June 06, 2014

Ancient mtDNA from pre-pottery Neolithic B

Figure on the left shows Fst values between the ancient PPNB population and modern populations.

PLOS Genetics DOI: 10.1371/journal.pgen.1004401

Ancient DNA Analysis of 8000 B.C. Near Eastern Farmers Supports an Early Neolithic Pioneer Maritime Colonization of Mainland Europe through Cyprus and the Aegean Islands

Eva Fernández et al.

The genetic impact associated to the Neolithic spread in Europe has been widely debated over the last 20 years. Within this context, ancient DNA studies have provided a more reliable picture by directly analyzing the protagonist populations at different regions in Europe. However, the lack of available data from the original Near Eastern farmers has limited the achieved conclusions, preventing the formulation of continental models of Neolithic expansion. Here we address this issue by presenting mitochondrial DNA data of the original Near-Eastern Neolithic communities with the aim of providing the adequate background for the interpretation of Neolithic genetic data from European samples. Sixty-three skeletons from the Pre Pottery Neolithic B (PPNB) sites of Tell Halula, Tell Ramad and Dja'de El Mughara dating between 8,700–6,600 cal. B.C. were analyzed, and 15 validated mitochondrial DNA profiles were recovered. In order to estimate the demographic contribution of the first farmers to both Central European and Western Mediterranean Neolithic cultures, haplotype and haplogroup diversities in the PPNB sample were compared using phylogeographic and population genetic analyses to available ancient DNA data from human remains belonging to the Linearbandkeramik-Alföldi Vonaldiszes Kerámia and Cardial/Epicardial cultures. We also searched for possible signatures of the original Neolithic expansion over the modern Near Eastern and South European genetic pools, and tried to infer possible routes of expansion by comparing the obtained results to a database of 60 modern populations from both regions. Comparisons performed among the 3 ancient datasets allowed us to identify K and N-derived mitochondrial DNA haplogroups as potential markers of the Neolithic expansion, whose genetic signature would have reached both the Iberian coasts and the Central European plain. Moreover, the observed genetic affinities between the PPNB samples and the modern populations of Cyprus and Crete seem to suggest that the Neolithic was first introduced into Europe through pioneer seafaring colonization.



bellbeakerblogger said...

This is a very exciting paper with some data on PPNB.

For one, the authors rightly torpedo the retarded Costa et al paper of 2013. The results show potentially a truly weird continuity in haplogroup K frequency among Jews and Druze within the same area since in the PPNB, albeit with very low samples - but still flummoxing.

The presence of African haplotypes is somewhat expected in this area. This is probably the pho-basal autosomal component indicated in the Near Eastern Farming populations stemming from a small component in the earlier Natufian ancestry. Glad that dna is backing up physical anthropology.

Interesting again, where it appears that the European Neolithic is descended from this PPNB population, it also appears that both the European Population (Brotherton) and the Near Eastern population (this paper) experienced massive changes in haplogroup frequency if not outright replacements. The only Near Eastern haplotype that remained appears to be the clades of K! wow.

eurologist said...

This is a really important study - hopefully one of many to come to characterize the initial Near Eastern (NE) and Greek/ Balkan Neolithic populations. I have been fairly critical about using extant NE population as a proxy, since such models assume that all changes only ever took place in Europe, and ignore the Balkans as a source. Also, there are very good archeological, climatic, and geographical reasons to assume that the Balkans and Anatolia had a shared population history since the Epi-Gravettian.

"...suggest that the Neolithic was first introduced into Europe through pioneer seafaring colonization."

This much has been clear from Greek archaeological studies for a while. However, trade was surely bidirectional, and so far there is no indication that pottery in SE Europe is anything but first and autochthonous.

It is too bad the researchers could not study more essential markers for a more highly resolved haplogroup determination. Also, the main result virtually is that the ancient PPNB sites had haplogroups which are either absent or fairly rare, today. So, that makes any conclusions tricky, since many pathways between then and today could have brought in such minute amounts into either/ any direction. However, the connection between PPNB, Cyprus, and Cardium appears pretty solid.

Finally, Fig. 4 makes clear what I have been talking about for a long time: the Neolithic arrived in the Balkans not just from anywhere in the NE, but specifically from S Anatolia and the N of the Levant, via seafaring in the Aegean and adjacent mainland Greece.

Mark D said...

This always made a lot more sense to me, and explains the distribution of J2 and other Med Y-halpogroups. It may even partially explain R1b and its presence in North Africa. Thus, Sardinia is not some "ancient" DNA repository but an amalgamation of multiple waves of Mediterranean colonization, both historic and Neolithic.

Locrian said...

I wonder if this was Y-hg J2?

apostateimpressions said...

They seem to assume that early Neo migrations into Europe can be tracked by that mtDna but they then say that there is no evidence for that, that in fact that mtDna is rare.

Who says that people from those three sites were responsible for the early Neo expansions into Europe? Maybe Neo went from them to others in the NE who then spread into Europe. Arguably it would have been those on the frontiers of Neo who would have been pressured to move northward.

Also do we know how that mtDna got into those three sites? Could a group of females have been absorbed into the relatively rich farmers, perhaps because they were a good looking breed?

eurologist said...


So far, archaeology tells us that the first pottery is in Greece, and that Greek sites, in a succession over centuries, adopted seeds/ plants (fruit/nut trees, pulses, cereals) that have their origin in southern Anatolia or northern Levant. So, this was not an agricultural package brought in by a wave of intruders, but the slow introduction of plant agriculture and later animal husbandry via seafaring trade. Only once the package was more or less completely assembled in the southern Balkans did farming spread via two distinct methods and routes (and distinct people): Cardium and LBK.

Of these, the Cardium people indeed may have a significant S Anatolian/ N Levantine contribution, but I very much doubt that about LBK. Also, there are numerous lines of evidence that point to the Balkans and Anatolia to have a shared population history since the Epi-Gravettian, anyway.

Dr Rob said...

@euro said " the Balkans not just from anywhere in the NE, but specifically from S Anatolia and the N of the Levant, via seafaring in the Aegean and adjacent mainland Greece."

Probably right . Indeed, NW Anatolia , ie just adjacent to the Bosopran land bridge appears rather "blank" in the Neolithic , although future research might change this picture

Guy Jacks said...

@bellbeakerblogger It seems like you're partially retarded, K1a9 is descended from U8b'k which originated in Northeastern Italy, and probably then migrated to the Near East, in fact it's not even supposed to be called K, the only reason it is called K is because people didn't know it was a subclade of U8bk. Best read a book mate.

andrew said...

1. An obvious put important observation about this data set is that the PPNB population in this region had a heterogenous mix of mtDNA haplogroups that are not particularly closely related to each other phylogenetically.

This is contrary to the general trend in which many historically earlier populations (e.g. Upper Paleolithic West Eurasian AMHs, at least outside Mediterranean coastal regions which are dominated by U5 and U4) tend to have little mtDNA diversity and to have closely related mtDNA haplogroups where there is diversity.

This suggests a Fertile Crescent population history in which multiple populations which had historically been distinct extended family clans from geographically distant regions from each other had already blended into a melting pot culture around or before 8000 BCE. The Mesolithic era in the Fertile Crescent was not boring, but involved obscure by not bland or simple population genetic modifying events. In other words, even in the hunter gatherer era, the was a lot of notable prehistory going on that is relevant to modern population genetics in West Eurasia.

In contrast, mainland East Asian mtDNA population genetics are less muddled. We do not see this degree of jumbling in much of East Eurasia until much later (often well into the East Asian Neolithic).

2. Could anyone suggest or provide a concise summary of the mtDNA distribution of the areas that are the source of the ancient mtDNA samples today for comparison sake? Is there any good way to quantify of the extent of the difference in mtDNA mix between this region ca 8000 BCE and today?

An ability to describe how Fertile Crescent population genetics have changed over time has wide applicability to a variety of studies that out of lack of any other reference points routinely use modern Near Eastern or West Asian populations as proxies for populations in the same areas from early prehistory and draw broad conclusions from these comparisons, despite the fact that we know for sure that this is inaccurate, although not necessary precisely how and to what extent it is inaccurate.

3. Also, query to what extent we can use the PPNB mtDNA population genetic sample to develop a better Baysean prior regarding the likely Y-DNA population genetic mix and autosomal genetic affiliations of PPNB populations in this region.

4. Keep it classy.

@ bellbeakerblogger "torpedo the retarded Costa et al paper"

@ Guy Jacks "It seems like you're partially retarded,"

There is no need for ad hominem attacks and crass insults in discussions of a technical academic subject. in each of these comments, resort to this kind of rhetoric when facts alone can make your case undermines your credibility.

Alexandros said...

Finally! Genetic data from the Neolithic Near East!! I have been waiting for such data for years and I was so busy at work that I have just spotted this paper! I am glad that Dienekes was so swift as usual and already included it in his blog.

So.. from what I can see this paper partly resolves (at least from the maternal lines point of view) a question that has puzzled many people, including me! Which modern population better represents the ancient Near Easterns and more specifically, the ancient Near Eastern innovators who brought agriculture to Europe? Being the administrator of the Cyprus DNA project, my hypothesis of course was that the modern Cypriots are a really good proxy for the early Levantine farmers. The clues were coming from 2 sources: (1) The very high frequency of mtDNA haplogroup K (linked to agricultural expansion into Europe) among Cypriots; and (2) a high proportion of 'Caucasus' autosomal admixture with lower admixture from SW Asia, relatively low admixture from continental Europe and practically no African admixture.

Many people writing in this and other blogs have derived a lot of theories regarding the modern population which better represents the early Near Eastern farmers. The most popular candidates were the Druze, the Syrians, the Palestinians, etc. No one gave any attention to the 'humble' Cypriots.

This study however proves without any doubt that the closest Near Eastern population to the first farmers are the modern Cypriots! In terms of mtDNA at least..

When the numbers speak the theories crumble.. Here are the proportions of individuals carrying the PPNB haplotypes of the ancient Near Easterns, as reported in the paper.

Cypriots: 13.2%
Ashkenazi jews: 11.3%
Druze: 7.8%
Jordanians: 4.0%
Turks: 3.8%
Beduins: 3.4%
Palestinians: 3.1%
Syrians: 2.5%
Sephardic Jews: 1.8%

I can't wait for autosomal data from these samples!

Simon_W said...

The table of the pairwise Fst values offers some surprising insights:
1. Modern Europe has a smaller distance to PPNB than either LBK-AVK and Cardial. This seems to be the result of later West Asian expansions.
2. LBK-AVK is closest to modern Near Easterners but almost equally close to modern Europeans.
3. Cardial is closest (and indeed very close) to modern NE Africans, what a surprise!