September 19, 2013

Ancient genomes mirror mode of subsistence rather than geography in prehistoric Europe (preprint)

This is an abstract of a preprint not currently available, and seems to build on previous work by the same authors.

Ancient genomes mirror mode of subsistence rather than geography in prehistoric Europe

Pontus Skoglund et al.

Recent ancient DNA studies have provided new evidence for prehistoric population structure associated with the contentious transition to an agricultural lifestyle in Europe. In this study, we infer human population structure and history in Holocene Europe by generating ancient genomic sequence data from 9 Scandinavian individuals associated with the foraging Pitted Ware Culture and the agricultural Funnel Beaker Culture (TRB). We obtained up to 1.1x coverage of the genomes for the nine individuals allowing direct comparisons of the two groups. We show that the Neolithic Scandinavian individuals show remarkable population structure corresponding to their cultural association. Looking beyond Scandinavia, we integrate this data with ancient genomes from Southern Europe and find that the Tyrolean Iceman from an agricultural context is most similar to Scandinavian individuals from a farming context, whereas Mesolithic Iberian hunter-gatherers are most similar to Scandinavian hunter-gatherers, opposite to what would have been predicted from their geographical origins. This finding shows that among these individuals, lifestyle is the major determinant of genetic ancestry rather than geography. Comparisons with modern populations reveal a latitudinal relationship where Southern European populations such as Sardinians are closely related with the genetic variation of the agricultural groups, whereas hunter-gatherer individuals appear to have the closest relationship with Baltic populations such as Lithuanians and present-day Scandinavians. Our results also demonstrate that while Middle Eastern populations are not the most similar to Neolithic farmers, this observation can be explained by African-related admixture in more recent times for Middle Eastern groups, which, once accounted for, reveals that the other major component of their ancestry resembles Neolithic farmers. While present-day Scandinavian populations are intermediate between the two groups, consistent with admixture, they appear genetically slightly closer to Neolithic hunter-gatherers than Neolithic farmers. This suggests a model where initial colonization by agricultural populations was followed by later admixture with hunter-gatherer populations or gene flow from other regions.

Link

19 comments:

sblog said...

October 18, 2013. Pontus Skoglund will defend his thesis: "Reconstructing the human past using ancient and modern genomes" see here http://www.ebc.uu.se/Research/IEG/evbiol/research/Jakobsson/news/

Davidski said...

Where is this pre-print currently "not available"? Is it about to appear at arXiv?

apostateimpressions said...

<< this observation can be explained by African-related admixture in more recent times for Middle Eastern groups >>

Is there any autosomal or haplogroup evidence for African admixture in the ME? Do they mean E3b (E-M215)?

Is E3b considered Caucasian?

Rokus said...

Or maybe Neolithic people rather mixed with a native population in the Middle East as much as in northern Europe. Why native Middle Eastern population wouldn't have been as similar to Africans as eg. Mesolithic Scandinavians were to Mesolithic Iberians?
The author makes a strange twist in favour of conventional propositions. I wonder this article will ever appear in a serious journal.

Jim said...

"Is E3b considered Caucasian?"

I'm not following. I thoght E originated in Ethiopia and most clades were in that area still. What do you mean by "Caucasian"? Originating in the Caucasus?

Fanty said...

@Aposta:

"Is there any autosomal or haplogroup evidence for African admixture in the ME? Do they mean E3b (E-M215)?

Is E3b considered Caucasian?"

As for Autosomal I recall the result of one tree making software that Dienekes once used.

It claimed that the Southern European component and the Southwest Asian (Arab) component had been the last of the "caukasoid" ("white") components to seperate.

And it claimed that the SW-Asian component got altered after the seperation from S-European by additional African admixture.

terryt said...

"Or maybe Neolithic people rather mixed with a native population in the Middle East as much as in northern Europe".

The author does specify that ,'Our results also demonstrate that while Middle Eastern populations are not the most similar to Neolithic farmers'. If the incoming Neolithic population had been already admixed with Africans the Middle East and Neolithic Europeans would not have been as different as they appear to have been. As the author futher says, 'this observation can be explained by 'which, once accounted for [African-related admixture in more recent times for Middle Eastern groups] reveals that the other major component of their ancestry resembles Neolithic farmers'. So the European Neolithic is much more similar to what earlier Middle Eastern populations were like. The Middle Eastern population has changed.

apostateimpressions said...

Jim,

"I thoght E originated in Ethiopia and most clades were in that area still."

Well, North Africa is heavy in E3b, Tunisia is over 70% and Morocco 80%, yet I think that they are usually considered Caucasian, although it seems that some taxonomists class them with south Asians. I dont think that anyone would class NAs with East Africans.

On the other hand, Ethiopia is 40% E3b. I suspect that might be back migration, they have hefty SW Asian admixture for one thing, but I would invite someone more learned to clarify on E3b.

"What do you mean by "Caucasian"?"

I think that it is to do with racial taxonomy, morphological traits, skulls, skeletons, temperament, psychology etc.

"Originating in the Caucasus?"

D postulated on the basis of the Fst distances of ancestral components that NAs, West Asians, SW Asians etc. inhabited an urheimat in the Near East geographically centered around the site of Göbekli Tepe until after 16,000 years ago.

quote:

It is therefore clear, that the common ancestors of the Six may have been a people living in some small area of Eurasia until well after 16 thousand years ago. Much like a bubble of space in cosmological inflation, their living space multiplied by orders of magnitude, coming to encompass a huge region of space within most of West Eurasia and North Africa. Within that region a fairly homogeneous population was established, the people traditionally called "Caucasoids."

http://dienekes.blogspot.co.uk/2011/12/womb-of-nations-how-west-eurasians-came.html

To answer my own initial question, as far as I remember, the admixture analysis showed little or no African admixture in modern Turkey/ Anatolia from which region anthropologists often postulate the expansion of the Neolithic into Europe. Thus I would question whether they can explain the differences between the Neolithic European stratum and the present Near East/ ME on the basis of African admixture. Besides, the West Asian component, which seems to have nothing to do with Africa, appears to be absent in the Neo E stratum. Perhaps the authors had some other ME population in mind. Syrians and Jordanians do have a little African admixture but a lot more W Asian.

http://3.bp.blogspot.com/_Ish7688voT0/TNBSb08B_DI/AAAAAAAACzk/ULDBu3eQvXM/s1600/ADMIXTURE10.png

eurologist said...

"Our results also demonstrate that while Middle Eastern populations are not the most similar to Neolithic farmers, this observation can be explained by African-related admixture in more recent times for Middle Eastern groups, which, once accounted for, reveals that the other major component of their ancestry resembles Neolithic farmers."

Or, perhaps, Mediterraneans have been very similar to "Middle Eastern farmers" without "African" = Semitic contribution even before LGM.

Valdric said...

Apostateimpressions, haplogroups are not races; and modern middle easterns do have some african-related autosomal DNA.

apostateimpressions said...

Fanty, OK thanks that is interesting. So the SWA component inclines toward Africa relative to the other Caucasian components. That implies admixture with Africans. SWA mixed after it seperated from the southern component that transmitted the Neolithic into Europe. That seems to fit with what the authors said.

Ironically some of that info was in the page by D that I went on to link to. Quote:

For example, the Northwest_African component is shifted (relative to the remaining five) towards the other African components; the Southwest_Asian is also thus shifted, but less noticeably.

[...]

In Arabia, the migrants would have met aboriginal Arabians, similar to their next door-neighbors in East Africa, undergoing a subtle African shift (Southwest_Asians).

Valdric, yes I agree haplogroups are not races (not today anyway) but they do tend to be associated with races. Eg. N/ O mongolian, I/ R/ J Caucasian. I suppose my question might have been whether E3b played a role in the original Caucasian population. I suspect that it is more likely that E3b became more or less detatched from its earlier autosomal make up through admixture and then transmitted Caucasian genes into Africa, especially north Africa. Likewise we dont really know what the earlier autosomal make up of R was. R is today associated with Northern Europe autosomes but seems to be introgressive into NE.

Thanks

Kurti said...

Damn, It's like they are copying my thoughts :D I was spreading this idea throughout Forums since almost a year. The reason why ancient farmers appear not similar to modern Middle Easterners (specifically South/Southeastern Anatolia and Levant, is because their was a more recent gene flow from Africa into this area (maybe coming with the Proto Afro Asiatic speakers) and also a more West Asian related component with Caucasian, Iranian plateau and to some extend probably with Indo-Europeans.

Kurti said...

@Rokus
"Or maybe Neolithic people rather mixed with a native population in the Middle East as much as in northern Europe. Why native Middle Eastern population wouldn't have been as similar to Africans as eg. Mesolithic Scandinavians were to Mesolithic Iberians?"

To consider this as possibility we would have to consider the Neolithic to have risen and started somewhere else, but how is that possible if all archaeological foundings point to a Middle Eastern origin of farmers? Also the major Haplogroups found among Neolithic farmers point to an Near Eastern origin.

It is actually your thoughts which are pretty twisted. farmers can't have mixed with a "native population in the Middle East" while farmers were the natives themselves and farming was spread throughout the Middle East 10000 BCE.

@apostateimpressions

Yes the Southwest Asian autosomal component is actually pretty much Mediterranean (the farmer component) slightly shifted towards East Africa.

apostateimpressions said...

If the Caucasus component, given its central MDS position among the other five Caucasian components, is most like the original Caucasian population, and the others are shifted toward non-Caucasian populations, the North European toward Asians, then toward whom is the Mediterranian component shifted? The Med component seems as far shifted from the Caucasus component as the South West Asian or the NE and I dont see why Meds would have more drift or selection than NEs.

http://4.bp.blogspot.com/-fw1LU3Ln7oM/TuYW3xfUu6I/AAAAAAAAEW0/hdBpE8PV7Jg/s1600/1_2.png

Perhaps the NE component mixed with older northern Europeans who were already shifted toward Asia, while the Med component mixed with old southern Europeans who were not so shifted toward Asia? Then the old NEs would have been more Asian shifted than the NE component? Modern NEs would be even less Asian shifted because of mixture with the Med component? The unexpected mongolian Mtdna in ancient N/ central European burials?

(Btw. D I suspect that my Firefox browser acquires adware, that causes Firefox to show pop under ads, from the advert at the top of your blog and without me clicking on the ad. I reinstalled FF and it was fine but the pop under ads returned after I visited your blog a few times and no other site with FF. The ads were initially along the lines of "lucky, you are the millionth visitor to D's site". Has anyone else experienced this? I use IE for most other sites and that remained fine.)

Ponto said...

I doubt North Europeans can identify past populations of Southern Europe or the Middle East by using SNP array results and admixture programs. The main problem is that to differentiate Northern Europeans from Southern European populations makes it a Northern European versus the world division. Lactase retention, light pigmentation and other diagnostic features of Northern Europeans are what separated them from the rest of humanity. The choice of SNP results produces a biased result.

It will be very interesting when European human remains, older than the Neolithic or its interface with Iberian or Swedish hunters/gatherers are tested for haplogroups. My opinion is that there will be no mtDNA U in those remains, and the Y chromosome haplogroups will be C, D and E. Everyone is jumping the gun based on some results of remains less than 10,000 years old. Results obtained on older remains have been questionable like those Italian results.

Caucasian is an antique, 300 year old, term for White Europeans and people who approximate them outside Europe. Look up Blumenbach the German on Wikipedia. E3b shows your age; it is obsolete, replaced by another designation. That haplogroup paragroup, not all its subclades in found all over Europe more so than J2 and its subclades. Hitler, Bonaparte...belong to that haplogroup. Yes, all negroes! Northeast Africa is overrated, given too much importance. It is essentially made up of mongrels of recent vintage.

AWood said...

@Ponto,

Not sure how to read into your post but I think your conclusions are false. My hunch is that you'll find North and South Europe are really two sides of the same coin, but different geography of hunter-gatherer. The only lineage in Europe during the mesolithic is R1. Despite E1b, J2, J1, and G2 being at higher nodes in the phylogenetic tree, I predict these are all "johnny come lately" Y lineages into Europe with the advance of farming. Let's throw I1/I2 in there for good measure as well. I would also say that ydna C and D were never in Europe and are purely East Asian lineages. I can't see any evidence putting them anywhere else. If we ever get aDNA from the Mesolithic, we should have an answer. It will only take 1 sample to prove the case.

Stenhög said...

The full thesis can be found here:

http://uu.diva-portal.org/smash/get/diva2:645462/FULLTEXT01.pdf

Dienekes said...

This looks like a thesis summary that references a number of papers that are the "meat" of the thesis.

Stenhög said...

Yes, you are right, sloppy of me. The doctoral thesis is referring to the upcoming paper with the posted abstract that is in your blogpost.

However there is some additional information from the thesis, like the sequencing of 8000 year old Gotlander with mtDna U4b1. Lets hope that you can get the raw data from that individual.