March 13, 2012

Neolithic expansions: how the European foragers were assimilated

From the paper:
Recently, mitochondrial DNA (mtDNA) from skeletal remains of European early farmers and late hunter-gatherers has been retrieved [7]–[13]. The frequency of mtDNA haplogroups, defined by substitutions shared by related mtDNA types (Phylotree.org-mtDNA tree build 12), in early farmers across Europe [7], [10]–[13] was found to be overall similar to those in modern Europeans (Figure 1, Figure S4, Figure S5), while pre-Neolithic hunter-gatherers appear to be quite distinct (Figure 1). In particular, 83% (19 out of 23) of hunter-gatherers analyzed to date carry mtDNAs belonging to haplogroup U [9], [10], [14] and none of the hunter-gatherers fall in haplogroup H. In contrast, haplogroup U has been found in only 13 of 105 (around 12%) individuals from early farming cultures of Europe and it occurs in less than 21% of modern Europeans, while haplogroup H comprises between 25% and 37% of mtDNAs retrieved from early farming cultures (Figure S4) and is in about 30% of contemporary Europeans (Figure 1). The mtDNA data thus suggest that the pre-Neolithic populations in Europe were largely replaced by in-coming Neolithic farming groups, with a maximum mtDNA contribution of around 20% from pre-Neolithic hunter-gatherers [8]–[10].
Age estimated by looking at nucleotide difference between genomes descended from a common ancestor. See left: haplogroup H is in green and shows low nucleotide differences; U is in brown, and it is clear that the former are much more recently related to each other than the latter. This is consistent with recent founders for H and older founders for U.

The following paragraph explains exactly what the mtDNA evidence says happened:
The high frequency of H-type mtDNAs in European Neolithic populations and its complete absence in pre-Neolithic hunter-gatherers suggests that H-type mtDNAs arrived with early farmers in Europe. The population size increase observed between 9,000 and 5,000 YBP likely represents the population expansion that accompanied the Neolithic revolution. In contrast, U-type mtDNAs show an increase in population size around 15,000 to 10,000 YBP, which coincides with the end of the last glacial maximum in Europe and a northwards expansion of hunter-gatherer populations. The data suggests that this population remained rather constant after 10,000 YBP until the onset of the Neolithic revolution. However, the H-type mtDNA population size seems to experience an exponential increase around 7,000 YBP, suggesting that both populations are not yet fused. After 4,000 YBP, no archaeological remains of hunter-gatherers were found in central Europe [22]. From approximately that time on, both H- and U-type mtDNAs expand in a similar way. This may reflect fusion of the two populations where these mtDNAs were prevalent.
So, in short: U experienced after the end of the last Ice Age, as European hunters found new room to expand in space, and economic opportunities to expand in population size. By 10,000 yBP they had maximized their population size. Subsequently, incoming Neolithic H-bearers started expanding as they arrived in Europe and started growing in numbers and filling up the landscape. At some point, (c. 4ky BP) all the hunter-gatherers had been assimilated, and the composite gene pool consisted of the descendants of mainly the Neolithic farmers and a minority of assimilated hunter-gatherers. Since H and U were no longer tied to different subsistence strategies, they underwent expansion in similar fashion.

PLoS ONE 7(3): e32473. doi:10.1371/journal.pone.0032473

Complete Mitochondrial Genomes Reveal Neolithic Expansion into Europe

Qiaomei Fu, Pavao Rudan, Svante Pääbo, Johannes Krause

The Neolithic transition from hunting and gathering to farming and cattle breeding marks one of the most drastic cultural changes in European prehistory. Short stretches of ancient mitochondrial DNA (mtDNA) from skeletons of pre-Neolithic hunter-gatherers as well as early Neolithic farmers support the demic diffusion model where a migration of early farmers from the Near East and a replacement of pre-Neolithic hunter-gatherers are largely responsible for cultural innovation and changes in subsistence strategies during the Neolithic revolution in Europe. In order to test if a signal of population expansion is still present in modern European mitochondrial DNA, we analyzed a comprehensive dataset of 1,151 complete mtDNAs from present-day Europeans. Relying upon ancient DNA data from previous investigations, we identified mtDNA haplogroups that are typical for early farmers and hunter-gatherers, namely H and U respectively. Bayesian skyline coalescence estimates were then used on subsets of complete mtDNAs from modern populations to look for signals of past population expansions. Our analyses revealed a population expansion between 15,000 and 10,000 years before present (YBP) in mtDNAs typical for hunters and gatherers, with a decline between 10,000 and 5,000 YBP. These corresponded to an analogous population increase approximately 9,000 YBP for mtDNAs typical of early farmers. The observed changes over time suggest that the spread of agriculture in Europe involved the expansion of farming populations into Europe followed by the eventual assimilation of resident hunter-gatherers. Our data show that contemporary mtDNA datasets can be used to study ancient population history if only limited ancient genetic data is available.

Link

56 comments:

apostateimpressions said...

After 4,000 YBP, no archaeological remains of hunter-gatherers were found in central Europe [22]. From approximately that time on, both H- and U-type mtDNAs expand in a similar way. This may reflect fusion of the two populations where these mtDNAs were prevalent.

Is it possible that Indo-European invaders melded the Mesolithic and Neolithic groups when they subjugated the whole lot of them during the Bronze Age and suppressed the hunter-gathering lifestyle? An elite widespread IE cultural and political dominance seems consistent with both the consistent replacement of pre-IE languages and with the sudden melding of the Meso and Neo populations and the sudden disappearance of Meso culture.

An analogy might be where African tribes that had previous remained distinct were suddenly melded into a single group in the US by the slave trade, when African language and culture suddenly gave way to European culture. The IE invaders may have also have formed a "master race" of sorts and with similar demographic and cultural effect. (We see historical cultural leaps in both cases.)

Dienekes said...

No need to invoke a special explanation. Hunters simply can't compete with farmers; we've seen that everywhere in the world. Their only remnants are in those places where farmers got late (like South Africa) or where the climate is so hostile to raising crops (such as Siberia).

Also, hunting/foraging was always a secondary strategy for farmers everywhere in the world. The farmers got more and more land and competed with hunters for game and other resources in the remaining wildlands. ALso, because of their village life they were organized in bigger units that could easily outcompete the scattered hunter bands that can't organize in larger units because local resources don't suffice. So, foragers slowly adopted farming or intermarried with farmers and became extinct as a distinct culture.

Maju said...

If U5 is mostly pre-Neolithic (this is essentially your theory by someone else), shouldn't there be a 20-25% of pre-Neolithic autosomal ancestry? At least 10%! But nope: the pre-Neolithic component has simply vanished altogether, according to you. Not serious!

"Hunters simply can't compete with farmers; we've seen that everywhere in the world".

There are a lot of places where hunter-gatherers have unquestionably become farmers or herders, for example the Khoikhoi, the Papuans, most pre-Columbian Native American farmers. Your claim would make some sense if you said "people with Iron Age tech" (or even Industrial tech) instead of "farmers" (early slash-and-burn farmers).

Anyhow this paper fails miserably because it has not collected any mtDNA samples from Paleolithic SW Europe and the only data collected so far (Chandler 2005, Portugal, which you arbitrarily prefer to ignore, does not say this).

apostateimpressions said...

Our analyses revealed a population expansion between 15,000 and 10,000 years before present (YBP) in mtDNAs typical for hunters and gatherers, with a decline between 10,000 and 5,000 YBP. [...]

After 4,000 YBP, no archaeological remains of hunter-gatherers were found in central Europe [22]. From approximately that time on, both H- and U-type mtDNAs expand in a similar way.


I took that to mean that the Meso population declined after the arrival of Neos, that the Neos did not gradually assimilate the Meos and that the two suddenly melded circa 4K bp. But I see that there is a thousand years between the times indicated for the decline of Neos (10-5000Kbp) and the melding (4000Kbp); perhaps they gradually melded during that 1000 years. I dont know if it is possible to be precise about whether the melding was gradual or sudden.

Dienekes said...

If U5 is mostly pre-Neolithic (this is essentially your theory by someone else), shouldn't there be a 20-25% of pre-Neolithic autosomal ancestry? At least 10%! But nope: the pre-Neolithic component has simply vanished altogether, according to you. Not serious!

Who says it's vanished altogether? Please don't imagine things. No one has ever claimed that it has vanished altogether, and I'm perfetly happy to consider 10-20% or thereabouts Paleolithic ancestry with some regional variation.

Anyhow this paper fails miserably because it has not collected any mtDNA samples from Paleolithic SW Europe and the only data collected so far (Chandler 2005, Portugal, which you arbitrarily prefer to ignore, does not say this).

I don't arbitrarily choose to ignore the unpublished Iberian conference paper, and, apparently I'm in good company in ignoring it, unless you think that you know more about ancient DNA than Svante Paabo &co.

Unknown said...

A cyclical argument. Populations are labelled hunter-gather or farming partly on their haplogroups. Therefore U is hunter-gather and H is farming. Sigh.

H is indisputable present in large numbers in North Africa before the neolithic arrived, it is reasonable to assume it was in Southern Europe (Iberia) too. The disputed Portuguese paper confirms it.

This paper is consistent with a South to North expansion (which I have long advocated). Similar to the R1b expansion out of Iberia.

If you look at the papers dating (Figure S6) the unbiased expansion of H was already accelerating away by 13,000 BC. This pre-dates even the neolithic in the East and neatly first with the Late Glacial maximum when refuge populations expanded. The H population had actually stabilized by 6-7kya when the neolithic is supposed to have arrived in Europe, and this is European only data! If H expanded dramatically when the neolithic arrived in Europe then the curve should still be accelerating at this point in time. Only U seems to expand when the neolithic hits Europe. Only one H data set shows a possible small expansion around 2kya, the biased set. I could not work out how they biased it (the unbiased sample had fewer samples)?

This is not IMO population replacement from the East arriving with the neolithic, it is an expansion from the existing southern European population.

Check out the very different haplogroup profiles of Eastern neolithic populations (Tell Hamad, Tell Kurdu, Tell Halulu).

http://www.buildinghistory.org/distantpast/ancientdna.shtml

Dienekes said...

A cyclical argument. Populations are labelled hunter-gather or farming partly on their haplogroups. Therefore U is hunter-gather and H is farming. Sigh.

Don't bother posting comments here if you don't actually read the papers or have anything interesting to say.

There is nothing "cyclical" about measuring nucleotide divergences between full H and U mtDNA genomes and discovering that the former expanded in Neolithic, the latter in pre-Neolithi, and since 4,000 years BP, they expanded together.

If you look at the papers dating (Figure S6) the unbiased expansion of H was already accelerating away by 13,000 BC.

That is in ky BP and neatly coincides with the onset of the PPN in the Near East.

jeanlohizun said...

I don't arbitrarily choose to ignore the unpublished Iberian conference paper, and, apparently I'm in good company in ignoring it, unless you think that you know more about ancient DNA than Svante Paabo &co.


Appeal to authority, nicely played Dienekes.

Anyhow, this is a very interesting paper indeed. Yet there are a few things that might be in good need for further explanation:
Table-1 showing the regions where the Haplogroup samples came from show that for the Basques only 8 mt-DNA were fully sequenced, all of them belonging to haplogroup H, they in turn reference it to the work of Alvarez-Iglesias et al(2009).

Here is an excerpt from the Alvarez-Iglesias et al(2009) study:
“Finally, eight samples carrying substitution C4592T in the sample set of 75 individuals from the Basque Country (bordering East Cantabria) screened in [19], and presumably belonging to a new minor clade of haplogroup H (here baptized as H2a5), were selected for complete genome sequencing.”

So turns out all the Basques Haplogroup H studied belonged to a sole clade H2a5. Out of the 332 clades of mtDNA Haplogroup H, only 33 came from Iberia, none from France, 31 from Finland, 119 from Italy, and 114 from the US/UK Caucasoids(Listed as unknown, yet sourced to the following study: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC447592/?tool=pubmed)

So there is nowhere near a fair representation of a descent gene pool for H. In fact only 9 countries were used to draw the sample of mt-DNA Haplogroup H, with an over-representation of Italy and the US/UK. On the other hand mt-DNA Hg U was sampled from over different 21 populations, with the greatest sample drawn from Italy at 74, for a total of 222 samples.

Also the so called unbiased sample were just 144 sequences of mt-DNA Hg H drawn from the UK/US sample, and 41 sequences of mt-DNA Hg U, which can be readily seen in table-2.

Not to mention the fact that they use the samples from Bramanti et al(2009) as representative of the whole pre-Neolithic European mt-DNA pool. But all in all, it’s a good study, gives some good insight as to the time of expansion of mt-DNA H in Croatia compared to mt-DNA U.

royking said...

Wow, what a heated dispute! I think the mesolithic mtDNA of SW Europe is an open and unpublished question. Also, I disagree with Dienekes about the inevitable march of the Neolithic demography. Here in California and the NW Coastal US, hunter/foragers/fishers resisted the conversion to a Neolithic corn economy from Mesoamerica until the post-Columbus Era. California, clearly could have supported corn; but the surplus of fish and acorns prevented the utilization of a Neolithic economy and supported a high population density of Mesolithic populations--in contrast to New Mexico and Arizona which did convert to corn based agriculture. I think a similar process may have occurred in SW Europe--Portugal/Spain/Midi of France.

Unknown said...

"Be polite. Use facts and arguments"

"Don't bother posting comments here if you don't actually read the papers or have anything interesting to say."

I have read all the papers as you well know as I posted comments on most of them on your blogspot. Are you only interested in the opinions of those who agree with you?

"There is nothing "cyclical" about measuring nucleotide divergences between full H and U mtDNA genomes "

We knew this already, U is older than H, I dont know anyone who disputes this.

"That is in ky BP and neatly coincides with the onset of the PPN in the Near East."

Its a bit early IMO but leaving that aside. There is no expansion of H corresponding to the neolithic in Europe. If H swamped nearly vacant Europe from the east with massive population replacement then the line should still be going up even more dramatically well past 5k.

It isnt.

GailT said...

We currently have 485 FGS test results in the U5 mtDNA project, and only a small fraction have been submitted to GenBank. The project test results are consistent both with an age estimate of nearly 30,000 years for U5b and a rapid expansion of U5b daughter groups around 20,000 years ago. We have a period of slow growth in diversity beginning with U5 which has only 2 daughter group, U5a and U5b, and only 5 groups at the next level, U5a1, U5a2, U5b1, U5b2 and U5b3. But beginning around 20,000 years the population appears to expand more rapidly, for example, U5b1 has 9 named daughter groups and it's very likely we will find more. Some of these old U5b1 result are also extremely rare with only 1 or 2 examples of mtDNA clades that date to around 20,000 years ago.

So, the U5 project results are consistent with the analysis in the new Fu et al paper: a post-glacial expansion of U5 followed by a period of slow growth or contraction as U5 was largely replaced by Neolithic immigrants, and I expect a larger sample size will confirm their analysis.

It will also be interesting to break down the analysis between U5a and U5b. U5a has less diversity and U5a1a1 especially is found in the near east as well as Europe. It's possible some U5a were among early farmers and expanded into Europe along with H.

It would be extremely helpful to see FGS testing of ancient remains, and of course testing of more ancient remains. I don't think we can rule out the possibility that some H were present in Europe ca 10,000-15,000 ybp.
But the theory of H as the primary, original hunter gatherers in Europe was based on nothing more than the current distribution of H in Europe. At this point it is really an act of faith to cling to that interpretation in the absence of any compelling evidence.

Grey said...

I admit a bias because i like the atlantic cowboy idea but..

alternatively
- both paternal and maternal neolithic dna in a cline from the SE and S
- both paternal and maternal paleo (turned paleo-cowboy) dna mostly around the western and northern periphery
- paleo-cowboy paternal dna and maternal neolithic dna in the middle

eurologist said...

I also think that due to the lack of data, at this point we can't exclude H having significant presence in the very south of Europe, such that it would not have been able to participate in the post-LGM expansion. Also, if U were the last to retreat, perhaps they were also poised to be the first to make a post-LGM comeback.

If you look at S7, U has been a very successful haplogroup for a very long time. The paper is very nice, but clearly at the limit of what can be done, today. The early neolithic mtDNA data show that farmers "picked up" or integrated women from all different local backgrounds. Given the diversity of paleolithic and neolithic mtDNA, concentrating on just two haplogroups - which perhaps cannot be that clearly labelled - gives us only limited information.

eurologist said...

So, to play the devil's advocate, let's assume that 1/3 of today's H is paleolithic southern European, 1/3 is Balkan, and 1/3 is intrusive neolithic near Eastern. This scenario is clearly still compatible with the effective population size curves of this paper. Also, perhaps U for most of the time (except just around LGM) was primarily a northern haplogroup. Then it is still possible that more than 50% of today's mtDNA lineages are paleolithic European...

Also, I would like to see how today's European H-tree (and K, T, J, for that matter) map compared to the near Eastern tree - there is a lot of information available here that has not been exploited, yet.

Dienekes said...

Also, perhaps U for most of the time (except just around LGM) was primarily a northern haplogroup.

There are region specific U subclades in North Africa, West/South Asia. There is no reason to think that U was ever a primarily northern haplogroup. It was a widely distributed haplogroup of Paleolithic people in West Eurasia, which happened to persist the longest in high frequency in Europe.

Maju said...

"No one has ever claimed that it has vanished altogether, and I'm perfetly happy to consider 10-20% or thereabouts Paleolithic ancestry with some regional variation".

Then why doesn't it show up in any sort of admixture runs? I'm yet to see in any such analysis, either by you, me, any other blogger or in academic papers: that component, which should show up very clearly and distinctly. It does not: nowhere!

As the remnant "Paleolithic component" necessary in the Neolithic replacement theory, does not exist, then this seems to debunk or at least show yet another fundamental contradiction in the Neolithic replacement model.

I think that the Neolithic replacement model needs to find that Paleolithic remnant component to prove itself true. Otherwise I understand that the Paleolithic continuity (with lesser Neolithic input) fits much better with the data, which clearly shows a Europe-West Asia duality at K=2 (or equivalent if more samples present), with some West Asian (roughly Neolithic) penetration specially in the Balcans and Italy.

"I don't arbitrarily choose to ignore the unpublished Iberian conference paper, and, apparently I'm in good company in ignoring it, unless you think that you know more about ancient DNA than Svante Paabo &co".

I know a bit about the literature and the sampled areas and the limits and reaches of our knowledge of aDNA in Europe (I've taken my time to consider it in detail, as you may know) and you are wrong in trying to dismiss my argument with an appeal to authority (which is always a weak defense and a sign of lack of a good rationale).

Experts can well be wrong. In fact they often are. That's why criticism and self-criticism are so important in science (which is very different from scholasticism, in which appealing to authority is the only rationale - faux rationale).

In this case, as others of similar nature, the fact that the sample is essentially limited to Northernmost Europe, with very limited integration of what was then and is still today most of the continent. We can't describe what happened on Europe based on a few hunter-gatherers (and "Forest Neolithic" peoples, not the same as Paleolithic) from the Baltic.

We need more data. And there is some of that, just that it is conveniently ignored on scholasticist-ideological reasons. Some people seem to think that if we hide the evidence down the rug (not by replacing or expanding it with better evidence, what might be a feat, but just ignoring it altogether) we can alter facts.

Sorry to inform you that facts are not reliant on what you or I may think.

eurologist said...

Well, of course I am talking about the U subgroups of relevance in this discussion - the ones that are present both in paleolithic and modern Europe.

What I am driving at is this: if those were mostly northern pre-neolithic, but have a substantial contribution today throughout Europe, then, again, not much paleolithic was lost (they gained in the south, H gained in the north).

If H was present in the South (we don't know at this point) with later dispersal, and then if U spread as well, but south, there is not much left of the neolithic replacement theory. Could instead be all intra-European mixing since post-LGM.

We desperately need more data from southern Europe.

Dienekes said...

Then why doesn't it show up in any sort of admixture runs? I'm yet to see in any such analysis, either by you, me, any other blogger or in academic papers: that component, which should show up very clearly and distinctly. It does not: nowhere!

For the same reason that ASI doesn't show up in any ADMIXTURE runs, because it has been completely absorbed and no longer exists in a distinct form in any population.

Well, of course I am talking about the U subgroups of relevance in this discussion - the ones that are present both in paleolithic and modern Europe.

The point is that if these subgroups are in Northern European and other subgroups are in North Africa and West Asia, it's very unlikely that Southern Europe was a "hole" without U of its own. So, your idea that U was primarily a northern haplogroup within Europe is probably wrong.

Dienekes said...

I know a bit about the literature and the sampled areas and the limits and reaches of our knowledge of aDNA in Europe (I've taken my time to consider it in detail, as you may know) and you are wrong in trying to dismiss my argument with an appeal to authority (which is always a weak defense and a sign of lack of a good rationale).

You don't have an argument, just a lot of opinions based on bad data.

Also, there's nothing wrong in trusting the opinions of expert. If I get the flu I'll go to the doctor not to the witch doctor.

If an expert says that 3+1 = 2, the fact that he's an expert will not make me believe it. But, I see absolutely no reason to think that the dozens of papers published since 2005 have a conspiracy to suppress the supposed ancient DNA results from pre-Neolithic Iberia, and -in the absence of a clear argument to the contrary- I'll take the opinion of experts over yours any day of the week.

War Lord said...

The results of the study are weird. H1 definitely expanded from Southwestern Europe after the end of the Ice Age (What about H1 in North Africa and in the Saami?)

Andrew Oh-Willeke said...

"Then why doesn't it show up in any sort of admixture runs"

I concur with Dienekes on this point. Paleolithic autosomal contributions have fused into the various ancestral groups created by that program's cluster analysis methods. This study dates the time when that happened to prior to 2000 BCE.

I don't discount the possibility that non-U mtDNA may have been present in the Mediterranean basin, but not the rest of Europe, however, mostly because there seems to be evidence of that in Moroccan ancient DNA which may describe a population as old as 12,000 years ago, and because not much of the ancient DNA data is from far Southern Europe. But, if it is there, it had to be someplace where it didn't participate in the post-LGM repopulation of Europe.

GailT's comment certainly does cast doubt on the accuracy of the paper's estimate for the U expansion.

apostateimpressions said...

This paper makes perfect sense. U expanded in population size 15-10KYBP after the glaciers retreated and foragers spread northward. Thus U represents the Mesolithic substratum. H expanded 9-5KYBP as agriculturalists spread NW from the Near East. Thus H represents the Neolithic stratum. U shrank in size 9-5 KYBP as H expanded, as we would expect. But U and H melded and expanded together from 4KYBP onward during the Bronze Age.

The frequency of mtDNA haplogroups, defined by substitutions shared by related mtDNA types (Phylotree.org-mtDNA tree build 12), in early farmers across Europe [7], [10]–[13] was found to be overall similar to those in modern Europeans.

That fits neatly with D's hypothesis that the Bronze Age IE were an elite component similar to the Neolithic components: thus they dont show up as a massive population replacement. My inkling is that the relatively abrupt merging of Neos and Mesos may allude to the impact of the IE invasion.

U and H coexisted for several millennia and then merged circa 4KYBP: why? The IE invasions would provide an additional factor, from outside, that might explain the almagamation. It seems that the IE were particularly warlike, highly mobile and liable to conquer all in their path with their superior weaponry and their military attitude, as recorded in the archeological record and in IE religions. Thus they may well have formed a dominant caste over the Neos and Mesos, which cause the two to merge and for Meso culture to end. We know that Bronze Age IE Indo-Aryans formed a caste system and we can imagine something similar, if perhaps not so complicated, in Europe.

truth said...

Isn't mtDNa H supposed to be at least, Mesolithic ? Based on the recovered 8 mtDNA sequences from several Mesolithic sites from the Sado estuary in central Portugal, and identified 4 individuals belonging haplogroup H (including one H1b and one possible H7).

mooreisbetter said...

Can we all just agree Dienekes that it was really lame for you to appeal to authority when you (quite nastily) wrote the comment about "Oh you presume to know more than Svante Paablo?"

The very point of this board is to discuss the theories that are published by experts in scientific journals. We can and will assail their logic. They may have more formal training in DNA, but no one has a monopoly on logic.

And as for knowing more than Mr. Paabo, yes, some of us do. You should recall that he was on the cover of Newsweek about 20 years ago declaring multi-regionalism completely dead. Those of us who disputed the findings, disputed the logic, or simply intuited him to be wrong came out on the winning end of that one. Denisova anyone?

Dienekes said...

Can we all just agree Dienekes that it was really lame for you to appeal to authority when you (quite nastily) wrote the comment about "Oh you presume to know more than Svante Paablo?"

No we cannot, because all those who parade the 2005 Iberian conference paper as evidence that there was plenty of H in pre-Neolithic Europeans have to explain why it was never published in a journal, and has been universally ignored by scientists.

And, it's not lame at all to "appeal to authority", because I do in fact trust Svante Paabo about 1,000x more on matters of ancient DNA than commenters of this blog, including myself.

eurologist said...

But, if it is there, it had to be someplace where it didn't participate in the post-LGM repopulation of Europe.

Andrew,

That's exactly what one would expect, though. The people on the northernmost edge of the LGM refugia had an incredible advantage: they would quickly make use of the new resources and multiply - leaving no space for latecomers who first had their own, smallish new areas (warming mountains or now no longer deserts in the south) to expand into, and then would have to traverse through hostile regions to get to the north, which would have been pretty much fully occupied in 200-300 years with healthy population growth. And we know that the groups in the North lever left during the Younger Dryas.

It seems that the IE were particularly warlike, highly mobile and liable to conquer all in their path with their superior weaponry and their military attitude, as recorded in the archeological record...

There is no such record of bronze age invasion ~west of central Poland - in fact, there is cultural continuity, instead. Everything points to IE already being present in parts of northern/Central Europe before the bronze age.

Yes, the date of parallel U and H growth would appear relate to the Corded Ware culture - but that culture is anything but homogeneous. I am not sure I trust the authors' timeline, because it is way too late to suddenly absorb a HG component almost all across Europe (and S6 shows a different timeline, anyway). Too me it seems more plausible that agriculture was adopted later in the North (the Isles, Scandinavia, Baltics), and not as successful, initially, but when it became, it enhanced the total fraction of U in Europe.

I mean, we have no evidence of HGs persisting in most of Europe fairly shortly after the dominance of agriculture. Yet, look at S7: the U population was anything but isolated, instead blossoming pretty much during the same time period as H.

jeanlohizun said...

Dienekes said:

No we cannot, because all those who parade the 2005 Iberian conference paper as evidence that there was plenty of H in pre-Neolithic Europeans have to explain why it was never published in a journal, and has been universally ignored by scientists.
So I presume you don't consider PhD Catriona Pickard, Professor Clive Bonsall and PhD Ben Pickar scientists then? Because not only did they not ignore the Chandler et al(2005), they expanded on the conclusions of the paper, and showed how it is possible that the Neolithic samples are a subset of the Mesolithic sample. So now we have a study from the University of Edinburgh acknowledging the presence of mt-DNA H in pre-Neolithic Portugal, and acknowledging the validity of the results of Chandler et al(2005).

http://edinburgh.academia.edu/CliveBonsall/Papers/251471/Reassessing_the_mitochondrial_DNA_evidence_for_migration_at_the_Mesolithic-Neolithic_transition_2008_

So what is gonna be next?

dalouh said...

"This paper makes perfect sense. U expanded in population size 15-10KYBP after the glaciers retreated and foragers spread northward. Thus U represents the Mesolithic substratum. H expanded 9-5KYBP as agriculturalists spread NW from the Near East"

this paper is not making any reference to the known evidence(s) from North Africa{Taforalt) and lets not forget the Sahel..

"The West Eurasian component observed in the Tuareg is highly interesting. A major proportion (94%) could be allocated to haplogroups H1, H3 and V, West Eurasian lineages of Iberian origin that spread to Europe7, 10, 17, 26, 29, 36 and most probably North Africa30, 31 with the improvement of the climatic conditions after the retreat of the ice sheets 15 000–13 000 years ago."

a must read paper..

http://www.nature.com/ejhg/journal/v18/n8/full/ejhg201021a.html

so,the H and V were agriculturalists in the Sahara ???
ever heard about the Ibero-Maurusian period 22 to 8 ky ago?

eurologist said...

As far as U is concerned, it simply does not fit a neat pan-European paleolithic distribution with subsequent widespread replacement. Some Us are widely distributed (2,3,4,K), others are more or less absent in the North (1,6,8), and U5 is slightly enhanced in the North (of course one really needs to drill down deeper for a clearer picture, since all of these are very old).

That fits with the idea that not all southern populations were able to participate in the post-LGM expansion, and thus also grew differently post agriculture. If this is the case for U, why not similarly for H? We won't know until we have data from Southern Europe.

And I would love to see a comparison of the European H tree with a west Asian one, say sufficiently removed, like Saudi Arabia, Iraq, and southern Iran. With the full genome studies, age estimates should now become more realistic.

Matt said...

Maybe determining identical patterns in other mtDna haplogroups associated with the Neolithic and not with the Mesolithic (ones that aren't even contentious, like V?) would end this debate?

...

This paper seems pretty definitive in terms of H being associated with a major populaton expansion between the Neolithic to 4000. That's not something H expanding due to selection can really explain (why not before and why not afterwards?).

...

One question, although maybe slightly off topic - do we have much reason now to think that the Mesolithic and Paleolithic hunter gatherers themselves didn't, predominantly, expand from the same Near East region that the early agriculturalists came from, at the beginning of their "increase in population size around 15,000 to 10,000 YBP" and "end of the last glacial maximum in Europe ... northwards expansion of hunter-gatherer populations"?

My understand was the refugium in Europe explanation was popular because it explained present day patterns in human variation, but a 5000-10000 year separation between European HGs and incoming Neolithic agriculturalists seems more compatible with their levels of similarity and the ambiguity people seem to find with the signal.

eurologist said...

do we have much reason now to think that the Mesolithic and Paleolithic hunter gatherers themselves didn't, predominantly, expand from the same Near East region that the early agriculturalists came from...?

Firstly, we still don't know that agriculturalists came from the Near East. The only thing we know for sure is they came from the Balkans. What make-up they had, we don't know. It could be 100% local Balkan when starting out, or it could be 50% Near Eastern, or what-not. And then, current clines indicate that much of that was lost (i) during westward migration, and (ii) at the collapse of LBK. And we know at the beginning, LBK picked up all sorts of local mtDNA (and thus autosomal) along the way.

Autosomaly, it's mostly a question of what time scale you believe in. Typical extant European populations could be largely >25,000 years old, or significantly neolithic - until we have better data and better gauges for timing.

Secondly, we do know for certain that the most successful, most populous culture before LGM (the Gravettian) was truly pan-European, and so likely dominated the northern fringes of LGM refugia from France to the Balkans and the Caucasus, and perhaps Anatolia. Post-LGM culture in the North is decidedly a mix of Magdalenian from the West, and additional cultures perhaps more related to Central Europe, with some Eastern European influence.

Historically, migrations from the East beyond Eastern Europe have not been very successful in penetrating Europe or making a dent on the genetic make-up - despite several language changes (Scandinavia, Baltics, western Poland, Hungary, ...). And despite three different languages, Germany, Hungary, and surrounding Slavic areas toady have extremely close Fst measures. There is no reason to believe all this was different at earlier times.

Maju said...

I have been just pointed out that the Chandler 2005 paper is indeed cited in at least another study:

CATRIONA PICKARD, BEN PICKARD AND CLIVE BONSALL, "REASSESSING THE MITOCHONDRIAL DNA EVIDENCE FOR MIGRATION AT THE MESOLITHIC-NEOLITHIC TRANSITION" (available at academia.edu - you need account to read)

The authors, as I do, find Haaks 2005 data very unsupportive of demic replacement leading to modern Central Europeans because the resulting gene pool is very different from modern (notably N1a). Discarded the drift hypothesis (mathematical models fail), there are only two options left: later replacement (archaeological support is arguable for Central Europe) or selection, which is the choice for Pickard et al.

But I'd rather suggest demic replacement from West (Megalthic peoples?) and East (Indoeuropean) in the Chalcolithic. While modern genetic pools are only observed in the Bronze Age, there is a massive void of data for Germany and nearby areas (excepting the Indoeuropean colony of Eulau in East Germany) for the Chalcolithic period.

So I'm leaning for complex replacement in the Chalcolithic in relation with Megalithism, Kurgans, etc.

Dienekes said...

Lol, ok, you found someone citing that paper in a Polish Festschrift, by a couple of archaeologists and a molecular biologist that doesn't specialize in ancient DNA. I guess we can now all agree that it is high quality data.

Onur Dincer said...

Ancient DNA research is unlike conventional DNA research in that poor quality data too often hinders reliable research. Poor quality data is worse than no data. I am writing this as a general note of caution, not as a comment on any specific sample.

apostateimpressions said...

> Then why doesn't it show up in any sort of admixture runs? I'm yet to see in any such analysis, either by you, me, any other blogger or in academic papers: that component, which should show up very clearly and distinctly. It does not: nowhere! <

/ > One question, although maybe slightly off topic - do we have much reason now to think that the Mesolithic and Paleolithic hunter gatherers themselves didn't, predominantly, expand from the same Near East region that the early agriculturalists came from, at the beginning of their "increase in population size around 15,000 to 10,000 YBP" and "end of the last glacial maximum in Europe ... northwards expansion of hunter-gatherer populations"? <

That might explain why the Meso substratum is common to the Neo components and thus doesnt show up as a distinct component. If the Meso stratum also originated in the Near East then conceivably the Neo components already absorbed or contained it before they diverged from each other. Then in Europe they absorbed the same stratum there with low population density and low genetic diversity.

jeanlohizun said...


Lol, ok, you found someone citing that paper in a Polish Festschrift, by a couple of archaeologists and a molecular biologist that doesn't specialize in ancient DNA. I guess we can now all agree that it is high quality data.


Ad Hominems!!! Really Dienekes, is that all you got. Here another paper very recently published by the same team on the Cambridge Archaeological Research:

http://edinburgh.academia.edu/CliveBonsall/Papers/994781/Autistic_Spectrum_Disorder_in_Prehistory_2011_

Here is a little biography on the folks who published it:

Catriona Pickard teaches archaeological science and manages laboratory facilities at the School of History Classics and Archaeology at the University of Edinburgh. Her publications to date have focused on coastal archaeology, palaeodietary reconstruction and archaeogenetics, and include Submerge Prehistory(2011).

Benjamin Pickard is a Senior Lecturer at the Strathclyde Institute for Pharmacy and Biomedical Science (SIPBS), University of Strathclyde, Glasgow. His research interests focus on the biological and genetic underpinnings of major mental illness.

Clive Bonsall is a Professor of Early Prehistory at the University of Edinburgh. He has carried out excavations in Britain and southeast Europe, and is a Corresponding Member of the Romanian Academy of Scientists and an Honorary Member of the Bulgarian Academy of Sciences National Institute of Archaeology with Museum. Among his publications are: The Mesolithic in Europe (1989), The Human Use of Caves(1997), The Iron Gates in Prehistory(2008), and Submerged Prehistory.

But you are right, why should we take anything this folks said as legit, after all they are only two Archeologist and a Molecular Geneticist. Moreover, they did not perform any aDNA analysis, but rather compared the literature from Haak et al(2005) and Chandler et al(2005), and I think that a Professor like Clive Bonsall who has written books like : The Mesolithic in Europe (1989) might just be qualified enough to talk about the data in an archeological/historical context, the same applies to Geneticists like Ben Pickard. I’m not trying to say that just because they are experts we should take whatever they say an accept it, that would be an appeal to authority. Now to try to disqualify their research and arguments based in that they do not have according to you enough “experience” on the aDNA subject is just preposterous and the very definition of an Ad Hominem attack. The fact is that you claimed that the validity of the Chandler paper was in question due to it not being published in any journal, or cited by any scientists. That claim is bogus, because certainly the Chandler et al(2005) data is as valid as the Haak et al(2005) data, and the fact that it has been ignored points more to either ignorance of the data by some in the community, or an Ad Hoc approach by others, where the data that doesn’t agree with what is being proposed is disregarded as faulty or ignored for an ideological purpose.

jeanlohizun said...

Moreover, there is recently published paper by Deguilloux et al(2012) which can be found here:

http://onlinelibrary.wiley.com/doi/10.1002/evan.20341/full

Take a look at this excerpt from the study:

Although specific gene pool transformations can be identified only through the comparison of continuous diachronic aDNA datasets, ideally from the same region, aDNA samples are, for the moment, rare, and data from crucial periods are sometimes lacking. Moreover, diachronic data can concern different regions, complicating interpretations in terms of genetic continuity.
The late hunter-gatherers analyzed by Bramanti and colleagues31 are from geographically diverse locales, including sites in Germany, Poland, Lithuania, and Russia. Since they are compared to LBK farmers from Hungary, Austria, and Germany, we can wonder if the samples are appropriate to test genetic continuity in north-central Europe. In this case, the geographical and temporal structures of the communities could remain inextricable during genetic analyses. In consequence, we can ask ourselves if there is any acceptable argument that allows us to assume that currently available Mesolithic data apply to the pre-Neolithic Danubian.


[…]

The representativeness of the hunter-gatherer sample studied by Bramanti and colleagues can, however, be questioned. This sample encompassed a large geographic area and temporal span, thus forming a highly artificial population. The chronological heterogeneity of the sample implies that it incorporates pre-Neolithic individuals from central and northeast Europe, as well as hunter-gatherers postdating the first farmers' arrival in the region (50% of samples, Fig. 2), who may already have incorporated variants from incoming Neolithic groups. In that respect, late hunter-gatherers, such as those of the Ostorf site in Germany (around 3000 cal BC, site 30 in Fig. 2)31 may already have incorporated mtDNA lineages brought by the first farmers, who arrived around 5500 BC. Adding even more confusion to the analyses, the hypothesis that Ostorf people were former farmers who became fishermen around 3300-3200 cal BC has yet to be resolved.32 The geographical heterogeneity of the sample may make the neolithization processes appear simple, when in reality they may have been quite variable at the regional scale

Milton Anderson said...

Fishing played a large role in the hunter-gatherer culture. Farmers not only raised crops but herded animals. These factors may have raised conflicts between the two cultures.

Both would have wanted access to the best river sites, either for fishing or for watering livestock. Herders would have wanted large areas for pasture, which encroached on hunting grounds. Herders would have been able to clear large areas by girdling trees, even though large areas might not have been farmed by ox-drawn wooden plows.

In other contexts, hunter-gatherers have been severely impacted by rancher-herders, not farmers.

Dienekes said...

jeanlohizun, I will not waste any more time on you. Anyone can easily see that the Iberian conference paper has about two orders of magnitude fewer citations than Haak et al. (2005). You can check Google Scholar.


They're both from the same year and deal with the same topic: prehistoric Europeans. One was published in some regional Iberian conference, the other in _Science_. One has been cited to death, the other has been ignored. Experts in ancient DNA who are in a better position than any of us to know what research is reliable and what is not, have ignored it.

You can continue to believe that it's a reliable paper, and I can continue to doubt that it is. You won't convince me that it's reliable because you've found a couple of archaeologists who mentioned it in some Polish Festschrift.

This conversation is over.

Samequeen said...

Dienekes!
Thanks for posting that work. I had only to look at the piechart to see that it must be true for us in northern Scandinavia, U5 and K.

Yes, hunters and fishermen, for sure gatherers too are U5 and K. The works in Upsala university by Anders Götherström et al (Helena Malmströms diputation) shows that these two mt-DNAs existed among the Hunters populations in Scandinavia. The people as they could see (studing skeletons and teeth) never lived on farming, even if there were in their vicinity farmers.

There are still in northern Scandinavia this people, U5 and K.

GailT said...

To clarify my previous post - the results of nearly 500 FMS U5 mtDNA results in the U5 project are generally consistent with the Fu et al. paper, and to the extent that they differ, I believe the U5 project results indicate an earlier expansion of U5, perhaps around 20,000 ybp rather than 10.000-15,000 ybp. (Also, Fu et al only considered the coding region, so there analysis will also have less time resolving power.)

The U5 project combined with GenBank results also shows high diversity of U5b in Iberia. But one problem with the U5 project is sample bias (with much higher sample frequency in western and northern Europe).

I downloaded the Pickard et al 2008 paper, and I think it had some interesting speculation for its time but it is outdated now (for example, note their discussion of the lack of mtDNA J among aDNA in the neolithic) and also, their speculation is more nuanced than you suggest. More importantly, any speculation that does not include recent aDNA testing is obsolete. Unfortunately,even speculation that does include new aDNA results, such as Deguilloux et al(2012), doesn't really help clarify, it only identifies the uncertainties. The Deguilloux paper analysis could be easily summarized as: "large uncertainty, we need more data."

We have a small sample of aDNA from hunter-gathers that is mostly U5 and U4. That favors one theory over others but is far from conclusive. We have no strong evidence for H as the mesolithic people who repopulated Europe as the ice retreated. Maybe we will find such evidence, but it's just not there a present.

Most importantly, we really need to have full genome mtDNA testing of ancient DNA. The HVR results raise more questions than they answer. In some cases HVR cannot even distinguish U from H. FGS testing of ancient remains will also help refine estimates of mutation rates.

The people who I trust as authorities are those who are doing high quality DNA testing and who are also doing competent numerical analysis of the data. In this very new and immature field, having a well known name and getting papers with no analysis published in a journal does not always count for much, in my opinion.

Gail

GailT said...

Is there a link to the Chandler et al 2005 paper? Jean lists the test results on her site, but no dates are given.

What procedures were used to avoid contamination? Is that why the results have not been published in a journal?

Dienekes said...

Thousands of farmers schlepping their farm animals invaded and crossed the entire continent of Europe and ended up wiping out the entire population of Ireland, so that it is now 90% R1b.

It didn't take thousands of farmers. A few of them would suffice, the rest is a matter of exponential progress.

I thought farmers were, well, sedentary.

Farmers have expanded throughout the world in search of new farmland. That is why, for example that forests around the world have been retreating and farmland has been increasing.

Funny, my Irish friends don't look Middle Eastern.

There is no single "Middle Eastern" look. It is certainly the case that starting in the Neolithic there have been massive changes in physical type in Europe, which eventually saw the absorption of the robust low-skulled, broad-faced, long-headed UP Cro-Magnoids of Western Europe, by the people that live there today.

Mark D said...

Dienekes, first let me say that I have followed your blog for several years now and have enjoyed and appreciated your article postings and comments, and I thank you for your efforts in bringing these issues up for the discussions that have taken place here. As a layperson I have never felt qualified to write a response until now. I have read many books on the subject related to this article (at least the Y part), such as Sykes, Oppenheimer and Cunliffe. It just seems to me that their analyses make more sense, that a significant portion of the Celtic West trace their ancestry to before any Neolithic invasion, if not before the LGM. I understand there are strong debates about the dating of SNP mutations, that the science is very much in its infancy, and that there is to date very limited data. I hope everyone can keep an open mind on the subject and appreciate the arguments on both sides of any given question. Thank you again for your efforts.

jackson_montgomery_devoni said...

From all that I have read about on this issue it appears that at least so far the only mtDNA haplogroups that are conifrmed as being Paleolithic/Mesolithic in origin in Europe are U2, U4 and U5. The Chandler paper does not seem to be credible at all as it was done in 2005 back when they did not take extensive prcautions to avoid contamination. Apparently scientists only really started taking extreme precautions to avoid contamination of ancient bones back in 2009. So anything before that can be seen as fairly irrelevant.

The so called haplogroup H samples from Paleo/Meso Morocco also are not reliable. As Gail T has mentioned these studies did not focus on FGS results either. They are very questionable. I suppose that most people on here though will not want to hear this though and will simply keep believing in unreliable results and data.

All in all so far the evidence is pointing only to haplogroups U4 and U5 as being the true Paleolithic/Mesolithic haplogroups of Europe. This may change in time though as more reliable ancient DNA testing is done.

GailT said...

"It just seems to me that their analyses make more sense, that a significant portion of the Celtic West trace their ancestry to before any Neolithic invasion, if not before the LGM."

As in the case of mtDNA H, the theory that y-DNA R1b was the dominant mesolithic haplogroup in Europe was based entirely on it's current distribution in Europe. This theory requires that we assume people have never migrated in significant numbers to already populated areas, which we know to be a flawed assumption. Even in the historical period we know that there were large scale migrations of Germanic and Slavic peoples in Europe. So the original assumptions by Sykes et al. were really just poor scholarship - they jumped to an unsupported conclusion without considering any alternate theories.

And now we have substantial evidence that Sykes's theories were in fact wrong, based on the diversity of R1b, and more importantly, the complete lack of R1b in ancient DNA in Europe. Perhaps a new study will find ancient R1b in Europe, but until that data is reported, there is really no evidence that R1b was in Europe in the mesolithic, or for that matter, even in the early neolithic. The earliest evidence of R1b in Europe is 1,000 BC at the Lichtenstein Cave:

http://www.buildinghistory.org/distantpast/ancientdna.shtml

With each new study, the lack of R1b in Europe becomes more striking and makes Sykes's conclusions seem more implausible.

wagg said...

"Apparently scientists only really started taking extreme precautions to avoid contamination of ancient bones back in 2009. So anything before that can be seen as fairly irrelevant."

Hilarious. Somehow the old testings found diverse gene pools in their tests. But we should throw everything before 2009 away. Right...

In the ancient testings they did try to avoid contamination and sometimes they checked the hgs' testers to evaluate the results and the risk of contamination.

"so far the only mtDNA haplogroups that are conifrmed as being Paleolithic/Mesolithic in origin in Europe are U2, U4 and U5"

The problem being that we have still no data from the most populated and interesting European area yet (south (beside caramelli 2005), south-east and south-west of Europe).

Caramelli 2005 found no U, instead they found some R0 and N* (I'll let aside some vaguer stuff like Sungir's DNA (Russia, maybe as old as 28,000 BCE) that is apparently strikingly similar to HV4a2, H17'27 or H1j (IIRC) but that could still possibly be something else IIRC - the testing was not precise enough to be sure).

eurologist said...

Even in the historical period we know that there were large scale migrations of Germanic and Slavic peoples in Europe.

Gail,

But we also know that these migrations often had little effect on the genetic makeup of the populations; you can include Hungary, the UK, and others on that list. Outside of the core area of Slavic expansion, e.g., in western and northern Poland and in the Baltics, at most 5% to 10% of R1a subgroups can be attributed to recent (1,500 ya) migration. The remainder likely has been there for over 4,000 years.

cris said...

mr Mark D
of course Irishmen dont look middleastern nor do they look Greeks

Middleasterners lack the northeuropean component wich evolved out of the westasian component somewhere around the caucasus

Also by a mechanism of natural selection, in the snowy grounds of eastern and northeastern europe there was an increase of the ratio of the blond people because darkhaired people were an easy target on the snowy grounds of eastern and northern europe

However before Anglosaxons came to England I suspect that the percentage of blonde people in britain was lower

This saying ,many middleasterners from levant anatolia armenia iran and caucasus could pass as southern europeans see for example recep ardogan, bashar asad, ramazan kadirov, mohamed khaminei, mohsen makhmalbaf, serj sarkisian, dukashvili th georgian, saliman of lebanon and netanyahu of israel

jackson_montgomery_devoni said...

@wagg,

Go ahead and laugh it up bud. Like I said some people (such as yourself) will continue to believe in unreliable results.

We need to wait and see what new testing comes out of southern Europe before any firm conclusions can be drawn about which haplogroups were present there during the Mesolithic.

truth said...

@libyia

"in the snowy grounds of eastern and northeastern europe there was an increase of the ratio of the blond people because darkhaired people were an easy target on the snowy grounds of eastern and northern europe"

Explain then why Siberian tribes or Eskimos are not blonde. Your theory doesn't make any sense.

cris said...

@truth
mr truth, it's simply because Eskimos are not caucasoids to begin with

As you know light eye and light hair mutation arose amongst caucasoids somwhere around iran-armenia-anatolia-caucasus and from there expanded northwards and eastwards southwards and westwards, Eskimos on the other hand are not Caucasoids

truth said...

@ libya

And what matters that they are mongoloid ? According to your theory blondism appeared because of the snow, then why it hasn't appeared in mongoloids such as Siberians or Eskimo, given they live in the same conditions ?

cris said...

mr truth I never said that blondism appeared because of snow

the light hair and light eye mutations arose amongst caucasoids of the mideast but the ratio of blondism increased when those caucasoids migrated northwards by a mechanism of natural selection because predators ate much of the darkhaired mideastern caucasoid migrants to northern and eastern europe wich contrary to gulf stream affected western europe could see months with snowy grounds and indeed there are more blonds in ukraine than same latitude france

as for mongoloids they were not affected by light hair light eye mutation wich was born in neareast after the birth of the caucasoid race there

jackson_montgomery_devoni said...

Here is something that does not make much sense either regarding a Paleolithic/Mesolithic presence of mtDNA haplogroup H in Europe. If H was present during those periods then why did it not expand into Central and Northern Europe after the ice sheets melted like U4 and U5 appear to have done? I highly doubt that H would have just stayed behind in the south and not moved at all? That really does not make much sense.

Rob said...

Dienkes (or anyone); is there as much of a debate as to how to date mtDNA (eg compared to the various proposed methods for dating Y Hg STRs) ?