New blogs

Leherensuge was replaced in October 2010 by two new blogs: For what they were... we are and For what we are... they will be. Check them out.

Sunday, May 9, 2010

Exploring the Neanderthal admixture episode (1)


I
mentioned the other day that a small amount of Neanderthal genes have been detected among non-African modern humans in a homogeneous way that strongly suggests it happened only within the process of migration into Eurasia and beyond.

The percentage of Neanderthal "blood" in non-Africans is estimated in 1-4% (average: 2.5%) what means the equivalent to one great-great-great-grandparent (3.13%) or great-great-great-great-grandparent (1.56%). In other words: the equivalent to one of 32-64 ancestors.

The homogeneity of the apportion across people of Eurasian ancestry (includes Oceanian and American natives) tells us that this was the situation in the Migrant Population at the arrival to South Asia, where it expanded very dynamically.

We must understand that, due to various environmental reasons (adaptation to warm climate, arid barriers and Neanderthal pressure) the Migrant Population must have been rather small between Africa and South Asia. At least that's what the genetic signature strongly suggests. However it was not too small or not too homogeneous because several lineages (2-3 at least per gender) arrived to the South Asian secondary cradle (a really small homogeneous population would surely have got its lineages more strictly fixated, i.e. one per gender).

There are many different detailed scenarios that could explain this limited gene flow but the following one is which seems to me as the simplest one on light of the available genetic and archaeological data:



Feel free to discuss, of course, it's very tentative and pretty much open to variants and alternatives.

Why did I get the main migrant population go via South Arabia? Because there are many clues that suggest so but it's again a personal choice: my best hunch.

Why did I place the main admixture event in Palestine? Because there's some archaeological evidence telling of at least contact between both species there c. 130,000 years ago (old stratigraphic date). But could have happened in many other areas of the Middle East, in Iran for instance (in which case, we could consider a single migrating population via South Arabia and consider the Palestinian episode genetically unproductive).

Just trying to figure out...

4 comments:

terryt said...

"In other words: the equivalent to one of 32-64 ancestors".

Valid, and interesting, comment. That's quite a lot really. Many breed societies do not allow the introduction of any new genetic material at all after the studbook is established. Some of these studbooks have now been established more than two hundred years, and the breeds are now nowhere near as useful productively as they were when first formed, they have become too inbred. But some breed societies do allow 'breeding up'. That is taking one breed and, after several generations of crossings with another breed, allowing the offspring to be considered purebred members of the second breed. The most common level allowed for breeding up is four generations: 15/16, just one generation less than the 31/32, and two generations less than the 63/64, of apparent proportion of modern/Neanderthal genes in non-African modern humans.

Andrew Oh-Willeke said...

It is worth noting that the oldest Near Eastern modern human/Neanderthal cohabitation was followed by a modern human absence from the Near East ca. 75,000 years ago, only to be re-established ca. 50,000 years ago (at any rate post-Southern Route).

We don't know if the humans in the first wave died out or retreated. If they retreated after gaining Neanderthal genes then, it might not show up anyway given the methodology of the study (which uses Africans as a baseline).

Maju said...

Let's see, Andrew:

What we know of the OoA population(s) is that it is well described by mtDNA L3 and Y-DNA CDEF or Y(xA,B).

This population did not only expand to Eurasia but did it in Africa too. Very extensively even if we look only at L3. Certainly they reached West Africa along with L2 lineages looong ago. If the pre-OoA L3 population was already 1-4% Neanderthal, Yorubas would have meaningful, detectable levels of that (maybe 0.5-2%).

They do not.

Using Africans (diverse ones) as baseline is just perfect to test this hypothesis: we know that both San and Yoruba have negligible recent (historical or even Neolithic) Eurasian admixture, we know that for all practical effects we diverged from them at this very OoA episode: at the mtDNA L3 node. But we also know that Bushmen diverged earlier than that.

So, unless some new data comes to tell us otherwise, the theory posited by the paper of admixture at the OoA migration, after the scatter of the L3 proto-OoA population, is the only consistent one.

terryt said...

"It is worth noting that the oldest Near Eastern modern human/Neanderthal cohabitation was followed by a modern human absence from the Near East ca. 75,000 years ago, only to be re-established ca. 50,000 years ago (at any rate post-Southern Route). We don't know if the humans in the first wave died out or retreated".

Bit of both probably. But if modern humans at more than 70,000 years ago could make it to the Levant surely it's reasonable to assume they didn't just stop there. If they retreated we can't assume they just retreated back into Africa.