Substack cometh, and lo it is good. (Pricing)

African genomics tells us about deep structure and history


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.

The results confirm at a finer-grain and with more precision what we’ve known before. The strangest result (to me), which has been confirmed over the past several decades or so (starting with uniparental lineages), is that the hunter-gatherers of Sub-Saharan Africa are broadly related more closely related to each other than they are to the agriculturalists of Sub-Saharan Africa. The relationship is deep (see the figure of the NJ-tree). The Baka and Mbuti Pygmies of the west and east of the Congo are deeply diverged themselves from each other, and even more so from the Khoisan people of southern Africa, or the Hadza of Tanzania.

But the missing piece of the puzzle has to be the “Bantu expansion.” As was pointed out to me by Nick Patterson about a decade ago, the Bantu-speaking people are genetically very closely related, as benefits a demographic expansion that began on the order of ~3,000 years ago from a small region of the Cameroon-Nigeria border. Curiously, a similar expansion did not occur to the west, as West Africa remains linguistically much more diverse, though genetically it is broadly similar to “Bantu Africa.”

Even on the furthest southern edge of the Bantu expansion, the majority of the ancestry remains similar to their linguistic relatives in Cameroon-Nigeria region, even if a substantial minority of the ancestry is Khoisan. Similarly, the Bantu speakers of East Africa often have some Eurasian ancestry, probably mediated by pastoralist Cushitic-speaking people who were already present.

The genetic variation of contemporary Africa is an artifact of the recent Holocene. This was clear after the publication of Skoglund et al. a few years ago. It also means that a lot of the fine-grained inferences from modern populations must be taken with a grain of salt, as there was probably more admixture and movement in the early Holocene which is not visible to us, let alone in the Pleistocene.

That being said, the figure at the top of the post captures the major features since 1) they’re in keeping with our understanding of the interaction of deeply diverged lineages 2) there has been other work with other methods detecting the same dynamics.

Most scholars did not accept the likelihood of Neanderthal admixture until the genome of this ancient hominin was published. Then we discovered the Denisovan contribution due to another genome. Both of these discoveries should update our prior expectations of any given model. The model above utilizes a complex ABC-framework. If you know what that means, you know I’m not going to get into the details. But I trust computers only to a point… I expect the ABC model does reflect realities of archaic admixture, but I’m not sure that the typology of the tree is totally settled.

That’s because of the reality that human genealogies are more like graphs with edges of different thicknesses. There’s always some gene flow, especially between close lineages. These models often infer that the Khoisan diverged ~200,000 years ago from other lineages. And, that the African ancestors of modern Eurasians diverged ~100,000 years ago. The model above even suggests that there is “anatomically modern human” (so African) admixture into Neanderthals. It is highly likely that this was detected in the Altai genome, so I think this is likely.

But is it likely that the Khoisan were a long and isolated branch over 198,500 years ago when some admixture with a Nilotic or Cushitic populace broad Eurasian ancestry? I doubt it. I suspect these divergence/coalescence times mask deeper and more recent branches, which fused together, and so give us an average mid-point coalescence. Archaics aside, I suspect that some of the Khoisan ancestry is probably deeply basal, and some of it is less basal.

Finally, the divergence time of the ancestors of Eurasians is  ~100,000 years ago. Archaeology tells us that the expansion of the ancestors of modern non-Africans occurred 50-60,000 years ago. What was going on in that interval? First, of course, all these estimates need to be taken with a grain of salt, even beyond the confidence intervals. But, I’ve read enough of these papers to see a number >>>60,000 years for the divergence to think it’s real. It strikes me that the ancestral proto-Eurasians were on the fringe of Africa or in Arabia for a long time before the expansion. A deep divergence between Eurasians and Sub-Saharan Africans is probably a function of the fact that most ancient African populations left no descendants (or very few). This seems common in large parts of Eurasia. If the proto-Eurasians were liminal, it is possible all the intervening populations between it and the ancestors of West Africans just went extinct.

The structure and history of African genetics is not just important for medical and population work. It is probably also essential to scaffold some of our inferences about cultural evolutionary processes. At a minimum, half of modern human man history has been within Africa (if modern humans were outside Africa in Arabia 100,000 years ago).

3 thoughts on “African genomics tells us about deep structure and history

  1. Razib, you’re podcast with Jeffrey Rose a few months ago serves as a perfect primer for this discussion. I don’t think there’s any need for an Out of Africa event at all 50,000-60,000 years ago. Human beings were well established in Southwest Asia 100,000+ years ago, at a minimum in the Levant and Arabia. I have no doubt that if Iraq and Iran were better surveyed more evidence of human habitation would be found there as well.

  2. Fascinating…would you say Southern Nigerians like the Yoruba are semi-Bantoid? I know linguistically they are nothing alike but genetically speaking.

Comments are closed.