search this blog

Wednesday, September 30, 2015

Cranial affinities of Mesolithic populations from Eastern Europe and Siberia (teaser)


This looks like an excellent example of modern physical anthropology work. Unfortunately it's only an abstract. Can't wait to see the paper.

Mesolithic populations from the Eastern Europe and Siberia: cranial shape analysis with the help of geometric morphometric methodology

Ekaterina Bulygina(1), Anna Rasskasova(1), Denis Pezhemski(1)

1 - Research Institute and Museum of Anthropology, Moscow State University, Russia

Several Mesolithic and early Neolithic populations dated to 10,000 – 6,000 years BC from Russia, Romania and Ukraine have been analysed by means of quantifying their 3D cranial shape. The whole sample comprised 85 individuals, including Mesolithic and Neolithic groups from Yuzhny Oleni Ostrov (Russia); Vasilievka, Voloshkoe and Vovnigi (Ukraine); Varasti (Romania); Itkul and Ust-Isha (South Siberia) and Locomotiv (East Siberia). A comparative set of modern populations was sampled to include representatives from Europe, Africa, Eastern Asia and (native) America. Apart from the standard geometric morphometric procedures, we cluster ordinated data to establish potential relationships between groups and use permutation of individual distances to establish the significance of the group differentiation. The method of analysis is first verified with the help of the modern populations that have varied geographical provenance. We establish that no cranial data, whether the face and the neurocranium are analysed together or separately, allow us to recover geographical relationships between the modern populations in our sample. Nevertheless, clusters that have been recovered with the help of the whole cranium data correspond well with the expected generic relationships between the sampled modern groups. As a result, we choose to analyse the shape of the complete cranium, where such is available, in fossil individuals as well. Our results highlight a high level of variation within Mesolithic and within Neolithic populations of the Eastern Europe and Siberia as compared with the pooled sample of the modern humans from different geographical locations worldwide. However, a certain structure among the analysed groups can still be revealed. The results suggest that Mesolithic groups from the Dnieper region have close morphological affinities with each other, while Yushny Oleni Ostrov have a large overlap with modern humans in general and with some of the mongoloid groups in particular. Neolithic groups are, on the whole, closer to modern populations than to the Mesolithic sample. At the same time, Siberian individuals show a complex pattern of morphological relationships which may be revealing of their genetic identity. On the whole, our results invite further discussion on the origins and affinities of the Eastern European Mesolithic and Early Neolithic groups as well as call for the research into the impact that the choice of data has on the results of 3D morphological analyses. Acknowledgements: This work has been supported by the grant of the Russian Foundation for Basic Research No № НК 13-06-00045.

Source: European Society for the study of Human Evolution (ESHE) 5th Annual Meeting PESHE4 final abstracts volume.

31 comments:

Nirjhar007 said...

Very Interesting but no aDNA, Mesolithic to Neolithic aDNA from Don area can be a massive thing of course.

Krefter said...

There's mtDNA from Yuzhny Oleni Ostrov, Locomotiv, and I believe Ust-Isha or at least near by and dating to the same periods.They're all a mix of Eastern(A, D, C, Z) and EHG mtDNA(U5a, U4, U2e). mtDNA also suggests the two could have been closely related because they shared mtDNA haplotypes.

Nirjhar007 said...

this is interesting, has this been discussed before?.
A New Model of Human Dispersal
Trevor Underwood1
1 - Independent researcher
This presentation examines previously unpublished allele counts obtained from the French-San-Neanderthal-Chimpanzee alignment
of the high quality DNA sequence of a Neanderthal from the Altai Mountains [1]. is analysis indicates the existence of an
unidentified third archaic ancestor of present-day Europeans, which diverged from its common ancestor with sub-Saharan Africans
and Neanderthals around 900 thousand years ago. It also shows that the relative proportions of derived alleles of Neanderthals versus
sub-Saharan Africans versus the third archaic ancestor, in the 0.0826% the European genome that is not shared with the common
ancestor of humans and chimpanzee, are 13.6%, 32.3% and 54.2%, respectively. In addition, analysis of the also unpublished allele
counts from the alignment of the 45 Kya fossil from Ust’-Ishim in western Siberia [2] and of the alignment of the 36.2 Kya Kostenki
14 (Markina Gora) fossil from Kostenki-Borshchevo in European Russia [3] show similar relative proportions, suggesting that these
individuals were closely related to the ancestor of present-day Europeans. ese results differ significantly from previous estimates
of the proportion of Neanderthal ancestry in present-day Europeans which range from 1.3% to 2.7% [1, 4, 5]. is presentation
also identifies a mathematical error in the derivation of the admixture proportion estimators used to generate the previous estimates
which explains this difference. e analysis of the allele counts together with anthropological and archaeological evidence suggests
a new model of human dispersal based on a Eurasian lineage in the Levant, which admixed with Neanderthals between 250-55
Kya as they expanded eastward, and subsequently with members or descendants of the African mtDNA haplogroup L3 aer their
emergence from Africa between 84-63 Kya. This was followed by radiation from a basal admixed population in the Levant from
around 55-50 Kya, with no subsequent major contribution to the European genome. Ancestors of the Ust’-Ishim individual, a
member of mtDNA haplogroup R, probably went northeast from the Levant into western Siberia around 47 Kya; and ancestors
of the Kostenki 14 individual, a member of mtDNA haplogroup U2, probably moved northward from the Levant into the Central
European Plain between 40-36 Kya. It is likely that other members of these hybrid populations with a morphology similar to
present-day Homo sapiens, including mtDNA haplogroups N, R, U, U2, U8 and JT, expanded westward into Europe along the
Danube and Mediterranean coast and replaced the already dwindling Neanderthal populations between 45-35 Kya, rather than
newly emerged sub-Saharan Africans as has been assumed. If this new model of human dispersal is correct, it has profound implications
for the interpretation of the anthropological and archaeological evidence, which has largely been framed within the Recent
Out-of-Africa model.
Acknowledgements: I would like to thank Nick Patterson and David Reich for assistance in providing the unpublished allele counts
for the French-San-Altai Neanderthal-Chimpanzee alignment; Qiaomei Fu, Janet Kelso and David Reich for assistance in providing
the Ust’Ishim-San-Altai Neanderthal-Chimpanzee allele counts; and Martin Sikora, Eske Willerslev and Robert Foley for assistance
in providing the Kostenki-San-Altai Neanderthal-Chimpanzee allele counts.

Shaikorth said...

Krefter, Yuzhny Oleni Ostrov has just C1 and U, and a one H. It is the Karelia HG's site, so safe to assume they're fully EHG autosomally and at least R1a is present in the Y-DNA.

Karl_K said...

"It is likely that other members... similar to
present-day Homo sapiens... expanded westward into Europe... between 45-35 Kya, rather than newly emerged sub-Saharan Africans as has been assumed."

Really?! Wow! Now that changes what everyone has been assuming.

Chad Rohlfsen said...

The Karelian shows ENA affinity above the Samara hunter. The other Karelian bust shows clear ENA admixture. If EHG is maxed in the Samara, the Karelian looks about 5% ENA.

Chad Rohlfsen said...

Hmm, what is that paper concluding for the archaic amount in Europeans?

Shaikorth said...

Taking into account stats like this:

Samara_HG Karelia_HG Ami Chimp 0.0001 0.018 8794 8792 203451

and the fact that Karelia_HG is the better sequence, it's likely that the differences between the two in ENA sense aren't significant. Granted, those features in old Soviet reconstructions might reflect the EHG phenotype variation.

Alberto said...

That abstract posted by Nirjhar sounds quite amazing. Is it really saying that there is a 3rd population still unknown (different from Neanderthal and Homo Sapiens) that is the main (54%) contributor to modern non-Africans? And Neanderthal admixture as high as 13.6%? Or am I misunderstanding those figures?

That hypothetical 3rd population should have been very similar to Homo Sapiens if it's gone unnoticed until now. But it says it diverged from the common ancestor or Homo Sapiens and Neanderthal 900K ya.

Very strange, and quite amazing if true.

postneo said...

@alberto
I think Those percentages are stated for the European genome not shared with all humans and chimps. How big is thAt?

Karl_K said...

@postneo

It looks bogus. (IMHO.)

Alberto said...

@Postneo

Yes, they refer to the part not shared with chimps, which is only 0.0826% of the human genome. But it says that from that tiny specifically human part, 54% of the alleles are derived from an archaic population that was not Homo Sapiens nor Neanderthal. While from Homo Sapiens (or at least modern Africans), only 32% of the alleles are derived from, and ~14% from Neanderthal.

Or maybe he's saying something else?

Grain of salt, of course, for an independent investigator suddenly coming up with such figures and from an unknown species. We'll see what experts have to say about it when the presentation is done.

PF said...

I've found the full paper uploaded here: https://www.academia.edu/15581930/A_New_Model_of_Human_Dispersal_ESHE_2015_Poster_

It's sparse and math-heavy; generally difficult to make any sense of it. Here's the concluding paragraph:

"This analysis, together with anthropological and archaeological evidence, suggests a new model of human dispersal based on a Eurasian lineage in the Levant, which admixed with Neanderthals, probably between 250-60 Kya, and subsequently with descendants of African mtDNA haplogroup L3, followed by radiation from this basal admixed population around 55-50 Kya with no subsequent major contribution to the European nuclear genome."

What I *think* he's saying is that there was a very archaic population admixing with Neanderthals in the Levant prior to any Sapien OoA migration. Then, an initial migration occurred (mtDNA L3), which finally spread "radially" from the Levant to all of Eurasia.

The crazy part is that he estimates that this archaic hominid contributed way more to the European genome than either sub-Saharans or Neanderthals. His other point is that Neanderthal admixture has been underestimated due to a "mathematical error."

Looking forward to finding out soon whether this is complete quackery or... Anyways, I remain skeptical. :)

Nirjhar007 said...

Thx. :).

Krefter said...

@Shaikorth,

I forgot to mention I'm referring to Karelian mtDNA dating 3500 YBP and Hunter gatherer mtDNA from Siberia ranging 8,000-4,500 YBP. See here "Hunter Gatherers" mtDNA spreadsheet.

https://docs.google.com/spreadsheets/d/1I6YdqMXZFW6m6eaaCuggUNpBPVuP5vkri9-ncUW8KGs/edit#gid=0

There's quite a bit of East Asian mtDNA in 3500 BP Karelians: C5b1, Z1a, and D. I've read Z1a is a European-specfic form of Z, so it's evidence ENA has been in Europe for a long time. The Siberian HGs had mostly East Asian mtDNA(including shared C5b1 and D) and a significant amount of EHG(U4, U5a, U2e).

Anyways, I've read mtDNA C1 looks Eastern Asian not Western Eurasian. There's an assumption in some C1 doesn't exist outside of Siberia/America, but I've already found examples of C1 in Volga/Ural Russia. Also, a unique basal branch of C1 was found in Iceland. IMO, the C1 in Karelia is from East Asia originally.

Shaikorth said...

The 3500 BP site is Bolshoy Oleni Ostrov, quite a bit more northern in location than the Yuzhny site. The reason I brought the issue up is because only the latter is in the new study, and we know there was no D or Z1a there.

About the C1 in Karelia: it is also a "basal" branch and not particularly related to the East Asian subclades.

https://drive.google.com/file/d/0B9o3EYTdM8lQNDdMamFaWTVEQ1k/edit?pli=1

andrew said...

"We establish that no cranial data, whether the face and the neurocranium are analysed together or separately, allow us to recover geographical relationships between the modern populations in our sample."

This is the most important sentence in the abstract from the original post and one that seriously reduces my enthusiasm for seeing its results in more depth.

Karl_K said...

So for that off topic paper... he estimates modern Europeans are 68% Neanderthal or 'Other Archaic' both diverging (at least) 900,000 years before the present. This would put Europeans diverging from Africans greater than 600,000 years ago (as an average).

In this case, Europeans would be almost as diverged from Africans as Neanderthals were. Not to mention the mitochondrial and Y haplogroup situations. Seems kind of out there...

Krefter said...

@Shaikorth,

It's safe to say some-type of East Asian-ancestry existed in Bolshoy Oleni Ostrov and Karelia in later periods. Besides the mtDNA and autosomal(slight but real affinity to East Asia) evidence we have skeltal evidence.

From abstract.
"while Yushny Oleni Ostrov have a large overlap with modern humans in general and with some of the mongoloid groups in particular."

The degree and nature of East Asian ancestry has many possibilities.

Shaikorth said...

The morphological affinity refers to Yuzhny, not Bolshoy. Yuzhny is an EHG site, so the morphological affinities are an EHG-related feature insofar as they can be connected to autosomal ancestry.

It's nothing surprising that they might have affinities to East Eurasian groups, the cranial analysis included in the recent Raghavan paper has the Austrian Berg sample cluster with Buryats (a Mongolic population) instead of European samples in its tree, and that is on top of PCA affinities. In a genetic sense Austrians like other modern Central Europeans are less related to East Eurasian populations than EHG, and IIRC even SHG.

Krefter said...

@Shikorth,

What is suggest about morphology is possible. But to me it seems East Asian features are very specific and it would be surprising that the same specific features existed in unrelated people. Amerindians have a lot of Mongoloid features and their non-Mongoloid features are probably from MA1-types.

South Siberians who were found to have mixed EHG and Eastern mtDNA and were said decades earlier to have mixed Caucasoid and Mongoloid features. In Bolshoy Oleni Ostrov some had Caucasoid-type skulls and some Mongoloid.

Karelia_HG may be one of the Caucasoid-types, and if DNA was taken from others who were Mongoloid-types they might show a lot of East Asian ancestry.

If you think about it: EHG contributed maybe as much as 25% ancestry to North Europeans and is closely related to WHG and EEF. They're very close to Europeans and West Eurasians in general. It doesn't make sense for them to have any Mongoloid features.

Shaikorth said...

I doubt Yuzhny site is anything other than EHG, Bolshoy is over 3000 years younger. The craniometric affinities some ancient samples may show to East Eurasians or other genetically distant populations are nothing surprising because physical anthropology is not as reliable as genetics. Abstract itself says here: "We establish that no cranial data, whether the face and the neurocranium are analysed together or separately, allow us to recover geographical relationships between the modern populations in our sample."

I already mentioned Austrian skulls clustering with Buryats in the Raghavan paper, and here's another case of physical anthropology making continent-wide misses on skulls of modern populations: http://www.diva-portal.org/smash/get/diva2:667495/FULLTEXT01.pdf
pages 39-41

It would be no surprise if the EHG skulls would look more eastern than something like Turkmens on a craniometric analysis.

RobertN said...

"So for that off topic paper... he estimates modern Europeans are 68% Neanderthal or 'Other Archaic' both diverging (at least) 900,000 years before the present. This would put Europeans diverging from Africans greater than 600,000 years ago (as an average)."

Isn't this a variant of the original multiregional hypothesis iterated by anthropologists like Carletoon Coon, who thought that the different races were subspecies whose differences dated back to the original dispersals of Homo Erectus out of Africa? I think even Coon posited that there was an European Erectus lineage that split and gave rise to neandertals on the one hand, and a more gracile type on the other. Maybe the gracile would be the "archaic" in this paper, breeding with the more "classic" neandertals, with gene flow from the "homo sapiens" emerging in Africa.

Karl_K said...

@RobertN

But clearly this is absurd. Just look at this PCA plot.

http://dienekes.blogspot.com.au/2012/03/neandertaldenisovan-admixture-using-pca.html?m=1

The variation among modern humans is actually quite small. If there was an additional archaic group that admixed extensively with any non-African population, we would already be very aware of this fact. For very small amounts of admixture, however, you would need a reference genome.

Let's suppose that chimpanzees were extinct and there was no reference genome.

If a human genome was 0.001% admixed with a chimpanzee, you would only have about 30,000 SNPs difference. This is less than the difference between modern popuations. So, it would be a terrible assumption to think that these 30,000 SNPs came from admixture with chimps.

But, if you actually had a chimp genome, then you could actually make this kind of calculation and be confident about the results.

Matt said...

@ Krefter, you are correct that East Asian cranio-facial features are specific -

http://www.unsworks.unsw.edu.au/primo_library/libweb/action/dlDisplay.do?vid=UNSWORKS&docId=unsworks_1471

"East Asians are distinguishable from non-Asians on the basis of their tall, round, vault, shorter cranial length, tall faces that are flattened in the upper and mid-facial regions, short malars (anteroposterior length), narrow interorbital breadth and orthognathism. A north-south East Asian cline was also detected, with the northern samples exhibiting tall, orthognathic faces, and a long low vault. This long, low vault shape is in contradiction to the purported shape of cold-climate adapted populations. Southern East Asians possess a tall, rounded vault and a short, projecting (prognathic) face. Island Southeast Asians inhabiting the Andaman and Nicobar Islands exhibit a 'mixed' morphology, possessing the southern East Asian facial form, but the long, low vault seen in northern East Asian samples. "

From what I know though the problems with finding specific linkages to East Asians that aren't spurious are:

- The features on which East Asians (Northeast and Southeast) are a truly distinct group, relating mainly to facial flatness measures and tall faces in the orbits height, are only a subset of a larger number of cranial features, which aren't population structured. So the really different variables can get "swamped" in analyses by the larger number of variables, and through this you can get a spurious link.

- Unless the different features are scaled to overall size, the differences between "continental" populations seem to end up getting dominated by contrasts between generally large and small skulls and generally proportionately longer or higher or wider skulled groups. This is like in Shaikorth's example, where the Berg (European) and Buriat (Northern East Asian) samples are linked by having large, broad, low crania, even though they are not close on many other variables relating to the facial shape that tend to track continental ancestry.

Krefter said...

Thanks for the useful info Matt.

Onur said...

Matt,

So what truly differentiate Mongoloids from Caucasoids morphologically are the facial features, is that what you, or the thesis you quoted, are trying to say?

Nirjhar007 said...

Onur,
and where do Dravidoids of yours fit into this equation ? :D

Onur said...

Nirjhar,

The thesis Matt linked to does not cover South Asian populations, but it covers the Andamanese (it also covers Australians and Melanesians but I guess those are not particularly related to the ASI population of South Asia). Since I have not read most parts of the thesis, I cannot speak for it.

Nirjhar007 said...

OK ;).

Simon_W said...

@ andrew

Abstract: "We establish that no cranial data, whether the face and the neurocranium are analysed together or separately, allow us to recover geographical relationships between the modern populations in our sample."

andrew: "This is the most important sentence in the abstract from the original post and one that seriously reduces my enthusiasm for seeing its results in more depth."

I think you misunderstood it. In this quote they were referring to cranial measurements, rather classical ones I suppose. But if I understood the abstract correctly they applied a new method, where they measured the whole cranium, that is, more than the list of classical measurements, rather the complete 3D shape. And as they wrote: "clusters that have been recovered with the help of the whole cranium data correspond well with the expected generic relationships between the sampled modern groups."