynamic in the founding of the Cape Coloureds had to be the pairing of Khoisan females with non-Khoisan males. The disjunction between European ancestry on the male and female lineages is stark, but should not be surprising in light of what we know from colonial history. And perhaps not just the colonial history of South Africa. The same pattern is evident in Latin America. Even societies which have transitioned from Mestizo to white, such as Argentina, seem to have done so through generations of male biased migration so that the indigenous mtDNA remains. And the same pattern can be found in some cases where we have no historical documentation because ethnogenesis occurred during the prehistorical period. In particular this seems the case in India, where male lineages show a strong West Eurasian bias, while female lineages do not (they are more closely related to East Eurasian lineages, though that connection is much more distant than Indian West Eurasia lineages have with other West Eurasian lineages).

A little over 10 years ago L. L. Cavalli-Sforza was coauthor on a paper titled Genetic evidence for a higher female migration rate in humans. The logic behind the results are simple, most human societies are patrilocal, so one presumes that gene flow would be mediated by the movement of women between local groups. Cavalli-Sforza found that female lineages seemed to be less localized than male lineages, implying greater gene flow. The literature since then seems rather muddled, and has not confirmed this original finding in a solid manner. I suspect that this is because one general dynamic can not capture the varied events which have characterized human genetic history. That is, there were periodic “shocks” to the basic patterns of worldwide genetic variation, but after those shocks passed then the dynamics which Cavalli-Sforza saw would come to the fore. Exploring the details of the balance between these varied forces is going to be where the future avenues of research lay. I predict that it is going to be in regions and populations which have gone through great cultural ferment since the last Ice Age that you will see this palimpsest whereby variation emerged as a synthesis of shocks interleaved between long periods of stasis and more conventional deme-to-deme gene flow. By contrast, isolated hunter-gatherer populations such as the Andaman Islanders may have missed out on the shocks, the period of “genetic revolutions” (though as I imply above, most hunter-gatherer populations show a great deal of admixture with the far more numerous agricultures who marginalize them and push up against their range, as is in the case among the Bushmen).

Finally, going back to South Africa one major issue is going to be the nature of the Afrikaners. Tentative earlier genetic and genealogical work suggests that ~5% of their ancestry is non-European, probably reflecting the movement of Cape Coloureds who could pass as white into the Afrikaner population (Cape Coloureds usually share language and religion with Afrikaners, so the cultural move would not have been insurmountable). Yet I have seen very few papers such as this, Deconstructing Jaco: genetic heritage of an Afrikaner. The author concludes that ~6% of his ancestry is from non-white slaves, in line with prior expectations. Though white Americans often take pride in their Native American ancestry (often genealogically attested, as with the descendants of Pocahontas) the total proportions are actually rather small, probably on the order of ~1% at most. In contrast the Afrikaners likely have more non-white ancestry because their founding population did not receive as much migration from Europe to dilute the original non-European element.

Addendum: The Cape Coloureds seem a real interesting population in light of admixture mapping, no?

Citation: Quintana-Murci, L., Harmant, C., Quach, H., Balanovsky, O., Zaporozhchenko, V., Bormans, C., van Helden, P., Hoal, E., & Behar, D. (2010). Strong Maternal Khoisan Contribution to the South African Coloured Population: A Case of Gender-Biased Admixture The American Journal of Human Genetics, 86 (4), 611-620 DOI: 10.1016/j.ajhg.2010.02.014

March 1, 2010

Ostrich shell art in South Africa 60,000 years ago

There’s a new paper in PNAS reviewing the tradition of etching on ostrich shells. Since it’s PNAS, the paper isn’t on the website, but Edward Edmund Yong is able to cover the major points thanks to his access. This stuff is of interest because there was a long time lag between the emergence of anatomically modern humans in Africa, and their subsequent expansion out of Africa ~50,000 years ago, the crystallization of “behavioral modernity.”


Powered by WordPress