search this blog

Tuesday, September 13, 2016

Paleo-Eskimos as founders of Na-Dene Amerindians

Very nice new perprint at bioRxiv:

Abstract: Prehistory of Native Americans of the Na-Dene language family remains controversial. Genetic continuity of Paleo-Eskimos (Saqqaq and Dorset cultures) and Na-Dene was proposed under the three-wave model of America's settlement; however, recent studies have produced conflicting results. Here, we performed reconstruction and dating of Na-Dene population history, using genome sequencing data and a coalescent method relying on rare alleles (Rarecoal). We also applied model-free approaches for analysis of rare allele and autosomal haplotype sharing. All methods detected Central and West Siberian ancestry exclusively in a fraction of modern day Na-Dene individuals, but not in other Native Americans. Our results are consistent with gene flow from Paleo-Eskimos into the First American ancestors of Na-Dene, and a later less extensive bidirectional admixture between Na-Dene and Neo-Eskimos. The dated gene flow from Siberia to Na-Dene is in agreement with the Dene-Yeniseian language macrofamily proposal and with the succession of archaeological cultures in Siberia.

Flegontov et al., Na-Dene populations descend from the Paleo-Eskimo migration into America, bioRxiv, Posted September 13, 2016, doi:

See also...

Eurasians: An idiot's guide


Karl_K said...

It's about time someone decided to look at this with some real hard data. It has been strongly hinted at for decades.

andrew said...

Great catch relevant to one of the biggest unsolved problems in modeling the pre-Columbian settlement of the Americas.

Kristiina said...

This article "DNA Analysis Reveals All Arctic Eskimos Migrated From Alaska's North Slope" is also interesting:

"This is the first evidence that genetically ties all of the Iñupiat and Inuit populations from Alaska, Canada and Greenland back to the Alaskan North Slope," said Northwestern's M. Geoffrey Hayes, senior author of the new study to be published April 29, 2015, in the American Journal of Physical Anthropology.

The new findings support the archaeological model that determines people arrived in the North Slip in an eastward migration from Alaska to Greenland. It also supports the hypothesis that two major migrations occurred from the North Slope to the east.

"There has never been a clear biological link found in the DNA of the Paleo-Eskimos, the first people to spread from Alaska into the eastern North American arctic, and the DNA of Neo-Eskimos, a more technologically sophisticated group that later spread very quickly from Alaska and the Bering Strait region to Greenland and seemed to replace the Paleo-Eskimo," Hayes said. "Our study suggests that the Alaskan North Slope serves as the homeland for both of those groups, during two different migrations. We found DNA haplogroups of both ancient Paleo-Eskimos and Neo-Eskimos in Iñupiat people living in the North Slope today."

The research team took saliva samples from 151 volunteers living in eight different North Slope communities. They sequenced and analyzed the mitochondrial DNA present, which is passed from mother to child. The analysis revealed 98 percent of the maternal lineages in this group were of Arctic descent, and contained all know haplogroups which included A2a, A2b, D4b1a and D2.

Until this study, D2 had only been found in ancient Plaeo-Eskimos. D4b1a is a known haplogroup of the ancient Neo-Eskimo that came after the Paleo-Eskimos and is believed to have replaced them in most of the Arctic.

"We think the presence of these two haplotypes in villages of the North Slope means that the Paleo-Eskimos and the Neo-Eskimos were both ancestors of the contemporary Iñupiat people," said Jennifer A. Raff, first author of the study and a post-doctoral fellow in Hayes' lab at the Feinberg School when the research was being done. "We will be exploring these connections in the future with additional genetic markers."

Kristiina said...

The article is available here but it is not free:


Mitochondrial lineages from the North Slope villages had a low frequency (2%) of non-Arctic maternal admixture, and all haplogroups (A2, A2a, A2b, D2a, and D4b1a–formerly known as D3) found in previously sequenced Neo- and Paleo-Eskimos and living Inuit and Eskimo peoples from across the North American Arctic. Lineages basal for each haplogroup were present in the North Slope. We also found the first occurrence of two haplogroups in contemporary North American Arctic populations: D2a, previously identified only in Aleuts and Paleo-Eskimos, and the pan-American C4.


Our results yield insight into the maternal population history of the Alaskan North Slope and support the hypothesis that this region served as an ancestral pool for eastward movements to Canada and Greenland, for both the Paleo-Eskimo and Neo-Eskimo populations.

An article on Indian Country website claims that "The origins of the Eskimos and Aleuts are, like the origins of American Indians, shrouded in mystery and the source of much controversy. According to the modern scientific consensus, the progenitors of the Paleo-Eskimos and Aleuts settled in Alaska sometime about 5,500 years ago from Siberia, although this date is now being disputed and there is little evidence that suggests they were originally from Siberia. Up until recently it was presumed that Greenland had been settled by Paleo-Eskimos only about 2,500 years ago, but new archaeological discoveries have shown Paleo-Eskimos have inhabited Greenland for at least 4,500 years and maybe much more.

This new research on Inupiat DNA has provided strong evidence that the earliest Inuit peoples, as well as the Paleo-Eskimos, came from Alaska. Where they came from before then is anyone’s guess."

Karl_K said...

"Where they came from before then is anyone’s guess."

I am gonna go out on a limb here and suggest they came from North East Asia.

andrew said...

@KarlK. Agreed. I wouldn't put big $$$ on the Eskimos came from Australia or Eskimos came from the Congo theories. ;)

terryt said...

"I am gonna go out on a limb here and suggest they came from North East Asia".

On what possible grounds do you base that claim on? Haha.

Kristiina said...

“I am gonna go out on a limb here and suggest they came from North East Asia.”

With all probability yes, but the timeframe is of interest here. Were Eskimo-Aleut languages spoken on the Aleutian Islands or somewhere else in the northwest America already before the Na Dene languages came from Siberia. If not, what is the culture that brought them over the Bering Strait? Is Q-NWT01 the yDNA that is specifically related to Eskimo-Aleut languages? If it is, then what is the relation of Inuit Q-NWT01 with Chinese Q-M120? Q-M120 is said to be recent in Southeast Asia. If Inuits are a more recent population than paleo-Eskimos, why in the admixture analysis, Saqqaq shows recent Siberian ancestry but Inuits do not. Autosomally Inuits look very ancient. They have two components: c. 20% Amerind and 80% Beringian/Alaskan ancestry which they share, by the way, with Ma1 24 000 years back.

If you can provide answers to these questions, it would be interesting.

FrankN said...

From the thread-opening study:
"Substantial admixture of 22.3 – 23.8% from Siberians (22 genomes) into Northern Athabaskans was revealed in our model, with only 6.5 – 7% in the opposite direction (95% confidence intervals are given). The SiberianAthabaskan admixture edge was dated at 6,575 – 7,030 YBP (Table 1). (..)
Our results are consistent with a gene flow from the Saqqaq Paleo-Eskimos (19 – 25% admixture ratio) exclusively into the First American ancestors of Na-Dene, and a much later and less extensive bidirectional gene flow was detected between the Na-Dene and Neo-Eskimo branches. (..)
The new wave of population from northeastern Asia that arrived in Alaska at least 4,800 years ago displays clear archaeological precedents leading back to Central Siberia. The rise of the Syalakh culture that flourished across much of Northeastern Siberia between 6,500 and 5,200 YBP involved migrants from the Transbaikal area who possibly mixed with local remnants of the earlier Sumnagin culture (10,500-6,500 BP), bringing the bow and arrow and new types of pottery to Northeastern Siberia. As the Bel’kachi culture (5,200-4,100 YBP) developed from Syalakh along the Lena and Aldan rivers, at least one group of these people might have crossed the Bering Strait into Alaska around 4,800 YBP, giving rise to Paleo-Eskimos. Thus, the Syalakh culture peoples, spreading across Siberia after 6,500 YBP, might represent the “ghost population” that split off around 6,500-7,000 YBP and later gave rise to migrants into America.
The geographic connection between Paleo-Eskimos and the related Siberian groups probably became severed as subsequent waves of hunter-gatherers entering Eastern Siberia from the west during the Late Neolithic (Ymyakhtakh culture, 3,700-2,800 YBP) brought new cultures and new language groups. This phase of North Asian prehistory most likely involved the spread of Yukaghir, Chukchi-Kamchatkan and Eskimo-Aleut languages, whose presence in the extreme northeast of Asia intervened geographically between Paleo-Eskimos, Na-Dene, and their Old World cousins. Notably, the dates of the Siberian-Arctic split obtained under our model (~4,000-4,200 YBP, Table 2) also agree with this scenario that links the spread of the Ymyakhtakh culture (after 3,700 YBP) with the Arctic meta-population, i.e. ancestors of modern Chukchi-Kamchatkan and Eskimo-Aleut ethnic groups.

I guess that answers some of your questions, Kristina. You might also want to look into Suppl. Text 3, where E.Vajda discusses the Dene-Yeniseyan hypothesis in´the light of the genetic findings.

Kristiina said...

Frank, I still think that there are many unresolved questions. Yukaghiric languages (split intransitive alignment system), Chukchi-Kamchatkan languages (ergative-absolutive, polysynthetic structure) and Eskimo-Aleut languages (ergative absolutive pattern, a structure based on root+postbase) are different language families with different lexicon and not very close to each other, and it is audacious to put them all together under the same migration.

Ymyakhtakh culture has been linked with Yukaghir, and it is sloppy thinking to link Ymyakhtakh indiscriminately with all three. It is a pity that we do not know the Yukaghir Q yDNA line more specifically. If it is the same as Inuit Q-NWT01, it is an indication of a connection.

The autosomal problem is also evident. Inuits lack yellow ancestry component what Yakuts and Yukaghirs have while Saqqaq doesn’t. Threfore, if Inuits came from Northeast Siberia, Yukaghirs and Saqqaq came from somewhere else.

The admixture analysis clearly shows that Saqqaq and modern Inuits are close to each other. Both share the green Beringian/Alaskan component (Saqqaq c. 60%, Inuits c. 80%). On top of that, Inuits have c. 15 Amerind and 7% Chipewyan, while Saqqaq does not have any Amerind and has less Chipewyan than Inuits. However, Na Dene and Inuit mtDNA is very similar as it is majority A2, and it looks like local.

Shaikorth said...

Kristiina, read page 31 of the supplements and what they say about modeling ancients and misleading tools. ADMIXTURE misleads here because it assumes admixed modern populations unadmixed and Saqqaq as a mixture from them. Moderns on those runs seem to behave like Khatri and Gujarati Brahmins in the example given in Falush et al. preprint, more endogamous population gets its own component:

"For ANI, ADMIXTURE finds that KSH receive almost all of their ancestry from ANI while Gujarati Brahmin (GBR) are more admixed but once the high relatedness of some KSH is accounted for along with other population specific drift, it is difficult to discern differences between the palettes of the two populations."

Kristiina said...

Sure, what they explain is probably true. However, they own result is that Saqqaq and Na Dene contain Siberian admixtures that Inuits lack. I am confident that Inuits will remain different from Yukaghir whatever method you use but closer to Koryaks and Itelmen, and if I remember correctly it was Itelmen who were the closest Siberian population to Ainus. Do you see any indications of Inuits being more recent than Saqqaq? The new Amur paper will be of interest here.

Shaikorth said...

The point they're making is that modern arctic populations are recently mixed while Saqqaq isn't that and Admixture has problems showing this.

Itelmen isn't the closest Siberian population to Ainu, they just had relatively elevated shared drift with them compared to other East Asians (Koryaks had the most). This did not extend to Native Americans.

Onur said...


I am confident that Inuits will remain different from Yukaghir whatever method you use but closer to Koryaks and Itelmen, and if I remember correctly it was Itelmen who were the closest Siberian population to Ainus.

The closest Siberian population to Ainu are Nivkh, who speak a language isolate.

Kristiina said...

This new paper detected a two-way admixture between Inuits and Athabaskans, so there is in any case recent Athabaskan admixture in Inuits.

This is what Raghavan et al, proposed in 2014 (

“The Siberian Old Bering Sea culture is the earliest expression of the Neo-Eskimo tradition ~2200 years before the present, developing into the Punuk culture around the sixth century CE. Almost concurrently, the Old Bering Sea culture developed into the Birnirk culture in the northern parts of the Bering Strait region. Interactions between people of the Birnirk and Punuk cultures gave rise to the western Thule culture on both sides of the Bering Strait, with contribution from the Paleo-Eskimo Ipiutak culture in Alaska.

“Present-day populations that are genetically closely related to the Paleo-Eskimos include the Greenlandic Inuit, Aleutian Islanders, and far-east Siberians.”

“Similarly, using SNP chip data from select Old and New World populations, masked for European admixture, we observe that the high-coverage Saqqaq individual forms a clade with the far-east Siberian Koryaks instead of with the Chipewyan, another Na Dene population that groups with the South American Karitiana. Outgroup f3-statistics and D-statistics show that the high-coverage Saqqaq individual is closer to Greenlandic Inuit than to the Na Dene speakers. Furthermore, the lack of support for the Saqqaq being closer to the Athabascans than to Karitiana is incompatible with a scenario where Saqqaq and Na Dene speakers share a fraction of their ancestries through a secondary Asian stream. Overall, our results support the Paleo-Eskimo migration into the Americas as being separate from that of the Na Dene Native Americans.”

“We additionally analyzed five ancient individuals dated to the 6th to 7th century CE and associated with the Siberian Birnirk culture, which is part of the Neo-Eskimo tradition and may be one of the cultural ancestors of the Thule. Evidence from both mitochondrial typing (hg A2a) and nuclear markers reveals that these individuals are genetically closely related to present-day Greenlandic Inuit, providing the first genetic evidence of an Old World population that was not only a cultural precursor of the Thule, but also either closely related to or a component of the ancestral Inuit gene pool. We also reassessed claims that the Sadlermuit population from Southhampton Island in the Hudson Bay region are remnants of the Dorset Paleo-Eskimo people on the basis of cultural similarities and mtDNA markers. We typed 10 Sadlermiut individuals, dating to the 14th to 19th century CE, to mtDNA hgs A2b and D3a2a, which are characteristic of the Thule/Inuit. Also, from their nuclear genome sequences, two of these individuals (Canadian Thule) form a clade with present-day Greenlandic Inuit and not the high-coverage Saqqaq individual.”

“Maximum likelihood trees with migration edges inferred with TreeMix show evidence for admixture between Paleo-Eskimos and Greenlandic Inuit, mediated by the Neo-Eskimo groups that include the Canadian Thule, Greenlandic Thule, and Siberian Birnirk. In all the cases, this gene flow event is among the first few migration edges to be inferred by TreeMix. Bootstrap support suggests that gene flow likely occurred in both directions between the Paleo- and Neo- Eskimo groups.”

“Furthermore, Paleo-Eskimo and Inuit peoples appear to have occupied the New World Arctic for more than 4000 years, with only a single population replacement (Thule) less than 700 years ago.”

If Raghavan proposes that Neo-Eskimos came from Birnik, it is not very far from Alaska, but I do not know if Raghavan really proposes any migration of Neo-Eskimos, or at least the majority of their genes, from Siberia. In their treemix tree Greenlanders are on a New World branch next to Karitiana and Anzick while Saqqaq is further
apart on a Siberian branch with Nivkhs.

As usual in science, disagreement seems to abound.

Shaikorth said...

The recent Athabaskan in Inuits that the paper detects is unrelated to older Paleo-Eskimo-related admixture in Na-Dene or that modern Arctic populations are more likely to be recently admixed than Saqqaq.

They link the Paleo-Eskimo to Dene-Yeniseian partly because Saqqaq was linked to Kets in a previous study.

Kristiina said...

The picture is not yet clear. More data is needed.

When I referred to Itelmen, I tried to remember the data included in these attachments:

Figure S15 - The genetic affinity of East Asian and Siberian populations to Nganasan and Itelmen measured by Patterson's D(Yoruba, X; Nganasan, Itelmen). (.pdf, 264 KB)

Figure S16 - Genetic affinity of East Asian and Siberian populations to Nganasan and Itelmen measured by Patterson's D(Yoruba, X; Nganasan, Itelmen). (.pdf, 264 KB)

Figure S17 - Genetic affinity of East Asian and Siberian populations to Nganasan and Chukchi measured by Patterson's D(Yoruba, X; Nganasan, Chukchi). (.pdf, 263 KB)

Figure S11 - TreeMix results with 0 to 5 migration edges. (.pdf, 693 KB)

Table S6 - Northeast Siberians (Itelmen and Chukchi) are more closely related to the Ainu than to the other East Asians (Ami, Atayal, Dai, Lahu and the Sherpa). (.pdf, 95 KB)

In Figure S15 and S16, Koryak, Chukchi, Eskimo, Karitiana, Surui and Ainu are closer to Itelmen while other East Asians, including Yukaghir, are closer to Nganasan. In Figure S17, Itelmen, Koryak, Eskimo, Karitiana and Surui are closer to Chukchi while Ainu and other East Asians, including Yukaghir, are closer to Nganasan.

So, in all these D statistics, Itelmen, Koryak, Eskimo, Karitiana and Surui behave in the same way while Yukaghir behaves differently. Therefore, I still insist, with the data we currently have, that Inuits did not come from the same direction as Yukaghirs but they belong to Beringian/Alaskan cultural area which also includes the Kamchatka Peninsula.

Kristiina said...

I had a closer look at the yDNA now that I have my material at hand. YDNA supports Raghavan model of a separate paleo-Eskimo migration.
Na Dene: Q-M3, Q-L54, C-P39, Q-NWT01
Inuit: Q-NWT01, Q-M3, C-P39
Chukchi: Q-M3, C-M48
Koryaks and Itelmen: C-M48, C-(xM48), Q1a-MEH2
(I omitted N1c and R1a.)

Saqqaq yDNA is Q-MEH2, and according to Malyarchuk et al, 2011 (
“An interesting cluster of Q1a*-MEH2 haplotypes was detected in Koryaks inhabiting the Sea of Okhotsk coast. Although such haplotypes are relatively frequent, exclusively in Koryaks (10.3%), they are characterized by low age, which corresponds to about 1.0 Ka. Nevertheless, it seems probable that this Q1a*-MEH2 cluster appears to be much older taking into account that Q1a* haplotype (xQ1a1, Q1a3, Q1a4, Q1a5, Q1a6) has been detected in extinct Palaeo-Eskimo individuals belonging to the Saqqaq culture (dated approximately 4.75–2.5 14C ky).1 It is noteworthy that according to results of SNP genotyping, the populations closest to the Saqqaq individuals are Koryaks and Chukchi.1 As that study suggested that the ancestral Saqqaq peoples separated from their Old World relatives about 5.5 Ka, it is possible that Q1a*-MEH2 lineage detected in Koryaks represents an ancient genetic component, which in the past united the peoples of Northeast Asia, North America and Greenland. In order to determine the range of Q1a*-MEH2 haplotypes detected in Koryaks we have performed a search for similar STR haplotypes in the YHRD 3.0 database. As a result, we did not find identical haplotypes among 91 493 haplotypes within 686 world populations, but the only similar 9-marker haplotype was found in Yukaghirs from Northeast Siberia (Yakutia). According to Pakendorf et al.,29 this haplotype (one-step differences from the Koryak Q1a*-MEH2 haplotypes are shown in bold) was detected in 4 out of 13 Yukaghirs studied and belongs to haplogroup Q-P36. Therefore, it is likely that the range of Q1a*-MEH2 may cover a distance of about 1000 km between the coasts of East Siberian Sea and Sea of Okhotsk. Coalescence age of the Koryak/Yukaghir Q1a*-MEH2 haplotypes is about 3.5±1.5 Ka, that is within the bounds of the Saqqaq culture dating.”

Therefore, it is Saqqaq, Koryaks and Yukaghirs who share the paleo-Eskimo yDNA while modern Inuits and Na Dene share Q-NWT01, Q-M3 and C-P39.

Shaikorth said...

The point is that modern Inuits have recent admixture (from Athabascans and Paleo-Eskimos), not that they have some connection to Yukaghirs.

Kristiina said...

However, this paper claims that Paleo-Eskimos and Na Dene descend from the same migration while Inuits descend from the Ymyakhtakh culture which would have brought Yukaghir, Chukchi-Kamchatkan and Eskimo-Aleut languages to Alaska/Beringia. To me it seems that Raghavan's paper does not support this view, neither the yDNA and mtDNA distribution of Na Dene, Inuits, Chukchis, Koryaks and Yukaghirs nor the D statistics of the Ainu paper.

This is what this new paper claims:
"The geographic connection between Paleo-Eskimos and the related Siberian groups probably became severed as subsequent waves of hunter-gatherers entering Eastern Siberia from the west during the Late Neolithic (Ymyakhtakh culture, 3,700-2,800 YBP) brought new cultures and new language groups. This phase of North Asian prehistory most likely involved the spread of Yukaghir, Chukchi-Kamchatkan and Eskimo-Aleut languages, whose presence in the extreme northeast of Asia intervened geographically between Paleo-Eskimos, Na-Dene, and their Old World cousins. Notably, the dates of the Siberian-Arctic split obtained under our model (~4,000-4,200 YBP, Table 2) also agree with this scenario that links the spread of the Ymyakhtakh culture (after 3,700 YBP) with the Arctic meta-population, i.e. ancestors of modern Chukchi-Kamchatkan and Eskimo-Aleut ethnic groups."

Ebizur said...


If I am not mistaken, the authors of that paper about Ainus have chosen those statistics to hint at the direction of a third genetic affinity of the modern Ainu population: Kamchatka (represented by Itelmens). The sampled Ainu population's other affinities point toward Sakhalin-Amur (represented by Nivkhs) and the rest of Japan (represented by non-Ainu Japanese). In other words, it appears that the sampled Ainus have not been genetically isolated from their neighbors in any direction despite their insular place of residence.

The Itelmens are linguistically related to Koryaks and Chukchis. Any difference between Itelmens and Chukchis in regard to their genetic relationship to Ainus that cannot be ascribed to differential gene flow from another group (e.g. Eskimos, Yukaghirs) should be ascribable to gene flow between Ainus and one or another of the Chukotko-Kamchatkan populations. In this case, the statistics seem to suggest gene flow between Itelmens and Ainus to the exclusion of Chukchis. I think such gene flow has been suspected based on other data, such as toponyms around the northern end of the Kuril Islands and the southern end of the Kamchatka Peninsula.

Shaikorth said...

Yukaghirs don't cluster with modern "Arctic" populations in either of their datasets according to their Finestructure results and they don't claim that they do. In fact, it's Saqqaq that clusters with Yukaghirs when Human Origins set is used.

Kristiina said...

After all this “crazy messing and digging around”, I would propose that Saqqaq spoke a Yukaghiric language. The Ymyakhtakh culture (4200–3200 years BP) that has been linked with the Yukaghiric languages was obviously very influential if it brought pottery, bow and arrow to Beringia and northern North America. Then, I would also suggest that these new inventions gave a boost to more autochthonous groups which gave rise to proto-Eskimo-Aleut, which may have started expanding from the Birnirk and Punuk cultures according to Raghavan, but we cannot exclude that similar languages were already spoken in Alaska. The fact that Ainus are somehow related to Koryaks and Itelmen could mean that the Chukotko-Kamchatkan protolanguage originated in the Okhotsk area which is located between modern Ainu-speaking area and Koryak-speaking area.

It is not a surprise that Chukchis behave differently from Koryaks and Itelmen in their relationship to Ainu as Chukchi yDNA is clearly different from Koryak yDNA. Chukchis are American-shifted as their main yDNAs are Q-M3 and Q-(xM3) and they have only a small amount of C-M48 while C-M48 is the main yDNA of Koryaks and Itelmen, and if I remember correctly, also Ainus carry C-M217.

I admit that this sketch leaves out Na Dene. I would wait for the ancestor of C-P39 to pop up in ancient samples on the Asian side to give an indication from where and when Na Dene languages arrived to America. I know that this interpretation does not answer to the question from where Na Dene got their Siberian admixture, which is linked with Saqqaq in this paper. Does anyone know if Kennewick man (Q-M3) did have any Siberian affinity comparable to that of Na Dene? I remember that in the admixture analysis, Kennewick man did carry the Siberian blue color that was lacking in Anzick and he was in America already c. 8,340–9,200 BP.

capra internetensis said...


Na-Dene and Inuit may both be paternally related to Koryaks and Paleo-Eskimos via C and Q, though at present the picture is far from clear.

Paleo-Eskimo and Koryak (and probably Nivkh) Q1a *also* belong under NWT01 (see Karmin et al), but as there are no Eskimo Q1a1a full sequences we can't tell how closely the Paleo-Eskimo and Neo-Eskimo Y chromosomes are related. But seeing what Dulik says below, they may not be very far apart; anyway, we can't use Q1a*-MEH2 and Q1a1a-NWT01 as a dividing line.

Dulik et al (2012): "From an NRY perspective, many Eskimoan Y chromosomes belonged to Q1a6[-NWT01], which has a TMRCA that predates the Paleo-Eskimo material culture. The Y chromosome of the ancient Paleo-Eskimo man was assigned to paragroup Q1a*, but the NWT01 locus was not sequenced.... In fact, four Koryaks also have Q1a* Y chromosomes, with the number of repeat differences being within the typical range of confirmed Q1a6 haplotypes. Thus, although a discontinuity in mtDNAs between the Paleo-Eskimo and modern Inuit has been shown, this finding may not be the case for Y chromosomes."

Q1a*-MEH2(xQ1a1a-NWT01, Q1a1b1-M25, Q1a2-M346) was reported in Koryaks by Battaglia et al (2013) and in Na-Dene (Northern Athapaskans and Tlingit) by Dulik et al as well. I'm not sure if that Koryak Q1a* is really Q1a or some kind of mistake for NWT01 though.

Regarding C2-P39, its closest cousin also might be found in Koryaks. P39 belongs to F1756 (C3g-B473 in Karmin et al). The B78 branch of this clade is found in Manchus, Altaians, and East Europeans, while the B77 branch is found in Koryaks. P39 and B77 are equally related to B78, but as yet we don't know how they are related to each other. But it's plausible that they'll be on the same branch.

Tlingit, the other extent branch of Na-Dene, have a little C2b-P39 (which could easily be assimilated from Athapaskans though) and also some C2-M217*. It's too bad we have only one Paleo-Eskimo Y chromosome (and all the Paleo-Eskimo DNA is from the eastern Arctic), possibly they had C2 as well.

The main haplogroup of both Chukchis and Siberian Eskimos is N-B202, but yeah, Chukchis have some Q-M3 (less than Eskimos but maybe not significant) and only a little C2b2-M48. Actually Koryaks have a little Q-M3 too but not as much.

In Tajima's sample Ainu (n=16) had 13% C2-M217 and 87% D1b-M55.

Tajima's Nivkh (n=21) had 19% P1(xR1a1), presumably all or mostly Q, as well as 38% C2, 29% O-M175, 10% R1a1, and 5% F(xK). In Lell's sample Nivkh (n=17) had 35% P1(xR1-M173, Q-M3), 35% C2b2-M48, 12% C(xC2b2), 12% K(xN1c1, O1a, O3, P1), and 6% O1a-M119. Rootsi's sample of Nivkh (n=55) had 20% O-M175 and no N-M231 of any kind.

The Nivkh P1* haplotypes formed a cluster which was shared with Nanai and neighboured non-M3 haplotypes from Chukotko-Kamchatka, so I guess is Q1a1 related to the Koryak and Eskimo types, but how closely I wouldn't venture to guess as the haplotypes are only 4 Y STRs! A Nivkh full sequence from Raghavan came out as C2b*-L1373, not belonging to any known branch apparently. I guess the O could be O2b most likely.

capra internetensis said...

On the mtDNA side there is a strong connection between Athapaskans and Inuit in the form of A2a. Athapaskans have predominantly A2a4 and A2a5, which they brought with them to the south and has diffused into other groups like Cree and Shuswap. Inuit have around half A2a1, which is also common among Yupik and Chukchi, and is even found at low frequency in Koryaks (even seen in a Selkup). Aleuts have their own branches of A2a as well as A2a1 (probably, this is a bit unclear). Alaskan Yupik have mostly A2b1 (from a small sample), which also found in Inuit, Siberian Yupik, and Chukchis, but rarely elsewhere.

The age of A2a is estimated to be about 4 to 7 thousand years.

Tlingit on the other hand mostly lack the Arctic-type A2. There are cases of A2a1 and A2b1 but they are known by oral tradition to be from Inuit adopted into Tlingit clans. One northern Tlingit clan (Humpback Salmon from Yakutat) is characterized by Athapaskan A2a5, but given the location this may well be an assimilated Athapaskan clan. (This is from Schurr et al "Clan, language, and migration history has shaped genetic diversity in Haida and Tlingit populations from Southeast Alaska".)

Known Paleo-Eskimos (too bad there are none from Alaska) were all D2a, and modern Arctic D2a seems to be nested within their diversity. The modal haplogroup of Aleuts is D2a1a, which is a very young clade. Though D2a is very rate or non-existent among Inuit of Canada and Greenland it does turn up rarely in Alaskan Inuit, and is quite common (as D2a2 and D2a1b) among Siberian Eskimos, more rarely seen in Chukchis. Basal D2a occasionally shows up in Tlingit too (but in at least one case this was from an adopted Athapaskan). The nearest Siberian relative is D2b, which has been found among Yakuts and Tungusics from Sakha Republic and also among Buryats and Kalmyks.

The age of D2a is estimated to be 4.67 ± 1.14 thousand years or 4.26 ± 3.06 thousand years, but since Saqqaq already had a fairly derived version it must be a good bit older than he is. Anyway, it could plausibly have been brought from Asia by the Paleo-Eskimos.

The other Arctic haplogroup associated with Neo-Eskimos is D3, found at low frequency among them and also Chukchis and rarely Athapaskans. Its Siberian relatives are found in the Altai-Baikal region and also among Yakuts, Evenks, and Nganasans.

Since Aleuts are linguistically related to Eskimos they have to be worked into any hypothesis about the Siberian origins, but it's hard to say just when they arrived. It could be as recently as a thousand years ago, which seems to be when D2a1a expanded; on the other hand it could go back to the invasion of the Paleo-Eskimos (There are lots of well-preserved Aleut remains which have already been tested for mtDNA, hopefully someone will give them a proper full genome treatment, it might be enlightening.)

Yukaghirs generally don't seem to be connected to Eskimos maternally. Sakha Yukaghirs from Federova et al (n=22) didn't have any American haplogroups at all. The Kolyma Yukaghirs from Volodko et al (n=100) also shared no haplogroups with Eskimos and Aleuts, only having a small amount of mtDNA related to D3. The Chuvantsi Yukaghirs of Chuktoka (n=32) did have ~35% Eskimo/Chukchi-type haplogroups (A2a, A2b, D2a, and D3), but that can be attributed to intermarriage.

capra internetensis said...

BTW Kristiina thanks for pointing out that Yukaghir Q1a information, I never saw that before.

I compared the Yukaghir haplotype in question (which was according to Malyarchuk close to the Koryak Q1a* cluster) to the haplotypes of Koryaks with Q2b from Karmin et al and it differs by only one step from Q2b-B284 and by 2 steps from Q2b-B283.

So the Koryak Q1a* is really a branch of Q1a1a-NWT01 and the Yukaghir Q1a* clusters with the Koryak type.

I looked for this haplotype in Alaskan Eskimo data and couldn't find anything very close to it.

Kristiina said...

Thanks, Capra. Let's see if that upcoming Amur paper gives more light on these issues. In any case, ancient DNA from Southern Alaska, Beringia, Kamchatka and Okhotsk would be the best option if we want to resolve the Arctic puzzle.