June 11, 2013

~60-50 thousand coastal migration to Asia from Africa affirmed

My own opinions on this matter have been repeated ad nauseam in this blog, so I will only briefly touch on a couple of points in this new article.

Contrary to the authors' claim that: "The size of the mtDNA database is very substantial: currently there are almost 13,000 complete non-African mtDNA genomes available, not one of which is pre-L3." there are plenty of pre-L3 mtDNA in Eurasia. Some (or indeed most?) of these might represent more recent African admixture, and one could argue why they believe that to be the case, but no one has ever studied pre-L3 Eurasian mtDNA to conclude that none of it had an ancient presence in Eurasia. The native non-existence of pre-L3 in Eurasia is a viewpoint, not a fact.

Second, the authors present the following table:


Note, however, that they have chosen to date the common ancestor of all African sublineages (to ~70.2ky using ML method), but they have not done the same for the common ancestor of all non-African sublineages (i.e., M+N). A recent study estimates M+N to be 77KBP and L3 to be 78.3KBP, with other possibilities depending on method and portion of the molecule examined.

At present, I see no good reason to think that haplogroup L3 originated either in Africa or Eurasia; one could argue for an African origin, but temporal priority for African L3 is not established. The only thing that seems to be established is that there are "more" L3 subclades in Africa, which is phylogenetically meaningless until the bifurcating structure of the L3 subtree is resolved. And, we should also not forget that the Toba eruption did not cause a volcanic winter in Africa, nor was Africa affected by the drying up of the Sahara-Arabia belt c. 70kya, both of which may have suppressed Eurasian mtDNA variation within L3, irrespective of its ultimate origins. And, as I've argued before, both Toba and the post-70kya ecological crisis are excellent candidates for a Eurasian bottleneck caused by pre-existing populations surviving in refugia such as this.

Finally, the authors dismiss the possibility of an overestimated autosomal mutation rate and its implications for human history: "Recent reestimates of the autosomal mutation rate from whole-genome pedigree data suggest a European–Asian split time of 40–80 ka, although they do not, as has been suggested, lend any support to a dispersal fromAfrica before 80 ka (36) (Genetics)." 

Actually, the most recent estimate might be consistent with a ~96ky split of Africans from non-Africans, and the mutation rate issue is material to the timing of the African/non-African split. It's not clear where the needle will settle when the issue is resolved, but there's plenty of room for both pre- and post-Toba Out-of-Africa as things stand.


PNAS doi: 10.1073/pnas.1306043110

Genetic and archaeological perspectives on the initial modern human colonization of southern Asia

Paul Mellars et al.

It has been argued recently that the initial dispersal of anatomically modern humans from Africa to southern Asia occurred before the volcanic “supereruption” of the Mount Toba volcano (Sumatra) at ∼74,000 y before present (B.P.)—possibly as early as 120,000 y B.P. We show here that this “pre-Toba” dispersal model is in serious conflict with both the most recent genetic evidence from both Africa and Asia and the archaeological evidence from South Asian sites. We present an alternative model based on a combination of genetic analyses and recent archaeological evidence from South Asia and Africa. These data support a coastally oriented dispersal of modern humans from eastern Africa to southern Asia ∼60–50 thousand years ago (ka). This was associated with distinctively African microlithic and “backed-segment” technologies analogous to the African “Howiesons Poort” and related technologies, together with a range of distinctively “modern” cultural and symbolic features (highly shaped bone tools, personal ornaments, abstract artistic motifs, microblade technology, etc.), similar to those that accompanied the replacement of “archaic” Neanderthal by anatomically modern human populations in other regions of western Eurasia at a broadly similar date.

Link

12 comments:

Lank said...

There have been efforts to date the L(xL3) lineages in Eurasia, or at least Europe, with the earliest mtDNA lineages dating to 11,000 years ago, which is far more recent than the spread of L3. Furthermore, African L(xL3) lineages are lacking outside of West Eurasia, excluding South Asians with recent documented African contact.

Middle Easterners carry higher levels of L(xL3) than other non-African populations, as would be expected, but I'm not aware of any comprehensive attempt to date these lineages. It would be difficult to argue that these lineages in the Middle East predate the migration of L3 females, since the Middle East hasn't been isolated from the rest of the world since OOA times, nor do I see any reason to suspect such a thing. If there were any surviving L(xL3) lineages in Eurasia that predate the Toba eruption, they would not be expected to be found solely in modern Middle Easterners (particularly in contrast to the nearby Europeans, who trace a major portion of their recent ancestry to the Middle East), considering ancient Middle Easterners were ancestral to all non-Africans at some point.

Dienekes said...

There have been efforts to date the L(xL3) lineages in Eurasia, or at least Europe

Europe is irrelevant because it was colonized by modern humans after 50 thousand years ago.

Furthermore, African L(xL3) lineages are lacking outside of West Eurasia, excluding South Asians with recent documented African contact.

That is true but really doesn't help us because the question is whether ancient pre-L3 existed in modern humans outside of Africa, which, until the UP revolution excluded most of Eurasia.

If there were any surviving L(xL3) lineages in Eurasia that predate the Toba eruption, they would not be expected to be found solely in modern Middle Easterners (particularly in contrast to the nearby Europeans, who trace a major portion of their recent ancestry to the Middle East), considering ancient Middle Easterners were ancestral to all non-Africans at some point.

Why not? Those who suppose a recent OoA suggest that all non-L3 lineages were lost during the OoA bottleneck. One could very well use the same argument and say that all non-L3 lineages were lost during the Out-of-Arabia bottleneck.

Personally I'm not convinced that any pre-L3 is native to the Near East, but in any case the statement of Mellars et al. that pre-L3 is absent is wrong. And, the existence of pre-L3 is not really necessary since the origin of L3 itself is not resolved conclusively.

eurologist said...

I strongly suggest the authors of this paper study logic and argumentation, if they ever want to write something at least marginally convincing.

1. Straw-man argument: the much earlier ooA does *not* hinge on (perhaps) questionable pre/ post-Toba Indian finds. It is instead based on well-documented finds in the Levant and in Arabia, as well as extremely detailed paleoclimatic studies and also relatively early AMH finds in NE Asia.

2. Straw-man argument: The peri-Toba finds are Middle Paleolithic, ergo they were not made by AMHs. Widely accepted counter argument: Middle Paleolithic tool assemblages are the norm for AMHs before ~50,000 ya, and have been amply found and documented to be a poor distinguisher between ancient humans and AMHs. Even "low-hanging fruit" early AMH exploiters in Europe (early Aurignacian, Uluzzian) used relatively simple tools - simply, because that was enough, then.

3. Anachronisms-a-plenty. (i) The authors question AMH dates in India that significantly post-date European AMHs who, by both archeological and genetic evidence derived from S and SE Asian progenitors. (ii) Their cited dates for S Asian and NE African mtDNA lineages are highly selective and not at all generally accepted.

One of the worst "scientific" papers I have read, this week.

bmdriver said...

Out of India, once again.

Many people question the results of such papers because it directly conflicts with their own biased beliefs that civilisation arose from the middle east, aka Abrahamic religious ideology.

Grognard said...

The entirety of out of africa is a viewpoint, not a fact, and it hinges entirely on mt DNA evidence.

Yet there's a problem here, and that's neanderthal admixture.

That means the mt DNA evidence for OoA has been debunked. It doesn't matter we have not found neanderthal mtDNA or Y DNA markers if we have the other DNA. In fact it shows how hard it can be to make a conclusion from these markers, ESPECIALLY mt DNA which has so little variation.

Out of africa is a pretty new theory and has been touted as pure fact but since it has zero credible evidence it needs to go.

Neanderthal DNA also spread everywhere but the remotest parts of africa. So it's clear why we see these ancient mtDNA show up in africa, remote places like the congo are a refugium. It says nothing at all, zero, zip about where anything evolved.

Since all the human chimp ancestor candiates are in asia or europe then I'd say the best place to look is anywhere but africa.

It's also a lie that africa has the most genetic diversity. It has more difference between sub groups which are largely homogenous.

There is a supposed "mutation acceleration rate" for europe that happened due to the tecnological era. There's way, way more rare gene variations in europe than the rest of the world. Except suddenly accelerated mutation makes no sense.

When you have large populations that are not insular, you get mutation. When you have small isolated groups they are highly inbred. You shouldn't look for birthplace of man in a tribal group like our distant ancestors but in the oldest ruins of cities we can find.

Beastmanager said...

The coastal route in asia is supported by genetic and ethnographic data. See for example:
http://www.academia.edu/3084088/The_Out_of_Africa_Tribe_II_Paleolithic_warriors_with_big_canoes_and_protective_weapons

They were not just beachcombers, they most likely were already experienced seafarers with big canues.

Umi said...

Most of the L3 found in the Middle East belongs to Sudanese and Southeast African varieties brought over from the slave trade (started in the Middle Ages up to the Modern era).

The only rather old L lineages in the Middle East seem to be found in the Hijaz (SW KSA) and Yemen (rare L4 and L6 varieties, which aren't linked to recent contacts). However, no ancient and unique L3* varieties have been found as of yet.

A recent study by Hirbo et al. associates subclade L3a (oldest branch of L3) with Palaeolithic East Africans. Despite its age, it's practically not found in others parts of Africa nor the Middle East. Additionally, its two base subclades L3a1 and L3a2 split very early on (60-70 kya), so it can't simply be considered a bottle-necked fluke clade but genuinely points to an East African origin of L3 in general.

http://drum.lib.umd.edu/handle/1903/11443

terryt said...

"When you have large populations that are not insular, you get mutation. When you have small isolated groups they are highly inbred".

But it is in those inbred small, isolated groups that new mutations become fixed. If a mutation gives rise to a recessive gene selection cannot operate until the gene forms double recessives. This is far more likely to happen as a result of inbreeding rather than in a large population that is not insular. I agree that more mutations will arise in a large population but such mutations are unlikely to be able to form double recessives.

"You shouldn't look for birthplace of man in a tribal group like our distant ancestors but in the oldest ruins of cities we can find".

But humans had spread almost completely around the world long before anything like a city appeared.

Grognard said...

"But it is in those inbred small, isolated groups that new mutations become fixed."

WRONG!!!

Absolutely wrong, can't highlight the importance of this enough.

It implies that selection doesn't happen and that's not true at all. We know of genes under heavy selection where their transmission has nothing to do with drift. Evolution is NOT random happenstance and out of africa relies ENTIRELY on random happenstance.

It's saying all differences in populations is just a serious of founder effects. In which case genes would not be spreading wildly through massively different populations even when the originator population for the gene has neven even contacted them.

"But humans had spread almost completely around the world long before anything like a city appeared."

Humans spread around the world before they were anything we'd consider humans. Out of africa selects some tiny arbitrary population and says all of the humans on earth came from them. Since we know for a fact this isn't true, we can discount it.

The point about cities is that in places where mixing and evolution are happening you don't retain archaic features. And before the cities arose they already population centers, middle east and indus were paradisical at the time not wastelands like today.

Basically to believe out of africa is to disbelieve in evolution. Evolution has been redefined so that drift is "evolution" but that's just more damage that this kind of thinking has done.

terryt said...

"Absolutely wrong, can't highlight the importance of this enough. It implies that selection doesn't happen and that's not true at all. We know of genes under heavy selection where their transmission has nothing to do with drift".

Surely you must realise that selection can only operate on a gene that is expressed. A mutation in a large population is unlikely to be expressed.

"Evolution has been redefined so that drift is 'evolution'"

Although genetic drift does occurr the term is often used when the evidence doesn't fit a prefered scenario. However, as I said above, a gene must form double recessives before selection is able to act. That is far more likely to occur in a small population with more chance of inbreeding than in a large population, in a city or otherwise. Once selection acts on double recessive version the phenotype it forms can then spread through the wider population. To me that seems to be the way evolution works.

"It's saying all differences in populations is just a serious of founder effects".

I presume you meant 'series'. And I agree this stance too is ridiculous. Founder effects obviously do happen but too often are used to explain evidence that doesn't fit some prefered scenario.

"Humans spread around the world before they were anything we'd consider humans".

I'm more than a little inclined to agree with you there. I have been a 'regional continuity' supporter for a number of years.

"Out of africa selects some tiny arbitrary population and says all of the humans on earth came from them. Since we know for a fact this isn't true, we can discount it".

Both the mt-DNA and the Y-DNA indicate both derive from Africa. But some-one commented in one of Dienekes' posts that we cannot forget that the 'out of Africa' people usually did not enter uninhabited land. Haplogroups give very little indication of 'race'. However they can indicate direction of general genetic movement.

terryt said...

Sorry. Me again. I thought I should support my statement about recessive genes. It is just as well most mutations that have any effect are recessive. Here is some information about defects in cattle for which ancestral information is available. You will understand why it becomes so difficult to breed only for production. And also why we need the presence of double recessive genes to enable the selection required to allow evolution.

http://animalscience.ag.utk.edu/beef/pdf/CurleyCalfSyndromeForBullBuyers.pdf

Quote:

"It has been determined that the mode of inheritance for this condition is a simple recessive gene and can only be expressed when both genes are present. Its pattern of inheritance is similar to coat color for black and red in cattle or horned or polled".

http://pubs.ext.vt.edu/news/livestock/2009/07/APS_07-10-09_06.html

Quote:

"Neuropathic Hydrocephaly (NH) is a genetic defect recently recognized in Angus cattle. NH-affected calves are dead at birth and have severe hydrocephalus (water on brain), skull malformation, absence of central nervous system tissue (brain and spinal cord), and joint fixation. Research has confirmed that NH is a lethal genetic defect which is inherited as a simple recessive and controlled by a single gene".

http://angusnz.com/nh_main.php

Quote:

"Estimates were made of the diagnostic sensitivity and specificity of a new DNA test developed at the University of Illinois by Prof. Jonathan Beever and his colleagues to identify carriers of a simple autosomal recessive mutation causing congenital contractural arachnodactyly (CA) in Angus and Angus-infused beef cattle, a heritable developmental disease formerly called 'fawn calf syndrome'".

http://www.dpi.nsw.gov.au/agriculture/vetmanual/specimens-by-disease-syndrome/diseases_of_livestock/cwh

"CWH is a congenital genetic disease with an autosomal recessive disease, presumed to involve defective structural protein(s) within the desmosomal complex".

fla88 said...

Wow.. I'm 100% European from Italy (according to prof McDonald) and my mtDNA is L3f :)