search this blog

Thursday, October 5, 2017

Upper Paleolithic genomes from Sunghir, Russia (Sikora et al. 2017)


Over at Science at this LINK. Not surprisingly, these four Sunghir individuals are very similar to another Upper Paleolithic Eastern European, Kostenki14, in terms of both genome-wide genetic structure and uniparental markers (Y-haplogroup C1a2, mtDNA-haplogroups U2 and U8c). If you can't access the paper, the supplementary materials are freely available here, and there's a press release here.

Abstract: Present-day hunter-gatherers (HGs) live in multilevel social groups essential to sustain a population structure characterized by limited levels of within-band relatedness and inbreeding. When these wider social networks evolved among HGs is unknown. Here, we investigate whether the contemporary HG strategy was already present in the Upper Paleolithic (UP), using complete genome sequences from Sunghir, a site dated to ~34 thousand years BP (kya) containing multiple anatomically modern human (AMH) individuals. We demonstrate that individuals at Sunghir derive from a population of small effective size, with limited kinship and levels of inbreeding similar to HG populations. Our findings suggest that UP social organization was similar to that of living HGs, with limited relatedness within residential groups embedded in a larger mating network.

M. Sikora et al., Ancient genomes show social and reproductive behavior of early Upper Paleolithic foragers, Science 10.1126/science.aao1807 (2017).

See also...

The genetic history of Ice Age Europe (Qiaomei Fu et al. 2016)

198 comments:

Anonymous said...

No ANE in Eastern Europe at 34.000 ya, according to admix graph. Apparently they wintered the LGM out further east.

Chad said...

ANE is these guys, plus East Asian. ANE is a hybrid and not some uniform pop. ENA admixture will vary by date and location.

Anonymous said...

Admix models from the paper:

Vestonice: 84% Sunghir + 16% Villabruna
El Miron: 48% Villabruna + 56% Goyet
Loschbour: 91% Villabruna + 9% Goyet

@Chad

It's remarkable that such a small pop is considered to have caused such a wide influence: American Indians, Iran and CHG, Indo-Europeans, Han chinese, etc etc

Matt said...

@epoch2013, if you go by the model Ancient North Eurasians being Sunghir+ENA (which seems worth testing eventually in qpGraph, though seems hard to know how anyone can know without testing), then it may not be a single small population and it may just be than all these populations are varying independent mixes of the Basal_Eurasian+Villabruna+Sunghir+ENA ancestries. I am not sure about this though.

MaxT said...

Y-haplogroup C1, mtDNA-haplogroups U2 and U8c

C1 Y-DNA again!? amazing to see how wide spread it was once.

@Chad
Very unlikely based on the paper. Although i agree that ANE could be hybird but Sunghir is not it.

@Matt
I agree, they appear to be too Kostenki14-like.

Chad said...

ANE is essentially UP Euro-related plus ENA. I showed this long ago. Reich & Lipson (2017) showed the same. Believe what you wish.

Matt said...

It'll be shown by dropping the Sunghir samples into a qpGraph, or it won't. (Question is whether these specific people are right, not whether some UP Euro related group divergent from the samples in Fu et al is!) Not much room to debate around it before anyone does that.

@MaxT:C1 Y-DNA again!? amazing to see how wide spread it was once.

Though these y chromosomes are not necessarily closer to those individuals today who are in present day C1 groupings, than R is to G...

Very distant branches whose whole lower branching tree has left no descendants today...

http://3.bp.blogspot.com/-5cWLu_X84is/VQRecy3MHyI/AAAAAAAAJ_0/8fnpuYYQ_4Q/s1600/figure1.jpg - in our history, the GT tree left lots of descendants all over Eurasia from East to West, while the C tree has left very few, only in the Eastern fringes of Eurasia. (I would say only due to luck, but who knows?).

(Sunghir population seems to had a tiny population size also, compared to Ust Ishim, who himself is the representative of an extinct lineage, albeit, as a fairly derived node on the GT tree closer to the people who did leave widespread descendants today).

Rob said...

So R1b expanded from SEE....and Italy

MaxT said...

@Chad

I'm well aware of what Reich & Lipson (2017) showed :- ANE as mix of Pre-K14 and Pre-Ami. They said whatever admixed into ANE is OLDER* than K14. This rules out both K14 and Sunghir.

There is also no ANE affinity in Sunghir as shown in the paper. That's all.

Mark B. said...

Not relevant to the topic at hand, but every time I see a science paper with unnecessary abbreviations (HG, UP), I want to kick a puppy. It makes reading the papers more difficult, and I can't imagine that the characters saved are being donated to illiterate children in the Third World. Unless you've got a good reason - deoxyribonucleic acid - write it out.

Anonymous said...

Chad, this is not so, you write something that contradicts the text. ANE is not a mixture, or rather, no mixes do not exist, all components are mixtures.
Guys of Sunghir is not ANE. They like Kostenki extincted.

Anonymous said...

@Rob

When Sunghirians lived, the R1b did not exist in nature. Even R was not.

Ryan said...

@Epoch - Keep in mind that the areas that likely had larger population sizes are also much harder to get samples from. We may be seeing a biased picture here. In fact, a lot of the best sites are probably under a few hundred feet of water.

Rob said...

Very interesting contextual info:

"The third individual in Burial 2 (SIV) was represented only by the diaphysis (midsection)
of a human femur, which had been polished, hollowed-out and filled with ochre, and
carefully placed next to SII. That it is a single element (and not an entire skeleton), and evidence from its bone chemistry which indicates “a different geographic origin for that individual and/or a contrasting post-mortem history,” suggests that the SIV femur section might have been an heirloom piece from an individual who died earlier than SII and SIII. Regardless, that individual presumably had not died significantly earlier than SII and SIII, since his remains were still available to them to collect. Hence, it is of similar age, perhaps someone who had died relatively recently or within generational memory of the individuals in the double burial"

He probably died out on a trek or hunt somewhere, and was brought back for burial. Or at least a piece of his leg was.

Rob said...

@ Supernord

"When Sunghirians lived, the R1b did not exist in nature. Even R was not."

Yes I agree. But what's that got to do with what I said ?

Chad said...

You don't pay attention. I never said they are ANE. ANE is a group like this, plus ENA.

Chad said...

Sister clade to K14 and a pre-Ami. It does not rule them out.

MaxT said...

Here is branching chart from Sunghir paper. ANE component is older** than both K14 and Sunghir. Again, in line with Reich & Lipson (2017), whatever admixed into ANE is older than K14.

http://oi64.tinypic.com/t6cvh5.jpg

Anonymous said...

They do an extensive paper, full of info and then put .jpg minuscule graphs inside, instead of vectors, which means that when you zoom you can't see anything.
No Bayesians as well. Is Reichlab that much better in formating their work?

Chad said...

That phylogeny is what happens when you don't account for ENA in MA-1. If you actually paid attention and used these programs, you would know that.

Chad said...

I'll post a better phylogeny from my work. Hold on.

Matt said...

I'm not sure what else "ANE is these guys plus ENA" could mean other than a special relationship between the Sunghir samples and ANE? At least we've got it cleared up.

@MaxT, right, as you noted, no sign of any enriched sharing with Sunghir and ANE derived (MA1, AG3, EHG) in any of their measures in the Supplement.

Not Fig. S26, not Fig. S19, not Fig. S18, not nuthin'.

The qpGraph do show that Sunghir, Vestonice, Kostenki, GoyetQ116-1 are all effectively a clade to MA-1 (none is more related and they all diverge from a population which had split from MA-1's ancestor).

Though as Chad would probably note, they do show a worst Z stat for relatedness between Villabruna and MA-1, which is likely mediated through ancestry from a population related to ENA.

Chad said...

You're cherry picking one comment over another. ANE is UP Euro-related plus ENA. Neither of these UP will be much closer to MA1 without ENA. They're not differentiated enough for any real significance.

Chad said...

If you want more specifics, the difference in drift between the samples and anyone else is quite small. They are not very differentiated. However, if you go f4 (Onge, MA1; UP_Euro, UP Euro), there is no real difference between Kostenki and Vestonice. However, both are about significantly better than Goyet as the admixing source. The trees do the same thing. A branch from by Vestonice and Kostenki are the admixing source for ANE. Also, drift lengths between these samples is very tiny when you put them on the same tree.

Chad said...

I'm pretty certain of this. I've done literally hundreds of runs trying to get the HG phylogeny right, from UP to Mesolithic.

AWood said...

So what is the relationship of Sunghir to modern groups today? I'm assuming a very equally weak relationship to all Eurasians today?

If so, what does that imply about La Brana and the entrance of WHG, presumably via NW Anatolia 14,000 ybp? Perhaps these Y lineages were gradually overcome by the I2, and R1 newcomers after the LGM?

Anonymous said...

"So what is the relationship of Sunghir to modern groups today?"

Sunghir (& Kostenki) treat Vestonice cluster (Y C1a2, mt U8c) which has no modern descendants. Essentially, these people almost did not participate in the formation of modern populations or their participation was minimal, through mt U2.

Ric Hern said...

How can we say that a certain population did not contribute or participate in the formation of modern populations when we see Neanderthal contributed without leaving any sign of their Y-DNA or MtDNA Haplogroups ?

Gioiello said...

"Rob said...
So R1b expanded from SEE....and Italy"

Thanx

Seinundzeit said...

Chad,

"ANE is essentially UP Euro-related plus ENA."

I think that makes more sense for WHG.

Or at least that's the impression I've gathered from my conversations with Mark Lipson. I guess the general idea is that WHG is mostly UP European-related, but with additional ENA and ANE admixture.

MA1/AG3 don't really have much (if any) UP European-related ("West Eurasian") admixture.

Rather, they are primarily descended from a stream of populations which constituted a "sister clade" to "West Eurasian" (basically, "North Eurasians proper"), but with additional ENA admixture.

These Sunghir genomes seem to be West Eurasians without ENA admixture, not North Eurasians without ENA admixture.

(We can't apply the modern conceptualization of West Eurasia to these ancient populations, since modern West Eurasians are a complex mix of ancient West Eurasian, North Eurasian, ENA, Basal Eurasian, and African ancestries)

Chad said...

WHG also has the relationship with farmers, not in MA1 or UP Euros. That is, I think, where the difference is. The difference between Kostenki and Vestonice from what is in MA1 seems very minimal if there is any at all. I don't like ghosts. One can simply materialize one anywhere on a graph for many things. It doesn't mean it's real.

Chad said...

Anyway, plug in the programs and try this yourselves. You will not get a West Eurasian source into MA1 without branching between Kostenki and Vestonice, or branching off from pre-Vestonice, after breaking with Kostenki. It's the only way to keep the Z below 3.

Tesmos said...

''So R1b expanded from SEE....and Italy''

How does this paper prove that R1b expanded from SEE and Italy?

Matt said...

@Chad: An effective almost trifurcation between the UP Euro related parts of MA-1, Kostenki-Sunghir and GoyetQ116-Villabruna with MA-1 shallowly on the K-S side seems very possible compared to the Lipson model of MA-1 basal to the other UP Euros that Sein relates.

The drift lengths (e.g. simple outgroup f3 statistics) just don't seem to fit with MA-1's West Eurasian ancestry drifting with the Sunghir-Kostenki subgroup for any substantial length of time.

I did want to say though about: "Also, drift lengths between these samples is very tiny when you put them on the same tree", this paper's supplement S10 notes:

"Sunghir / Kostenki 14 - We find that SIII shows substantial population-specific drift with all tested individuals, except the other individuals from the same site. The lowest estimates outside Sunghir are obtained with Kostenki 14, consistent with results from the ancestry analyses. However, despite their affinity, the results also show substantial amounts of drift specific to Kostenki 14 after its divergence, therefore rejecting a directly ancestral relationship to Sunghir. Estimates are high for both Sunghir and Kostenki 14 when compared to later European HGs, suggesting that despite their shared early European ancestry, they did not form a direct ancestral group to the later European HGs in our dataset."

Anonymous said...

@Chad

"WHG also has the relationship with farmers, not in MA1 or UP Euros. That is, I think, where the difference is. The difference between Kostenki and Vestonice from what is in MA1 seems very minimal if there is any at all. I don't like ghosts. One can simply materialize one anywhere on a graph for many things. It doesn't mean it's real."

But for this case (farmer relation) the ghost may be real. I read that Ofer Bar-Yosef considers the Levantine Aurignac to be very real, to have a very real connection to very early West-European Aurignac. If you take a look at the D-stats in Fu et al that paper uses Iraqi-Jew. If you do the same D-stats but swap Iraqi-Jew for Anatolina, Natufian, Iran_NL and Iran_CHL you'll find that Anatolian and Natufian show similar affinity to WHG as Iraqi_jew, Iran_NL shows nothing and Iran_Chl show some.

Couldn't there have been a ghost population in Europe around the LGM, apart from the usual suspects, with roots in the Aurignac but different from Goyet/Magdalenian? Something must link WHG to Natufians without Natufians coming to Europe because there is no Basal in WHG.

Rob said...

@ Tesmos

Slightly lighthearted remark, but looking at it R1b- L754 & I2a-L46o do seem to correlate with proto-Villabruna at a GW level; and they can have only expanded from SEE (sensu latu).

Nirjhar007 said...

From the paper :

''Sunghir 3 clusters with an individual from Nepal (nep-0172; 96/100
replicates) carrying the C1a2-defining V20 mutation, albeit with an early divergence close to the split with haplogroup C1a1 (represented by individual JPT-NA18974 from Japan) (Fig. S8). The deep divergences and widespread geographical distribution observed in the descendants of these haplogroups suggest a rapid dispersal of these lineages during the Upper Palaeolithic.''

Anonymous said...

R1b and I2a come from completely different sources. I2a is a local pan-European haplogroup leaving the roots in the West Asia, R1b it came in epipaleolithic from Siberia or the Urals. The fact that they were distributed in the Epigravettian
culture, it does not say that they further spread from Italy or from SEE. The eastern Epigravettian culture was widespread in the Northern Black Sea region also, where we see R1b and I2a in the Mesolithic and Neolithic.

Anonymous said...

You are too. My was not. Siberian Mal'ta is R*, I & I2a is in all Europe, J is the West Asian.

Rob said...
This comment has been removed by the author.
Rob said...

^ neatly

Blogger said...

I2a is pan European no?

Anonymous said...

Rob, you are wrong. Y-hg IJ > I has roots in the West Asian.
Siberian Mal'ta 24000 BP is R*, I & I2a is in hole Europe, J is the West Asian.
European Paleolithic mtDNA extincted.


Your reasoning is that R1b spread from Italy and SEE nothing at unreasonable, it's just nobody knows. Your reasoning about R1b and I2a somehow related by descent or distribution at all unjustified. You contradict all the studies. I wrote to you that the spread of I2a has nothing to do with the spread of R1b, they have completely different sources & routs & times of spread and of fates.

Blogger said...

@ supernord

I agree. Some idiots try and lump Haplogroup I in with R, when they are completely different. It's the same bs with people who lump haplogroups G,J, E and T together with all being of them being "Neolithic". Just not a fact.

Rob said...

@ supernord

"Rob, you are wrong. Y-hg IJ > I has roots in the West Asian. "

No duh, sherlock. But that´s not what you said, you said I2a is ´leaving its roots in West Asia´, whatever that means. So youre inaccurate, I2a had long been in Europe when it expanded.

" Siberian Mal'ta 24000 BP is R*, I & I2a is in hole Europe, J is the West Asian."'
You are talking about MUP. i am talking about the LUP -glacial expansions.
Different epochs, go look it up.
And I2a was NOT in ´hole´(i gather you mean whole) of Europ, because the Aurignacians and Magdalenians are not I2a, not even the Dolni Vestonice Gravettians were.

-"Your reasoning about R1b and I2a somehow related by descent or distribution at all unjustified"
no its not unjustified, you just cant grasp the details. For a start, i never suggested that R1b and I2a are phylogenetically related, I do know how the alphabet goes
Rather, i suggested that present evidence points to L751 and I2-L460 closely correlating with WHG/ UHG / and WHG-containing late glacial -mesolithic groups.
To repeat, this does not mean I am saying R1b originated in SEE or Europe, nor that they exactly correspond.

@Richard

"Some idiots try and lump Haplogroup I in with R"
If you weren´t an ignorant snowflake cry-baby, you´d gather what i meant.

Ebizur said...

The 95% confidence intervals of YFull's TMRCA estimates do not provide any reliable evidence that might allow one to reject a hypothesis of a simultaneous origin Y-DNA haplogroups C1a2 and I, perhaps concomitant with the entry into Europe of AMHs and the appearance in that region of Upper Paleolithic industry.

Focusing on Haplogroup C1a2:
(B2-M182 TMRCA 59,200 [95% CI 56,100 <-> 62,400] ybp)
A1b1b-M32 TMRCA 54,800 [95% CI 51,100 <-> 58,600] ybp
E-M96 TMRCA 52,500 [95% CI 49,600 <-> 55,400] ybp
B2b-M112 TMRCA 51,000 [95% CI 48,000 <-> 54,200] ybp
E1-P147 TMRCA 49,600 [95% CI 46,400 <-> 52,800] ybp
F-M89 TMRCA 48,800 [95% CI 46,300 <-> 51,400] ybp
GHIJK-M3658 TMRCA 48,500 [95% CI 46,200 <-> 50,900] ybp
C1b-F1370 TMRCA 48,300 [95% CI 45,000 <-> 51,600] ybp
D-M174 TMRCA 47,000 [95% CI 44,100 <-> 50,000] ybp
C1a-CTS11043 TMRCA 46,700 [95% CI 42,300 <-> 51,300] ybp
B2b1-M192 TMRCA 46,200 [95% CI 43,200 <-> 49,400] ybp
K-M9 TMRCA 45,400 [95% CI 41,400 <-> 49,600] ybp
K2-M526 TMRCA 45,400 ybp [95% CI 41,400 <-> 49,600] ybp
S1-B255 TMRCA 45,100 [95% CI 41,800 <-> 48,500] ybp
IJ-P130 TMRCA 42,900 [95% CI 40,300 <-> 45,700] ybp
LT-M2603 TMRCA 42,600 [95% CI 39,700 <-> 45,500] ybp
E1b-P177 TMRCA 42,400 [95% CI 39,600 <-> 45,200] ybp
C1b2-B477/Z31885 TMRCA 41,900 [95% CI 38,600 <-> 45,300] ybp
K2a1-M2313 TMRCA 41,500 [95% CI 37,400 <-> 45,600] ybp
E1b1a-V38 TMRCA 40,100 [95% CI 37,200 <-> 43,000] ybp
H-M69 TMRCA 39,600 [95% CI 37,000 <-> 42,300] ybp
(E2-M75 TMRCA 38,300 [95% CI 35,200 <-> 41,500] ybp)

Focusing on Haplogroup I:
(GHIJK-M3658 TMRCA 48,500 [95% CI 46,200 <-> 50,900] ybp)
C1b-F1370 TMRCA 48,300 [95% CI 45,000 <-> 51,600] ybp
D-M174 TMRCA 47,000 [95% CI 44,100 <-> 50,000] ybp
C1a-CTS11043 TMRCA 46,700 [95% CI 42,300 <-> 51,300] ybp
B2b1-M192 TMRCA 46,200 [95% CI 43,200 <-> 49,400] ybp
K-M9 TMRCA 45,400 [95% CI 41,400 <-> 49,600] ybp
K2-M526 TMRCA 45,400 ybp [95% CI 41,400 <-> 49,600] ybp
S1-B255 TMRCA 45,100 [95% CI 41,800 <-> 48,500] ybp
IJ-P130 TMRCA 42,900 [95% CI 40,300 <-> 45,700] ybp
LT-M2603 TMRCA 42,600 [95% CI 39,700 <-> 45,500] ybp
E1b-P177 TMRCA 42,400 [95% CI 39,600 <-> 45,200] ybp
C1b2-B477/Z31885 TMRCA 41,900 [95% CI 38,600 <-> 45,300] ybp
K2a1-M2313 TMRCA 41,500 [95% CI 37,400 <-> 45,600] ybp
E1b1a-V38 TMRCA 40,100 [95% CI 37,200 <-> 43,000] ybp
H-M69 TMRCA 39,600 [95% CI 37,000 <-> 42,300] ybp
E2-M75 TMRCA 38,300 [95% CI 35,200 <-> 41,500] ybp
B2a-M150 TMRCA 38,100 [95% CI 35,100 <-> 41,100] ybp
(NO-M214 TMRCA 36,800 [95% CI 34,300 <-> 39,300] ybp)

However, the interclade TMRCA of I1 and I2 is significantly less:
I-M170 TMRCA 27,500 [95% CI 25,200 <-> 29,800] ybp

The coalescence age of I2 and the coalescence age of C1a2a (extant European C1a2 plus a basal branch found in an ethnic Armenian) are significantly younger still, and not significantly different from each other:

I2-M438 TMRCA 21,800 [95% CI 20,100 <-> 23,500] ybp
C1a2a-V86 TMRCA 21,400 [95% CI 19,200 <-> 23,600] ybp

The Y-DNA ancestry of the tested La Braña specimen seems to have diverged from that of extant European & Armenian members of C1a2a during the course of C1a2's expansion into Europe with Upper Paleolithic industry. I would like to see the Y-DNA of the Sunghir specimens, the Nepalese individual (nep-0172) mentioned in the passage quoted by Nirjhar, and a rumored Algerian member of C1a2 added to YFull's tree to see where they fit in vis-à-vis La Braña.

Anonymous said...

@Rob, you wrote "So R1b expanded from SEE....and Italy" that it is not fact. This unfounded statement in any case.

You wrote "looking at it R1b- L754 & I2a-L46o do seem to correlate with proto-Villabruna at a GW level; and they can have only expanded from SEE" that it is not fact. This unjustified statement in any case.


Rob said...

Perhaps hypothetical, even speculative, but not unjustified.

Where is the earliest eivdence of UHG / prot-VB admxiture ? - Dolni Vestonice in ECE, none in western Europe Aurignacian (Goyet), none in EE plain (SUnghir, Kostenki´s).
Where is the next earliest evidence of its presence - El Miron & VB.

So where did UHG - and its associated lineages - expand from ?

Rob said...

Then observe the distribution of the lineages themselves during the LUP:
I2a1 - West europe, Italy
I2a2 - ECE / central-nrth Balkans
I2c - Scandinavia, Balkans, Anatolia

What's the likely centre ?

Have you had a look at R1b -L754 ?

L754
- V88

- L389
--CTS1010
-- P297
--- M73
---- M269
---- PF7562
---L23

Ryan said...

@Rob - "So where did UHG - and its associated lineages - expand from ?"

Probably the Balkans or the Near East. I think that's what we're all betting on here at least, no?

Grey said...

Richard Holtman said...
"Some idiots try and lump Haplogroup I in with R, when they are completely different"

Although I agree I and R have mostly different histories don't some of the I-E expansions (e.g. Germanic) seem to have picked up ydna I along the way (as allied tribes maybe)?

Grey said...

wildly speculatin but iirc i read somewhere that some of the mammoth herds who traveled to America came back again. i wonder if any mammoth hunters followed them back?

Rob said...

@ Ryan

Apparently not everyone (see Supernords comments; although I think he doesn't include the Black Sea region as part of the "south of eastern Europe").

Ryan said...

@Rob - He's just talking about R1b though, no? I would agree that at best that was a waystation for R1b. R1b pretty clearly arrived from the east, and from the East Eurasian side of ANE's heritage.

Rob said...

@ Ryan

R1b pretty clearly arrived from the east,

Yes, Obviously . But Siberia ?
Where's all the R1b in Siberia ? We have 20 + LUP- Meso-Neo samples. Zero R1b

And my point was that L754 lineages seem to have diversified around s o e Europe, which they clealry did.

"and from the East Eurasian side of ANE's
heritage."

I think it's more Complex than that
Villabruna, Blatterhohle (pre-M269 WHG -EEF hybrid) dont have any ANE
The R1b in iron gates and Latvia have minimal EHG/ ANE.
So what evidence is your statement based on ?

AWood said...

@Rob

The latest Mathieson paper did model Latvian HG as partially EHG. The Ukranian HG were also similar to EHG and R1b, rather than WHG. I suspect that the meeting point between WHG and EHG was extremely early, such that some early branches of R1b, like the NE Italian "Villabrunan" looked exactly like their western I2 counterparts genetically over a few thousand years.

Recall that even some I2, and C-V20 that appear in the European Neolithic look almost identical to their EEF counterparts, a process of admixture over a few centuries. It must have really depended on the tribe, and which HG groups were absorbed while others remained isolated and died out.

Ryan said...

@Rob - Villabruna had ANE (~5% according to David I believe), and measurable ENA ancestry too.

Ryan said...

I don't think we need to invoke EHG here at all either. R1b seems to be a WHG lineage, and EHG has it by virtue of being ~half WHG. Villabruna's ANE is just from a much older migration.

Rob said...

@ Ryan


@Rob - Villabruna had ANE (~5% according to David I believe), and measurable ENA ancestry too.

Yeah maybe it had 2% ENA in it. Big whoop

Tesmos said...

@Ryan,

I would be bit more nuanced about R1b being a WHG marker, I think it's more safe to say that some subsets of R1b may be connected with WHG. The ''ancestral'' subset of R1b could be still of EHG origin. All in all, R1s seem to be likely interconnected.

Rob said...

@ Awood


"
The latest Mathieson paper did model Latvian HG as partially EHG."

They had even less EHG than Motala

- "The Ukranian HG were also similar to EHG and R1b, rather than WHG."
Curiously these Ukrainian Mesolithcis that are shifted toward EHG are virtually all I2a2.
When R1b starts appearing in Neolithic, there is a slight WHG shift.

Also, we have blatterhohlewhich is basically WHG / EEF, with no ANE.

And again, there are samples from Mesolithic & EN Siberia. Are you aware of the absence of R1b of any sort ?

Ryan said...

@Rob - If that 2% ENA is where R1b came from - which it almost certainly is - then it is significant.

I do think you may be right about SEE as the home of R1b, or at least most of R1b though, with Lake Baikal as the home of R1 only.



@Tesmos - I highly doubt that EHG as a population is older than WHG.

New Neanderthal genome btw - http://science.sciencemag.org/content/early/2017/10/04/science.aao1887.full

Better fit than Siberian Neanderthals for introgression into us. I'd be curious to see if it's a better fit than Iberian Neanderthals for the extra Neanderthal people in East Asia have.

Rob said...

@ Tesmos

"The ''ancestral'' subset of R1b could be still of EHG origin"

Its interesting you think that EHG even existed 20 kyBP

Ric Hern said...

@ Rob

Just remember that I2a dominated the earlier European samples and very few untill recently expected R1b to be anywhere near the Balkans during the Mesolithic....There is still a huge area between Southern Siberia and the Balkans. R1b might surprise us again...

Anonymous said...

Rob, you are confusing the territory of Siberia & Ural with the ANE component. Siberia does not mean ANE component. Villabruna cluster does not like Vestonice cluster. Villabruna cluster much closer to Siberia and the Urals, as evidenced by the D-statistics:
Mbuti Karelia_HG Villabruna ElMiron -0.0434 -9.616 568639
Mbuti Karelia_HG Villabruna GoyetQ116-1 -0.0654 -12.425 625624
Mbuti Karelia_HG Villabruna Kostenki14 -0.0733 -14.719 827456
Mbuti Karelia_HG Villabruna Vestonice -0.0619 -16.221 701000
Mbuti Karitiana Villabruna ElMiron -0.0089 -2.279 589338
Mbuti Karitiana Villabruna GoyetQ116-1 -0.0127 -2.887 658456
Mbuti Karitiana Villabruna Kostenki14 -0.0132 -3.16 907963
Mbuti Karitiana Villabruna Vestonice -0.0157 -4.594 736700
See also Fig. 3. B in this study.

Anonymous said...

@Rob "see Supernords comments; although I think he doesn't include the Black Sea region as part of the "south of eastern Europe""

"south of eastern Europe" is not equal your SEE=SouthEastern Europe, see https://en.wikipedia.org/wiki/Southeast_Europe , Northern Black Sea region is not its part.

Rob said...

@ supernord

NO I agree that Siberia isn.t synonymous with ANE, and to clarify again, its abundantly clear that R1b is ultimately eastern.
So if you're suggesting the Urals ad Siberia, - where exactly ? Is there any clear evidence for a uralian refuge ? I know there are some cave´type sites in the Urals, which might vae been visited during the LGM, but anything else ?

@ Ric

"Just remember that I2a dominated the earlier European samples and very few untill recently expected R1b to be anywhere near the Balkans during the Mesolithic"

Yes, but a few of us did expect it ;)

Rob said...

"south of eastern Europe" is not equal your SEE=SouthEastern Europe, see https://en.wikipedia.org/wiki/Southeast_Europe , Northern Black Sea region is not its par"

Which is why I clarified abundatly "sensu latu" - as in the wide meaning, also taken by the recent Mathieson paper. The Black Sea littoral & adjacent Anatolia, the northern Balkans etc. This was a unit during the LGM. Obviously I am not referring to Crete or Peloponessus.

Ryan said...

Also I'd suggest taking a look at the qpAdm runs in Mathieson that compare autosomal DNA to X chromosomes. They have the increase of EHG in the Middle Neolithic coming entirely from male WHG. Admixture of 16% male WHG and 0% female WHG. Pretty stark contrast there.

Doing a similar analysis with Bell Beakers as a mixture of CWC + the Neolithic population that fits best (or maybe Neolithic + WHG) could tell us for sure to what extent Bell Beakers received their steppe ancestry for men and to what extent it came from women.

capra internetensis said...

@grey

http://www.nature.com/articles/srep44585?WT.feed_name=subjects_evolution for mammoths

The Asian mammoth mitochondrial DNA clade which seemingly back-migrated from America (hg DE) is estimated to have split from American sister branch close to 500 000 years ago, and the Eurasian DE clade (which contains some European specimens) looks to be around 200 000 years old.

According to the 2008 paper Siberia went from being dominated by haplogroup A before 44 000 years ago to being all DE afterward (but DE was already present as a minority earlier). But there was only one European sample in that one. In the new paper there are a bunch of European samples; it seems that some DE showed up but Europe was still predominantly B2 in the Upper Palaeolithic.

This is all mitochondrial DNA though. Obviously they need to step up their game and get into full genomes.

So anyway in a nutshell probably mammoth herds moving from eastern Siberia toward Europe yes, but from America in the time frame you are thinking of no.

Olympus Mons said...

@Ryan,
Sure from man CWC that became R1b in the process.

Rob said...

@ Ebizur

Yes, it is interesting that C1a survived in the 'peripheries' of Europe - Iberia & Anatolia, with its offshoots in EN Bulgaria & Cardial Dalmatia, eg.
As you said, the modern Europeans who are C1a are derived from the Anatolian branch rather than La Brana, and the 2 diverged as much as 43 ky ago.

Gioiello said...

@ Rob
“Yes, but a few of us did expect it ;)”

Actually till a little ago only one… but I recognize you having contacted me the first of these eurogenicians already a long ago.

Ryan said...

@Olympus Mons - I think probably the opposite but at least we can test it.

MaxT said...

@Chad

Lipson (2017) did account ENA admixture in the study, and came to Pre-K14 and Pre-Ami conclusion. Some kind of pre-UP and some kind of divergent ENA.

Chad said...

I know Reich & Lipson accounted for ENA. Fu et al didn't. Again, you are not paying attention. Also, no where in Reich & Lipson do they say pre-K14. It is a sister branch. Which works well as a split from Vestonice. Once again, you refuse to look at anything I type. If you aren't going to actually read what I actually type, I will not respond to you.

Anonymous said...

Upper Paleolithic Europeans certainly influenced Mal'ta and Afontova Gora a whole lot.
Actually, it goes around 40~50% for Mal'ta and 60~70% for Afontova Gora.

MaxT said...

That is not a sister branch but something older as their Fig.4 shows. Lazaridis also shows something similar.

You said you dont like ghost population, but studies make good use of them for a reason. In Lazaridis, we can see that there is a ghost population that was ancestor to ANE and W.Eurasian.

http://oi64.tinypic.com/imrxiq.jpg

MaxT said...

Chad, I have previously commented on your qpAdm work on Karitiana.

@Chad Rohlfsen
https://drive.google.com/file/d/0B962TtPkX1YnQjMtRGpmbE53U28/view?usp=sharing"

You did not use ANE in this run. Karitiana who score 45% ANE (in Basal-rich K7 cal), we see majority of their ANE was disappearing into ENA in your graph, Karitiana end up being 83% ENA and rest 17% K14 (instead of 45% ANE+ 55% ENA). K17 itself appears to have 23% ENA on your qpAdm.

Based on qpAdm score, i can see that it was a successful test. We see a pattern where ENA ancestry exists in both groups. Again pointing to some ghost population, older, less divergent than K14.

Chad said...

You're off-base. Just trust what I am telling you. It isn't older, but a sister clade. Fig 4 is the same thing. You're not understanding at all! It doesn't branch off before Kostenki's line, but at it. Forget some old qpAdm. I have no time to argue this with you. If you want to see a tree, go to Anthrogenica on this subject. I posted one and will put up several others tomorrow if I have time.

MaxT said...

@Chad

I understand quiet well what Reich, Lipson, Lazaridis are saying, I'll stick to their peer-review journals on this on subject. I'm not here to argue with anyone either.

Chad said...

Right..... It's not an older branch, but a sister branch. Think of it as a tuning fork. One side doesn't come off before the other. I don't know how much easier I can try to explain this for you. The West Eurasian side of MA1 wouldn't be a little closer to Vestonice than Kostenki if it were diverged earlier than Kostenki. You would be getting into the Ust_Ishim time-frame. But, believe whatever you wish.

Grey said...

@capra
"So anyway in a nutshell probably mammoth herds moving from eastern Siberia toward Europe yes, but from America in the time frame you are thinking of no."

that's a shame - it would have made an entertaining story

Anonymous said...

Vestonice cluster (Vestonice, Kostenki, Ostuni) WHG belongs to East Gravettian culture aka Kostenki-Willendorf culture. It has not ANE and was not involved in the formation of ANE.
Villabruna cluster WHG belongs to Epigravettian culture (& to other final paleolithic cultures). It has ANE-like participants (EHG?).
Epigravettian means "after gravettian", it did not come directly from the Gravette.

Anonymous said...

@supernord

I might be mistaken but I don't think these D-stats mean Villabruna doesn't like Vestonice, it means Villabruna shares considerably more drift with EHG than Vestonice does. Considering that Villabruna can be modeled as having a small ANE contribution and considering that EHG is modeled to be part WHG I'd say that is normal. That doesn't mean Villabruna is linked to Siberia. See the differences between EHG and Karatiana, both roughly half ANE:

Mbuti Karelia_HG Villabruna ElMiron -0.0434 -9.616 568639
Mbuti Karelia_HG Villabruna GoyetQ116-1 -0.0654 -12.425 625624

Mbuti Karitiana Villabruna ElMiron -0.0089 -2.279 589338
Mbuti Karitiana Villabruna GoyetQ116-1 -0.0127 -2.887 658456

Anonymous said...

epoch2013, Villabruna cluster and Vestonice cluster are both WHG, but they different.
I wrote "Villabruna cluster ... has ANE-like participants (EHG?).", "Vestonice cluster ... has no ANE"-participants, in their formation.

Mbuti AfontovaGora3 Villabruna Vestonice -0.0199 -3.844 213665
Mbuti AfontovaGora3 Villabruna Kostenki14 -0.0281 -4.465 218311
Mbuti AfontovaGora3 Villabruna ElMiron -0.0144 -2.382 187772
Mbuti AfontovaGora3 Villabruna Bichon 0.0009 0.109 218941
Mbuti AfontovaGora3 Villabruna Loschbour -0.0042 -0.569 216528
Mbuti AfontovaGora3 Bichon Loschbour -0.0045 -0.639 229661
Mbuti AfontovaGora3 ElMiron Vestonice -0.0081 -1.169 179675
Mbuti AfontovaGora3 Kostenki14 Vestonice 0.0085 1.217 199942


Clustering look at Fig. S18. of this paper. You can see that Villabruna cluster and Vestonice cluster don't alike, although both is WHG.

Grey said...

obsessing about mammoth some more...

this post always gets me

http://eurogenes.blogspot.co.uk/2016/05/following-mammoth-herds.html

the green dot mammoth clade looks like it might have been separated from the rest by some kind of mammoth barrier around the Urals which then disappeared around the 14-24K mark allowing mammoth clades from the east to move west.

i was originally thinking that would make a very convenient r1a/r1b split but apparently the dating doesn't fit but as a time for R1 generally to follow herds west maybe?

if the R1b/R1a split was some geographical barrier further east and R1b were on the western edge of it they might show up first?

(e.g. the pink vs dark blue mammoth clades on the maps - what barrier might have separated them?)

(and after all the megafauna were killed off they had to move off the steppe and settle in or near the various maritime refuges?)


Samuel Andrews said...

My mtDNA database keeps growing and now I have enough mitogenomes to show irrefutable of recent Steppe mtDNA in Iran. Several pan-European U4 and U5 clades show expansions around 5,000 years ago and several of them are also found in Iran. All of them have been found in LNBA Europe.

Ric Hern said...

@ Grey

I don't know how accurate this is but there is a theory about the Volga River and Caspian Sea being absolutely huge covering the whole Khvalynsk Basin and almost reaching as far North as the Urals when the Glaciers started to melt. Maybe this created the substantial barrier before and after the LGM ?

MaxT said...

@Chad

"You would be getting into the Ust_Ishim time-frame."

Younger than Ust_Ishim, as various studies show.

"But, believe whatever you wish."

Reich/Lipson et al, Lizardi et al, Fu et al, are all wrong according to you.

Figure S6.4 from Fu et al. also agrees with others.
http://oi65.tinypic.com/6zvwcx.jpg

I'll go with what they have to say. ANE component formed after Ust_Ishim and before W.Eurasian, even if it's admixed.

Anonymous said...

@supernord

Just to be sure: Most of us use the abbrevation "WHG" for Loschbour because Lazaridis used it in his paper. De abbrevation obviously means "Western Hunter-Gatherer", which would certainly include Gravettiab, but I think it is used as a synonym for the Villabruna cluster, which excludes Gravettian.

Matt said...

@MaxT, thing is neither Fu's nor Reich/Lipson's models actually resolve relationships between ancient UP Euro related HGs (inc. MA1 and AG3) at the same time as resolving relatedness between them and East Eurasians.

Ancestry from a population which is an outgroup to the main UP EuroHG clade could well be making MA1 look like more like an outgroup to the UP EuroHG clade, if it's quite significant, and if there's very little shared drift with MA1's ancestry as a member of the UP EuroHG clade (and it would have to be very little).

This is currently just not resolved at all in the published papers.

Grey said...

@Ric Hern

yeah, that's the sort of thing i was imagining - not necessarily a barrier to humans but a barrier to animals they hunted / scavenged and so indirectly a barrier.

Matt said...

@Rob, not confident enough in my knowledge of the Upper Paleolithic European samples to answer. More adna is always better. Tianyuan Man's sample will maybe shed more light on what is happening with differential relatedness to East Eurasians, which is a major question here (since it is at the root of where MA-1 and Villabruna cluster members fit with other UP Euro-Siberians).

Semi on topic ramble, when motivated to go back to Fu et al, I had a look over their qpWave analysis again. qpWave is something I've sort of ignored before as telling us what we already know. The function is use f statistics on a two sets of "left" populations and "right" populations, to get an N of how many sources are needed for differential relatedness in "left" pops to "right" pops.

("The papers that introduced qpWave showed that if the Left populations are mixtures – in various proportions – of N sources differently related to the Right populations, then all f4- statistics of the form of Equation S13.1 will be consistent with being a linear combination of N vectors of statistics that correspond to the mixing populations, and the matrix will have rank N-11,2. We can test whether this is the case using a Hotelling’s T test").

Anyway, in the case of the Villabruna cluster, this produces "Rank 2 is consistent with the data (P=0.41). This again supports the model of few as three ancestral populations, although the truth could of course be more."

"The qpWave analysis is able to document that at least three sources are necessary to account for the allele frequency correlation patterns in Villabruna Cluster samples. However, we did not succeed at convincingly modeling the ancient relationships among these sources."

They go on to suggest that they were unable to totally model these sources, however, they suggest 1) a "common thread" (main ancestry), 2) a source for "variability in the degree of allele sharing with ElMiron/GoyetQ116-1" and 3) finally, a source which relates to the variability in sharing with East Eurasians (it is important to note that V cluster is not homogenous on this, nor does this simply covary with sharing to Magdalenian / Aurignancian).

Apparently, they do not interpret that separate sources are required for relatedness to Near East or MA-1 itself, within the VB cluster, to the limits of their resolution (which we could consider more extensively now with the greater numbers of Euro Mesolithic HG samples published in 2017, e.g. Iron Gates, Latvian, etc. and to note they do just use WHG proper and not include Motala_HG, Karelia_HG, etc.).

So perhaps the question for V cluster is these three sources really, and where they fit on a tree. Then understanding how this integrates with the archaeology of the Epigravettian...?

On that, wonder if the labs are using qpWave in this way - first identify sets of subclusters, then run qpWave on each cluster to understand the minimal set of sources each cluster breaks down into, then try and fit them all on onto a qpGraph, with the constraint that members of each cluster should have the number of sources on the qpGraph as required by qpWave.

Angantyr said...

@Ric Hern

It's beyond theory; geology doesn't allow any doubts about the early Khvalynean basin filling up and overflowing into the Black Sea for a while post-LGM.

And while the overflowing Caspian and (no to be ignored) Aral basins, and the spillways these created, did form barriers, I think one must consider that this not that dry Central Asia also provided a more hospitable environment for both HGs and perhaps also early pastoralists than what you find there today.

Ric Hern said...

@ Angantyr

Yes I agree. Central Asia could have looked much different.However how did the Khvalynsk Basin look like Pre-LGM ? Was this not also a warmer period with some melting maybe not as extensive as Post-LGM but enough to maybe create huge marshes which most animals avoid travelling through ?

Ric Hern said...

@ Angantyr

And how did the landscape North of the Caspian look like during Winter times ?

Did the Volga Delta and Volga River freeze over and allow some animals and humans to cross this landscape during the Post-LGM ?

Anonymous said...

@Ric

http://www.donsmaps.com/icemaps.html

Ric Hern said...

@ epoch2013

Thanks.

Rob said...

Anthony's chapter in Black Sea flood book summarises it well

"The Khvalynian-era Caspian Sea separated the western steppes–those
west of the Ural Mountains–from the eastern steppes of Kazakhstan and Central
Asia for about 2000 years, from 16,000 to 14,000 BP. East-west travel on foot
was possible only across the Ural Mountains to the north. Earlier, during the cold
Last Glacial Maximum, the Upper Paleolithic cultures of the Siberian steppes
east of the Ural Mountains already were somewhat different from those west of
the Urals (Boriskovsky 1993; Lisitsyn 1996), but during the Khvalynian
flooding, the expanded sea cut off contact, isolating the cultures of the western
steppes even more from those of Kazakhstan and Central Asia. As a result, the
differences between them intensified. At the end of the Pleistocene, the spread
of forests across the former mammoth steppe created an ecological contrast
between the southern steppe cultures, which continued to hunt wild equids on the
open plains, and the cultures of the north, which slowly adapted to life as forest-
zone foragers.
After the Caspian retreated to approximately its modern basin (after
about 12,000 BP), the landewly exposed by the K hvalynian regression, the
North Caspian Depression, became a dry and challenging environment (Figure
1). A large sand desert called the Ryn Peski, or Red Sands (now the northern-
most sand desert in the world), lies north of the Caspian and northeast of the
Volga delta. "


There seems to have been no "South Ural refuge" either, there are like 3 sites there during the LUP, but maybe supernord knows something new ?

Ric Hern said...

@ Rob

Thanks

Rob said...

@ Matt

Thanks.
I guess we really need some samples from between 30 & 15 ky BP from east of Europe.

The EPiGRavettian network did stretch from Iberia to the Caucasus, so it could be a mediator of transmissions. As I 've said earlier in the thread, the Siberia route seems to come into play fairly late on - ~ 11 000 YBP arriving in Volga region with microblade techniques.
Until then west of the Ural region was repopulated from the west - southern aspect of the Russian plain, and some notable movements from the Caucasus to Stavropol region.

Its interesting the La Brana has that Eat Asian affinity, being so far southwest, whilst Villabruna lacks it.
How did Han affinity vary with affinity to MA-1 ?

Rob said...

RIP Vyacheslav Vsevolodovich Ivanov
Famous Indo Europeanist passed away

Davidski said...

RIP Vyacheslav Vsevolodovich Ivanov Famous Indo Europeanist passed away

Well that depends. Where was his favored urheimat?

Gioiello said...

Rob, I am sorry. Russians were always great linguists.

Rob said...

Dave
It's somewhat irrelevant. His contributions were important, even for the proponents of the Chalcolithic Tinder theory.

Slumbery said...

Davidski

He was in favor of the Armenian Highlands - Lake Urmia region, according to some biography I have just found on the internet.

Ric Herm
There is an article about Caspian and Black Sea sediments that I liked. Good starting point if you want to dig into the details, because in contains a lot of references.
http://www.sciencedirect.com/science/article/pii/S027737911630227X

It says that the level of the Caspian Sea was probably higher than now in the entire 50 000 ya (give or take) to end of Pleistocene period. They show that in the 19 000 ya - 13 800 ya period there was a considerable glacial meltwater inflow to the Caspian. It stopped around 13 800 ya, because the ice retreated outside of the Volga catchment area. They could not say anything for sure about before 19 kya, because their sediment samples were not old enough for that.

Rob said...

@ Slumberry

Before 19 ky BP, at the peak LGM, whatever rivers would have been frozen over and Caspian-Aral seas very low, making crossing across it possible, we can presume.
But this was definitely cut off after the LGM, until fairly late.
The Mesolithic groups which re-colonized the north Caspian region show no links to Siberian foragers across the Urals.

Anonymous said...

Rob said...
"The Mesolithic groups which re-colonized the north Caspian region show no links to Siberian foragers across the Urals."

It is a mistake, because no Mesolithic groups from the north Caspian region were tested! Samara EHG has link to Siberian Mal'ta/Afontova Gora men.

Ric Hern said...

@ Slumbery

Thanks.

Rob said...

@ Supernord

"It is a mistake, because no Mesolithic groups from the north Caspian region were tested! Samara EHG has link to Siberian Mal'ta/Afontova Gora men.'

It's funny how confident you are with yourself, given that you often fail to grasp the nuanced details.

The early Mesolithic "After the Caspian retreated to approximately its modern basin (after
about 12,000 BP), the lannde wly exposed by the K hvalynian regression, the
North Caspian Depression, became a dry and challenging environment (Figure
1). A large sand desert called the Ryn Peski, or Red Sands (now the northernmost
sand desert in the world), lies north of the Caspian and northeast of the
Volga delta. During humid periods, the Ryn Peski was covered by steppe, and
in dry periods, as today, it reverted to desert. The Caspian Depression elsewhere
had a floor of variegated clay anndd sya soils, dotted with brackish lakes and
covered by dry steppe containing salt-tolerant stands of Artemisia. Occasionally,
winds exposed fossil shoals of oDldre issena shells. Herds of saiga antelope
(Saiga tatarica), onagers E(quus hemionus ), and horses (Equus caballus) were
hunted across these saline plains by small bands of post-glacial hunters. Their
camps have been found among the dunes northeast of the Volga at places such
as the Early Mesolithic site of Je-Kalgan (Dzhe-Kalgan) and the Late Mesolithic
site of Suek-Te. Similar flint tool kits, containing geometric microliths in lunate
and trapezoidal shapes as well as end-scrsa poenr small blades, were used in sites
south of the Volga, as at Kharba. Igor Vasiliev (Vasiliev et al. 1996; Vasiliev
1998) has compared these tool traditions to those of the North Caucasus (at
Mesolithic sites such as Satanai and Tomuxlovka) and has suggested that the
Caspian steppes were re-occupied after the Khvalynian flood by forager groups
from the south. It is probably unwise to rely solely on lithic tool kits to identify
the ethnic and geographic origins of forager societies, but on the strength of the
evidence, similar microlithic tool kits were used during the Mesolithic over a
large region that included both the Caspian Depression and the North Caucasus.
"

by contrast "Earlier, during the cold Last Glacial Maximum, the Upper Paleolithic cultures of the Siberian steppes east of the Ural Mountains already were somewhat different from those west of the Urals (Boriskovsky 1993; Lisitsyn 1996), but during the Khvalynian
flooding, the expanded sea cut off contact, isolating the cultures of the western
steppes even more from those of Kazakhstan and Central Asia. As a result, the
differences between them intensified." & again "In contrast, Mesolithic foragers living east of the Caspian Sea and east of the Urals made quite different kinds of tools and
seem to have belonged to distinct social networks"

'Samara EHG', dates to 5500 BC, which is Late Mesolithic, making it some 3, 500 years after the appearance of Afantov-Gora microblade industries in northeastern Europe, which accounts for the ANE admixture seen in 'EHG".

Thus the earliest foragers (pre-9000 BC) probably lacked the ANE signal.

Ric Hern said...

Well I think Mammoths were hunted with Heavy Spears and Javelins so maybe the Migrants from Siberia adopted the Lighter weapons on their way to the West past the Southern Urals....

Anonymous said...

@Rob, your categorically allegations are unreasonable ("The Mesolithic groups which re-colonized the north Caspian region show no links to Siberian foragers across the Urals."). You just wrote Mesolithic, but not early Mesolithic. No genetic data on early (and too late) Mesolithic of the Northern Caspian Sea, the Urals, Siberia is present, therefore your assertions are unfounded. Any personal opinions not related to the distribution of people don't have weight. Just do not understand archaeology may think that the archaeological assumptions have in this issue big weight.
I wrote you that your declaration that "groups" "show" "no links" is mistaken. They may have a southern connections, but it does not means that they don't have еastern connections. Nobody knows.

by contrast "Earlier, during the cold Last Glacial Maximum, the Upper Paleolithic cultures of the Siberian steppes east of the Ural Mountains already were somewhat different from those west of the Urals (Boriskovsky 1993; Lisitsyn 1996), but during the Khvalynian
flooding, the expanded sea cut off contact, isolating the cultures of the western
steppes even more from those of Kazakhstan and Central Asia. As a result, the
differences between them intensified."


Early Kvalynsk transgression 17000-10500 BP had been interrupted Eltonsk regression. Phases of transgression and regression of the Caspian sea were changed constantly every few thousand years. But even it doesn't mean anything because people still could pass freely in any time, no physical barriers for this was not. The Eppigravettian culture of Moldova started approximately 20000 BP.

jv said...

@ Samuel Andrews. Came across this in my research on MtDNA H6.”MtDNA H6a was involved in a late Upper Paleolithic expansion from the southern Caucasus or Near East.” Is there a possibility that H6a lived in the Zagros Zarzian Culture? Eupedia.com includes H6a in Mesolithic Ukrainian MtDNA.

jv said...

Are there any similarities to Upper Paleolithic Eastern Siberian & Native American lithic reduction techniques? Behind my house, in the woods, there is an chert quarry that the Archaic Native Americans utilized. ( the chert is super chalky and was “cooked” in Earth ovens. I have found 3 Earth ovens in the woods) The chert cores left by these Native Americans resembles a long rectangular core with 4 distinct sides. I was wondering if this lithic reduction technique was used in Upper Paleolithic Eastern Siberian sites.

Anonymous said...

@supernord

Om anthrogenica some one connected the Kapova Cave to ANE or EHG survival during LGM. We know that EHG is roughly half ANE and half Villabruna. I have a hard time finding anything on Kapova cavem but we know (Epi-)Gravettian in Moldova came from the west IIRC.

Any thoughts or information?

Rob said...

Even mobile hunter-gatherers were tethered to their ecological niches. They didn´t just run around chasing mammoths, as some of you guys comically claim, & end up getting ''lost'' south Europe.

Rob said...

@ supernord

You need to stop obfuscating, because it severely undermines your relevance. And try not using words which you don.t understand the meaning of.

My ´´allegations´ are sourced materials from relevant specialists. I know what samples we have and don't have, the entire point of the discussion was hypothesizing based on archaeology. IF we had aDNA, we wouldnt be having the debate. So, no need for irrelvant side-points.


'But even it doesn't mean anything because people still could pass freely in any time, no physical barriers for this was not.'

Yes i did note that point above!. But the archaeological picture is nevertheless one of increasing divergence between the tehcnocomplexes on either side of the Urals.


So here it is again '"Earlier, during the cold Last Glacial Maximum, the Upper Paleolithic cultures of the Siberian steppes east of the Ural Mountains already were somewhat different from those west of the Urals (Boriskovsky 1993; Lisitsyn 1996), but during the Khvalynian
flooding, the expanded sea cut off contact, isolating the cultures of the western
steppes even more from those of Kazakhstan and Central Asia. As a result, the
differences between them intensified." & again "In contrast, Mesolithic foragers living east of the Caspian Sea and east of the Urals made quite different kinds of tools and
seem to have belonged to distinct social networks"

I presented literature, if you want to be taken seriously, do the same.


Here's more ´´The Trans-Urals and apparently even the north Siberian Arctic remained ice-free even during the LGM (Astakhov, 1992) and except for water-saturated and frozen boggy lands (Fig. 19) there were no major natural obstacles for movement of Palaeolithic groups further east into NW Siberia prior and after that time interval´´ http://www.sciencedirect.com/science/article/pii/S104061820900278X

Again, I note that diretion must have been bidirectional, but the main Afantova impact came the Terminal Palaeolithic.

If you have literature showing something else or new - please share it.

'The Eppigravettian culture of Moldova started approximately 20000 BP.´
Yes, and ?

Anonymous said...

@David

Are you checking these samples yet?

Anonymous said...

@epoch2013, In the LGM age in the Urals and behind it were boreal forests such as in the Balkans and in the South of Spain. Behind the Urals at that time was not of the glacier in contrast to Europe. In contrast to Western and Central Europe where there were glaciers and tundra, there was widespread steppe. Sungiric so generally lived near the glacier.


@Rob, I think you do not understand the texts referenced, confusing times and it is not clear that, so the answer something impossible.

Samuel Andrews said...

@jv,
"Is there a possibility that H6a lived in the Zagros Zarzian Culture? Eupedia.com includes H6a in Mesolithic Ukrainian MtDNA"

No H6a was found in Ukraine Mesolithic, I don't know why Eupedia claims so. I guess it's possible H6a lived in Zagros Zarzian culture. There's really no way to tell. My opinon on H6 han't changed. I've gotten more European mitogenomes and it keeps telling the same story: H6 in Europe is uniform and is the result of a "recent" expansion from the same source (Steppe).

jv said...

Thank you!

Ryan said...

Rob - I'm not sure what you are trying to say with the tethered thing. If your ecological niche covers a wide area why would you be stuck to one corner of it?

Grey said...

Rob said...
"Before 19 ky BP, at the peak LGM, whatever rivers would have been frozen over and Caspian-Aral seas very low, making crossing across it possible, we can presume."

might depend on weight?

Aram said...

Sunghir samples are uploaded to Yfull.
Soon we will learn their exact position.
https://www.yfull.com/tree/C-F3393/

Btw the first Armenian C1a2 is also uploaded. As expected it is in basal position toward Europeans. This could mean that current C1a2-s are of farmer origin rather than HG.

Ryan said...

@Rob - "Before 19 ky BP, at the peak LGM, whatever rivers would have been frozen over and Caspian-Aral seas very low, making crossing across it possible, we can presume."

From what I've read the Caspian and Aral seas were actually very high during the LGM. There's cave art of people hunting whales and dolphins in the Caspian so it can't ever have been too small either.

Rob said...

@ ryan

"I'm not sure what you are trying to say with the tethered thing. If your ecological niche covers a wide area why would you be stuck to one corner of it?"

Which is why, IMO, ANE probably first came via the boreal zone of Eastern Europe- the middle
Volga region.
It's quite easy to trace: microblades develop in TransBaikalia 20-18ky BP, a new, slightly different population (AG-like vs MA-1 like) with a distinctive material culture re-emerges from the refuge, reaches western Siberia by 16-14 ky BP, and then onto Eastern Europe after a couple G years delay.

They have nothing to do with EpiGravettian mammoth hunters with mammoth bone houses, which is an extinct culture

Ric Hern said...

What if MA-1 predecessors migrated from further West and more R* popped up near the Urals or just East of the Volga River ?

Rob said...

@Ric

Which sites Ric ? Did the Malta -Buyret culture originate in the Urals ?

Ric Hern said...

@ Rob

By 14 000 years ago R1b was already near Italy as you already know.So how could their ancestors only be in Western Siberia by 14 000 ? Or are you referring to some other Haplogroup ?

Ric Hern said...

@ Rob

So they migrated earlier....

Anonymous said...

@Rob
"Its interesting the La Brana has that Eat Asian affinity, being so far southwest, whilst Villabruna lacks it."

The Iberian refugee had Solutrean culture, distinct from the Epigravetian.
I guess a barrier of sort prevented cultural and also genetical influence between the two refugees during the LGM.

What would you think about a group of ANEs, living an Eskimo-like way on the sea edge of the glacier and spreading over it up to Iberia, bring the East Asian affinity, the Clovis-like blades, and maybe also R1b to there?

Rob said...

@ Ric

Right on - VB, and its haplogroups, could not come from Siberian Afantova Gora like groups.
But it could come from the meta-population which migrated to south Europe, Siberia , & Near East b/w 35 & 25 kya .

Ric Hern said...

@ Rob

Migrate from where ? Where was the epicentre of this meta-population ?

Ryan said...

@Rob - "They have nothing to do with EpiGravettian mammoth hunters with mammoth bone houses, which is an extinct culture"

Yes, I'm sure people who carved pictures of mammoths onto mammoth tusks had nothing to do with mammoths. /s

Matt said...

@kony numbers

What would you think about a group of ANEs, living an Eskimo-like way on the sea edge of the glacier and spreading over it up to Iberia, bring the East Asian affinity, the Clovis-like blades, and maybe also R1b to there?

In the context of Rob's comment ("Its interesting the La Brana has that East Asian affinity, being so far southwest, whilst Villabruna lacks it"), there's not so much sign of higher ANE in La Brana relative to VB though, is a drawback.

I was toying with the idea that, patterns in VB cluster could be explained by:

ENA patterns could be a combination of the

- GoyetQ116-1 ancestry with Tianyuan path of shared drift (seems + East Asian + Oceanian; -Native American, -Near East... based on abstract of Fu's latest on Tianyuan Man)

- EHG (+Native American, East Asia, Oceanian; - Near East)

- Villabruna's main UP descended ancestry (+ Near East; -East Asia, -Oceania, -Native American)

On the negative about that is that Mathieson's 2017 qpAdm puts about 16% EHG (8-10% ANE?) into Koros_HG (Hungarian KO1), only (https://imgur.com/7at24HX / https://imgur.com/mFjBBT1), and that is in theory seems like it would be the most extreme population on the WHG->EHG cline - https://imgur.com/M4TkAmV (from Lipson 2017) of those surveyed by Fu et al 2016.

(Though in a separate model Lipson 2017 tree models this same populations, KO1, as 85% Villabruna + 15% Anatolian - https://imgur.com/FM8eT8g).

Might have been easier to understand if some of the El Miron cluster had higher quality, and we could see if sustained raised relatednes to ENA is typical of the cluster.

Rob said...

@ Ryan

LOL. When you describe it like that, it does sound crazy.
But hear me out: The RUssian plain mammoth hunters which c. 16 kyBP could have arrived from further west, Moldova and even Moravia, Lower Austria, etc, literally chasing the last of the mammoths, east.
So it could be that the Siberian hunters were potentially distinct, although Of course they'd have some kind of contact.
Interestingly, the faunal assemblages in Siberia during the Late Palaeolithic were dominated by reindeer, Siberian mountain goat, bison, red deer, etc.
Only few sites had mammoths, like Afantova Gora.

@ Ric
Still need more aDNA, but there were probably a chain of movements in Late Palaeolithic, and there'd be multiple vectors for spread of specific lineages. Maybe R1 was near the east Casian during the LGM ?

Davidski said...

The idea that R1 came from LGM Central Asia was forced by some assumptions that have been proved erroneous by ancient DNA. For instance, that R1a was native to South Asia and that R1b-V88 was native to the Near East.

We now know that these lineages are native to Europe, so R1 could well be native to Europe as well, having arrived from Siberia as pre-R1 in Eastern Europe sparsely populated by Kostenki/Sunghir-related groups.

Rob said...

@ Dave
No, I was not basing my guess on studies from 2004, if your reply was in regard to that.
More to the possibility of now a rich biomass zones whose sites were destroyed by sea level changes, etc
But sure a pre-R1 scenario makes sense

Chad said...

Villabruna isn't UDG treated, unlike the others. So, it gets statistical noise away from Crown Eurasians. Nothing major and worth discussing. All WHG has an ENA affinity.

Ryan said...

@David - "We now know that these lineages are native to Europe, so R1 could well be native to Europe as well, having arrived from Siberia as pre-R1 in Eastern Europe sparsely populated by Kostenki/Sunghir-related groups."

I think it would be a bit academic to call it pre-R1 versus R1* since it'd just be a question of what SNPs are used to define R1 going forward if we find ancient offshoots from this branch.

@Chad - "Villabruna isn't UDG treated, unlike the others. So, it gets statistical noise away from Crown Eurasians. Nothing major and worth discussing. All WHG has an ENA affinity."

Both ENA and basal, right? At least in the case of Villabruna.

@Rob - I think big game hunters can be pretty mobile still. Take the indigenous peoples of the Great Plains in North America for example. I don't see why their relatives on the Eurasian steppe would be very different.

Ric Hern said...

@ Ryan

Yes Alaskan Caribou travel as much as 5000 kms during their migration and some African desert elephants migrate up to 500 kms.

So huge distances could have been covered by Hunter Gatherers if they followed the Herds and it is not difficult to imagine that herds would have shifted their migration path if conditions gradually deteriorated in their original range.

Ric Hern said...

@ Rob

I only imagine living in a world where there are only an estimated 5 Million people. It was not as if they trampled each others toes back then....

Rob said...

I have no doubt they were mobile and chased game around. I'm just pointing out that the only thing we appear to see in Western Russia at this point (25-16ky BP), is migration from the west and south.
So instead of carrying on like a broken record about game hunters, why don't we reflect on what the evidence is actually showing ?

Ric Hern said...

@ Rob

As you know there were huge changes going on on Earth and people back then did not take into consideration where they should build or camp just for us to find their remains easily.

So I think a bit of constructive speculation is not really the end of the world.

And yes I have been reading many books and papers for more than 30 years about this and some I even forgot about till someone mentioned something to trigger my memory....

Matt said...

Re: ENA patterns in VB cluster, current published material by lab (which sampled these ancients and invented the techniques):

Villabruna cluster shows variability in relationship ENA, not driven just by Villabruna sample alone. (Ranchot88, Rochedane).

Villabruna shows different relatedness to ENA even when restricting to transversion sites, not artefact of UDG treatment.

Relevant quotes: https://imgur.com/a/qNj7M

(If publishing by Fu, etc. changes, perhaps I will change my mind...)

Rob said...

@ Ric
Okay. Maybe you can scrounge around for some eastern / Siberian industries appearing c. 20'kya , because that would fit the bill for R1.
Because I'm not doubting the possibility, even likelihood, but I can't recall ever reading about such a phenomenon.

@ Matt
Thanks

Ric Hern said...

@ Rob

I think a lot of things happended near the Altai Mountains. I see a possibility of Haplogroup R forming somewhere between Ust Ishim and the Carpathians.

If K2a* could be in that area I can not see why K2b* and their later relatives could not have been there to. We see Denisovans, Neanderthal and Modern Humans all in the Central Asian Altai melting pot.

We know that Neanderthals which contributed to the Modern Population were of West Eurasian stock so there had to be a migration from the Carpathians/Crimea/Urals or the Caucasus/Zagros to where Ust Ishim Man eventually ended up. I think the Neanderthal admixture could be a clue from where the ancestors of Mal'ta Buret
originated from. So those women figurines could still be connected to Europe after all....

Ryan said...

@Ric - "We know that Neanderthals which contributed to the Modern Population were of West Eurasian stock so there had to be a migration from the Carpathians/Crimea/Urals or the Caucasus/Zagros to where Ust Ishim Man eventually ended up." Huh?

Ric Hern said...

@ Ryan

Well as far as I understood it Altai Neanderthals did not contribute genetically to Modern Humans....

Grey said...

Ryan
" I think big game hunters can be pretty mobile still. Take the indigenous peoples of the Great Plains in North America for example. I don't see why their relatives on the Eurasian steppe would be very different."

sure but limited by the extent of the biome. if hunters followed herds of x they'll remain within the bounds of the biome that supports x.

(seems to me the idea of bands following herds 1000s of mile might explain how a more or less homogeneous population could be maintained over a huge area i.e. if different bands swapped brides when they bumped into each other on their travels.)

Ric Hern said...

If all West Eurasian ancestors migrated from Southeast Asia how did they dodge Denisovan admixture ?

Tesmos said...

We really need Adna from Southeast and/or East Asia etc. it's already proven that modern diversity and frequencies can be misleading about it's origin and expansion like the R1a situation in India/Iran.

Slumbery said...

Ric Hern

1. I pretty sure nobody here arguing for _all_ West Eurasians migrating from SE Asia. There is an argument for an ancient migration that had significant impact on male lineages, but even that is only for branches under K. I am yet to see any argument for G, I, J or E coming from SE Asia. And the genome-wide impact of K branches is not necessarily as big as the Y-DNA impact.

2. The assumed SE Asian source population for the K lineages being Denisovan admixed is not self evident at all. Most East Asian have a lot smaller Denisovan ancestry than Neanderthal one and populations from the most likely source area, continental S-SE Asia, are not very well studied from this angle, as far as I know. So how did they dodge Denisovan admixture? The answer is that we do not have any information whether there was anything to dodge.

Ric Hern said...

@ Slumbery

It is just a bit strange for me that most K* and P* are found among people who have Denisovan and some other yet unknown humanin admixture and no trace exist of this in Haplogroups R*....

Ric Hern said...

@ Slumbery

So if I am not totally confused this tells me that Haplogroup K originated somewhere else than Southeast Asia and split into different migrations....? Maybe rather from the Caucasus or the Zagros Mountains...

Slumbery said...

Ric Hern

Why the Caucasus or the Zagros mountains? Why not Afghanistan, or Pakistan, or Mezopotamia, or India, etc.? What is the difference? This is just random guessing at this point. I could say India is a better candidate, because of its central position between IJ and K and also H is pretty India specific. But with the time passed since then this might mean nothing.

Also I still do not understand what is your problem with Denisovan admixture.
- We do not know where the Denosovans lived.
- We do not know where the Denisovan mixture happened and when.
- We do not know what later migrations carried around or diluted the Denisovan ancestry, and where and when.
- We do not know if there was any Denisovan ancestry in the part of SE Asia where and when the supposed P (or P precursor) population might have lived. (And the fact that we do not know where and when this population lived is not helping either...)

With this many "we do not know" the Denisovan ancestry vs. P ancestry is pretty much just a phantom, not a real problem.

MaxT said...

@Ric

MA1/ANE does have Denisova admixture as showed in Fu e al and Lipson et al 2017

"Denisova-related gene flow into MA1, with an inferred mixture proportion of 1.2%, or 1.0% Denisova-related ancestry"

Ryan said...

@Ric - Has anyone even checked Mal'ta for Denisovan admixture? I can't find anything.

That being said, Denisovan admixture isn't ubiquitous in SE Asia. The Onge seem to have managed to dodge it. Keep in mind that the indigenous populations in the Americas with the most Denisovan admixture have the least Oceanian admixture too. If that Oceanian ancestry came from ANE, then it lacking Denisovan admixture is exactly what we would expect.

The Neanderthals responsible for most of our Neanderthal ancestry probably just haven't been sampled yet, so I wouldn't read too much into which fringe population of Neanderthals is closest to us when all we have are samples from fringe populations.

If memory serves though, prior to the recent sequencing of this Croatian Neanderthal, Altai Neanderthals were the best match for the Neanderthal ancestry in all Eurasians, but Iberian Neanderthals were the best match for the Neanderthal ancestry specific to East Asians. So at the very least Neanderthal population structure wasn't very intuitive.

@Grey - The biome spanned Eurasia!

Ryan said...

K2 also probably split up before there was differentiation between different East Asian populations, or even much between East and West Eurasians.

Ric Hern said...

Thanks everyone.

Anonymous said...

@Slumbery

"Why the Caucasus or the Zagros mountains? Why not Afghanistan, or Pakistan, or Mezopotamia, or India, etc.? What is the difference?"

We may not have any good idea on where Denisovan lived. But we have a very good idea where Neanderthals lived. And *all* Eurasians have a considerable Neanderthal admixture proven to be from non-Altaic Neanderthals, and dated to 55k ya.

The Caucasus and Zagros mountains are inside known Neanderthal area, the rest isn't.

capra internetensis said...

@epoch

Do we actually know that Neanderthals didn't live in those places? There were Neanderthals in Iran and Uzbekistan and the Levant, so why not Afghanistan or Mesopotamia?

Though the obvious place would be West Asia.

Arza said...

@ Livestream of a Science of the Human Past talk

Prof. Reich

44:35 - Map of the spread of steppe ancestry and Beakers. I don't know how meaningful (or -less) this is, but "Yamnaya pastoralists arrive from the steppe ~5,000 ya" arrow points from Ukraine to the North-Western direction.

56:20 - Isotopic analysis of Beakers from Britain shows that less than 0.5% of people had non-local signature. Yet DNA shows a massive migration. How it's possible? Strontium isotope signature from the Netherlands and Northern France is virtually indistinguishable from the British one.

50:30 - Temporal data allow to better understand the dynamics of population change and they are moving in this direction. E.g. they have now a time-series of data from Hungary that shows continuous HG admixture in farmers.

Karl_K said...

@Slumberry and @epoch2013

"We do not know where the Denisovan mixture happened and when."

"But we have a very good idea where Neanderthals lived. And *all* Eurasians have a considerable Neanderthal admixture proven to be from non-Altaic Neanderthals, and dated to 55k ya."

We actually do have strong constraints on when and where most Denisovan admixture occurred. We know that eastern Neanderthals, but not western ones, had Denisovan admixture. And we know that the ancestors of the people who crossed the Wallace line had to have picked up their Denisovan component after the shared '55k ya' Neanderthal event in the west, but before crossing the water in the southeast about 50k ya.

We also know that some African-like 'modern humans' and some very divergent (erectus?) were around Asia at some time within the last 200,000 years, but before the major Out of Africa event.

And where were all the Basal Eurasian populations for so many years? I suspect in North Africa, away from Neanderthal contact, and only arrived in Eurasia in somewhat later OOA events that couldn't get past the fertile crescent without admixing with the locals hunter gatherers they encountered.

Davidski said...

@Arza

Can you screen cap the map and post a link here? I'm not able to watch the video right now.

Arza said...

https://s1.postimg.org/12vme0gjkf/Beaker_Phenomenon.jpg

If I understand correctly what he says, new article about interactions between HG and farmers is "in press".

Davidski said...

Yeah, he's referring to this Lipson et al. paper...

http://eurogenes.blogspot.com.au/2017/03/neolithic-europe-its-complicated-lipson.html

And about the arrow from Ukraine: to me it just looks like an arrow pointing generally from the PC steppe.

Chad said...

Denisova > MA1 flow may be an issue of using Altai instead of Vindija as the Neandertal. Altai has AMH admixture, so a bit of deeper archaic may be asked for. When I use Vi_merge, or combine Altai and Vi_merge, there is no Z >2 asking for Denisova ancestry in MA1.

Slumbery said...

epoch2013

Why should P lineages come from a place where Neanderthals lived? The bulk of the mixture probably happened earlier.

Slumbery said...

Karl_K
OK, but that still give a lot of room.

German Dziebel said...

Minimal presence of the main Amerindian component (RED) in one of the Sunghir individuals at 34,000 (https://genetiker.wordpress.com/). Tianyuan (40,000) has more. Kostenki even more. Yuzhnyi Olenij Ostrov (karelia) (7000), Okunev, Khvalynsk, etc. even more. Mal'ta has the most.

A secondary Amerindian (or Beringian) component (ORANGE, peaks in Eskimos and Paleosiberians) is also found in Sunghir individuals, and it's more pronounced than the main Amerindian component.

Matt said...

Re: the talk and Bronze Age Beaker Brits, not really much change at all in the broad picture (almost total replacement), but Reich's talk did show expanded sampling compared to Olalde's original preprint, and there are a few details from their extended panel of sampling:

https://imgur.com/a/f1kZZ

Series of images contrasting preprint with newer data.

Looks like this firms up complete replacement in early samples, with some "choppy" admixture, then a low level of Britain Neolithic in the later samples. But it is a very low level - 10% ish. (Could even reflect just more migration from Bell Beaker groups in e.g. France).

Of course, this is basically the same as preprint, but higher sample size firms up the trend.

jv said...

Thank you for the link! However, the male migration from the Steppe WAS NOT entirely male! MtDNA H6a1a and other “new” to Central & Northern Europe MtDNA lineages were introduced during these migrations.

Grey said...

@Ryan
"The biome spanned Eurasia!"

well seems to me that's the question - do the different mammoth clades on these maps

http://eurogenes.blogspot.co.uk/2016/05/following-mammoth-herds.html

imply significant barriers to mammoth travel and if so would that indirectly limit human populations who lived off them?

Grey said...

@Matt
"Looks like this firms up complete replacement in early samples, with some "choppy" admixture"

which would allow for the survival of some archaic genes in some regions which could explain this study

http://www.cell.com/ajhg/fulltext/S0002-9297(17)30379-8

"Arza: a time-series of data from Hungary that shows continuous HG admixture in farmers"

Ric Hern said...

Were there any human remains found in Northern Siberia that dates to roundabout the same time as the slaughter of that Mammoth 45 000 years ago ?

Slumbery said...

Ric Hern

Why do you specify "North" Siberia? Those mammoths were there is South Siberia too. And we have Ust-Ishtim exactly from the time you requested. (Not even that South, more like Central Siberia.)

Ryan said...

@Grey - that seems to show mammoths as migrating across Eurasia just fine. Keep in mind too that minor geographic barriers can create a fair amount of structure in populations. Take elephants for example - the European elephants that went extinct in Europe ~30,000 years ago were more closely related to African forest elephants are to African bush elephants. Their mtDNA actually falls within modern African forest elephant's mtDNA variation. That doesn't mean that it's harder for an elephant to walk from the savanna to the adjoining forest or visa versa than it is for a forest elephant to migrate all the way to Europe. It just means forest elephants happened to be the ones that made that migration.

klevius said...

Isn't there some sort of confusion re. AMH and what we used to call HSS? As I understand it we don't have a clue about AMH dna - only 45,000 bp HSS dna from Ust-Ishim.

Slumbery said...

@Ryan

It does not really effect your argument, but those European elephants were probably hybrids. It is just that the Asian elephant admixture was male mediated. Of course this still mean that some African forest elephants had to make that migration.

Grey said...

@Ryan
"that seems to show mammoths as migrating across Eurasia just fine"

sure, at a later date - hence the possibility of there being barriers which later disappeared and specific clades going extinct behind a barrier but that space getting refilled later with different mammoth from beyond the barrier.

Matt said...

Re; the part of Reich's talk on Bell Beaker, analysing it further to death:

Counting pixels on graphs on the slides (yes, I was that bored / impatient for updated paper), proportions of Britain Neolithic work out as:

Britain CA-EBA (Whole Period) - N=56 - Average: 7%, Median: 4.7%. Range: 0% - 39.7%.
Britain CA-EBA, Earliest (Sample 1-40) - N=40: Average: 5%, Median: 0%. Range: 0% - 39.7%.
Britain CA-EBA, Tail End (Sample 41-56) - N =16: Average 13%, Median: 11%. Range 6.4% - 35.4%.
Britain MBA - N = 22: Average: 10.8%, Median: 11%. Range: 0% - 20.9%.
Britain LBA - N = 5: Average 12.8%, Median: 13.2%. Range: 5.1% - 17.9%.

Comparing to the paper, we had:

Britain BC-EBA (Whole Period) - N=24 - Average: 5.5%, Median: 2.8%. Range: 0% - 38.2%.
Britain BC-EBA, Earliest (Sample 1-17) - N=17 - Average: 4.4%, Median: 0%. Range: 0% - 38.2%.
Britain BC-EBA, Tail End (Sample 18-26) - N=7 - Average: 8.3%, Median: 8.9%. Range: 3.9% - 13.7%.
Britain MBA - N=7 - Average: 6.8%, Median: 5.3%. Range: 0% - 14%.

The stated proportions of the paper were: "93±2% local population turnover by the Middle Bronze Age" (e.g. 7% British Neolithic).

So this does seem like a very slight increase, from just over 90% replacement to just under 90%.

Comparing Y in the slides to the above autosomal proportions:
Britain MBA: I2 - 8.9% vs Britain Neolithic - 10.8%

Assuming all I2 from Britain Neolithic, suggests slight sex bias with MBA having potentially 13% Britain Neolithic female ancestry and 8.9% male, which averages out at 10.8% autosome. But it's not really looking sex biased, mostly just low for both, esp. if we assume a bit more noise on the y (sample size only 11 and uniparentals are generally noisier).

....

Interesting to compare to Spain, which has survival of Iberian CA that is roughly 2.5-3.5 times higher (looks about 25%-35% SpainCA).

In Spain it looks like there is that early wave who have low steppe ancestry, followed by a general increase and also likely admixture from a eastern / southeastern vector.

Perhaps in Iberia, there was a difference that the first wave of migrants from the northern Beaker complex assimilated into a more shared culture which had already received cultural influence from them (e.g. used the same beaker pottery?), and then adopted proto-Basque Iberian speech and mythology?

Later on there was then another "top up" of steppe ancestry from Celtic migrations in the late Bronze Age?

While in Britain, only one wave with some low survival of British Neolithic people. Perhaps with a few ideas along the way - Razib Khan speculated that cthonic and fertility gods of NW European mythologies (Germanic, Insular Celtic) may derive from cultural transmission from Middle Neolithic farming peoples (though this is a really speculative) - and maybe some continued use of some of the British Neolithic sacral sites (the henge temples, the temple complex at Skara Brae, etc.)...

Matt said...

Just to reiterate about the y vs autosomal stuff I mentioned about, comparing to the actual sample size in ancients, that actually means that frequency of I2 in the slides was:

Britain CA-EBA: 2/39, or 5.6% Y against 4.8% autosome Britain Neolithic
Britain MBA: 1/11, or 9% Y against 10.8% autosome Britain Neolithic

The Y frequencies are not as granular as the autosome, so using these small comparisons to estimate sex bias is actually meaningless. Most likely there was just no sex bias in this particular situation.

Ryan said...

@Matt - "The Y frequencies are not as granular as the autosome, so using these small comparisons to estimate sex bias is actually meaningless. Most likely there was just no sex bias in this particular situation."

You can use the X-chromosome, which Mathieson did, and found a HUGE sex bias.

Anonymous said...

@Matt

" Razib Khan speculated that cthonic and fertility gods of NW European mythologies (Germanic, Insular Celtic) may derive from cultural transmission from Middle Neolithic farming peoples (though this is a really speculative)"

Vanir versus Aesir? It looks to me very attractive to tie the name "Aasir" to something related to "Aryan". Like the etymology of the name Ossetians is explained.

Anonymous said...

@Matt

Forget about that. It turns out its etymology is completely different. I should learn to google *before* responding :-)

https://en.wikipedia.org/wiki/%C3%86sir#Etymology

Matt said...

@Ryan, not for these particular populations Mathieson didn't.

If someone eventually looked at it, there could be a X bias in absence of any lower Y than autosome, somehow. It doesn't seem very likely though!

I still have questions about whether that method really works though. I'd rather see it demonstrated through simple f4 ratios and stat comparisons on X and autosome a la Charlston Chiang's paper on Sardinia, prior to these complicated qpAdm methods.

Jaap said...

Can't discuss lineages with you guys. One thing caught my attention though. 'Following the herds for 1000s of miles' is not within the realm of probability. That would be a really strenuous and one-sided existance. There's other things to do apart from running. Herds travel fast!
Hunter-gatherers have quite a full agenda along a range that is unlikely to supersede 100 miles. Crossing the trail of herds is important, but so is keeping in touch with the grapevine (mating!), attending the fruittrees and honeycombs and salmon-runs, etc etc. Losing touch with the smoke of the camps of your kin, and finding the tryst-places void of messages spells mortal danger for your group!
That said, cultures may spread very long distances within five or six generations when the numbers are swelling, and the lands are empty of 'others'. And to us that looks like the blink of an eye.
There is always Murphy's Law, and those that have survived knew how to circumvent adversity. I often speculate about the shaman's role here: paleological consciousness is so far apart from ours.
But it was obvious to me from the start there was something fishy about the Clovis-hypothesis. This was hardly a theory of well-meaning scientists, but obviously pushed by people with a very different agenda.
Let's face it: no group is gonna make it from Beringia to Vuelta del Fuego in 1000 years! They would have had to borrow motorcycles from the future to race down (non-existant) sandy beaches, all the while muttering: 'Can't keep Dillehay waiting, now can we?' Boats then! They could have made it in boats! Ah me! How many? Motivated by what? Why not stop off in sunny California and sprawl in the sun?
South America is much much older than the LGM, but North American academia simply wouldn't have it. Budgeting, I presume.
Academia has laws of its own, and they're unwritten, but understood by all that want a career in this field. Our David here understands that better than most. But he's also committed to the 'truth' (whatever that is!?). There are more agendas out there: crackpots who want to unmask crackpottery, hobby-horse riders, identity-(re-?)searchers, the Indo-Europeans (Indo versus European, so basically us-and -them guys), and searchers. By that last I refer to people for whom many things are still up in the air. They have a stake but are unsure about it. The searchers are all nice guys! There's also the observers, like me. They have no stake. They're probably cowards, looking for a bit of action when it is cheap. They won't be there when payment is required.
So here I am now with my finger on the post-button. Have I said anything meaningful in this post? Going from distance-covering to academia? Structure of a child's essay ... Well, what the heck ...

mike said...

I blame this site for the incredible amount of times I have opened a site on bull semen.

Ric Hern said...

@ Slumbery

Because that is what the link said which David posted. A kill site in Northern Siberia specifically that dates to 45 000 years ago.

rozenblatt said...

Note: the claim about mammoth kill site in Northen Siberia 45 000 years ago was disputed at another paper:

http://www.sciencedirect.com/science/article/pii/S1040618216302105

" The few minor traces of carnivore scavenging, the little disturbed condition of the carcass, and the absence of bone modifications made by human actions, along with the social status of this young male animal, are interpreted here as highly probable evidence that the Zhenya Mammoth died from unrecoverable injuries inflicted during a bull-to-bull fight."

Ric Hern said...

@ rozenfag

Thanks.

Speculation : However I can see how some human interaction could have been present without leaving a significant trace.

Eg. A small band of hunters encountering two fighting bulls or an injured bull which could have triggered their opportunistic instinct. All they maybe needed to do was cut the trunk artery to secure their prize. They could have been on the hunt for something else and didn't expect or prepare for this scenario and could have cut only a few choice pieces of meat because of a significant gathering of carnivores.....

Ric Hern said...

@ rozenfag

I have seen here by us how people in a hurry operate....

They kill a cow and cut all the choice pieces without disturbing the layout of the carcass, most of the times when the cow is still alive with only the lowerleg tendons cut.....

Ryan said...

@Matt - Yah, I'm just saying it could be done. And it'd be nice to do both approaches to compare them.

@Chad - "Denisova > MA1 flow may be an issue of using Altai instead of Vindija as the Neandertal. Altai has AMH admixture, so a bit of deeper archaic may be asked for. When I use Vi_merge, or combine Altai and Vi_merge, there is no Z >2 asking for Denisova ancestry in MA1."

Vindija's DNA has only be out for a week though hasn't it. If you've done any analyses using her DNA I'd be curious if anything interesting jumped out at you.

Sofia Aurora said...

@Matt

A good point there Matt!
Thanks

Unknown said...

>epoch2013, if you go by the model Ancient North Eurasians being Sunghir+ENA (which seems worth testing eventually in qpGraph, though seems hard to know how anyone can know without testing), then it may not be a single small population and it may just be than all these populations are varying independent mixes of the Basal_Eurasian+Villabruna+Sunghir+ENA ancestries. I am not sure about this though.

While ANE was of course admixed it seems to have a different type of West Eurasian ancestry than the Vestonice Cluster, and the Villabruna Cluster did. So they must have had split very early. Pre-Kostenki14 is a good bet. Due to Bronze Age invasions Modern Europeans seem to have more of that ANE-type West Eurasian admixture than Villabruna/Vestonice types of ancestry. It's indeed possible Paleolithic Europeans were admixed with other Eurasians, but it's also could be that all of these paleolithic pre-LGM Eurasians shared a significant part of their alleles because their divergence dates weren't that distant, yet.
That may be the reason why ADMIXTURE results for UP Europeans are so mixed, yet QpGraph suggests they in fact weren't admixed with other populations.(Though Sunghir is supposed to be admixed with Proto-WHG)
However gpGraph can create many different models so it's impossible to tell without Paleolithic and Mesolithic samples from East Asia, South Asia/Oceania, and West Asia, along with pure Paleolithic and Mesolithic Basal Eurasian samples. It's possible all Paleolithic Eurasians were initially Ust'-Ishim/Oase/ASI-like.
For one, some formal statistic tests do suggest that GoyetQ116-1 was admixed with something closely related to Tianyuan, and Tianyuan is supposed to be the ancestor of mongoloids. So other UP Europeans could also had>epoch2013, if you go by the model Ancient North Eurasians being Sunghir+ENA (which seems worth testing eventually in qpGraph, though seems hard to know how anyone can know without testing), then it may not be a single small population and it may just be than all these populations are varying independent mixes of the Basal_Eurasian+Villabruna+Sunghir+ENA ancestries. I am not sure about this though.

While ANE was of course admixed it seems to have a different type of West Eurasian ancestry than the Vestonice Cluster, and the Villabruna Cluster did. So they must have had split very early. Pre-Kostenki14 is a good bet. Due to Bronze Age invasions Modern Europeans seem to have more of that ANE-type West Eurasian admixture than Villabruna/Vestonice types of ancestry. It's indeed possible Paleolithic Europeans were admixed with other Eurasians, but it's also could be that all of these paleolithic pre-LGM Eurasians shared a significant part of their alleles because their divergence dates weren't that distant, yet.
That may be the reason why ADMIXTURE results for UP Europeans are so mixed, yet QpGraph suggests they in fact weren't admixed with other populations.(Though Sunghir is supposed to be admixed with Proto-WHG)
However gpGraph can create many different models so it's impossible to tell without Paleolithic and Mesolithic samples from East Asia, South Asia/Oceania, and West Asia, along with pure Paleolithic and Mesolithic Basal Eurasian samples. It's possible all Paleolithic Eurasians were initially Ust'-Ishim/Oase/ASI-like.
For one, some formal statistic tests do suggest that GoyetQ116-1 was admixed with something closely related to Tianyuan, and Tianyuan is supposed to be the ancestor of mongoloids. So other UP Europeans could also had been admixed with many different things from other continents to varying extents.

Unknown said...


This statement supports that.

>We demonstrate that individuals at Sunghir derive from a population of small effective size, with limited kinship and levels of inbreeding similar to HG populations. Our findings suggest that UP social organization was similar to that of living HGs, with limited relatedness within residential groups embedded in a larger mating network.

Then, the question is how it is posssible that WHGs were so genetically uniform and minimally admixed?
Maybe the LGM bottleneck made them less diverse?

https://imgur.com/a/OAInt

https://genetiker.wordpress.com/