search this blog

Tuesday, May 30, 2017

Ancient Egyptians less Sub-Saharan than present-day Egyptians


Over at Nature Communications at this LINK. Emphasis is mine:

Abstract: Egypt, located on the isthmus of Africa, is an ideal region to study historical population dynamics due to its geographic location and documented interactions with ancient civilizations in Africa, Asia and Europe. Particularly, in the first millennium BCE Egypt endured foreign domination leading to growing numbers of foreigners living within its borders possibly contributing genetically to the local population. Here we present 90 mitochondrial genomes as well as genome-wide data sets from three individuals obtained from Egyptian mummies. The samples recovered from Middle Egypt span around 1,300 years of ancient Egyptian history from the New Kingdom to the Roman Period. Our analyses reveal that ancient Egyptians shared more ancestry with Near Easterners than present-day Egyptians, who received additional sub-Saharan admixture in more recent times. This analysis establishes ancient Egyptian mummies as a genetic source to study ancient human history and offers the perspective of deciphering Egypt’s past at a genome-wide level.


Schuenemann et al., Ancient Egyptian mummy genomes suggest an increase of Sub-Saharan African ancestry in post-Roman periods, Nature Communications 8, Article number: 15694 (2017), doi:10.1038/ncomms15694

See also...

A homeland, but not the homeland

96 comments:

Salden said...

They're like Copts in the northern areas. It's damn obvious that Egypt as we know it was settled by migrants from Eurasia who mixed with already settled populations in Lower Egypt.

Strandloper said...

Was the Old Kingdom population different than the New Kingdom?

Salden said...

There wasn't any mass dispacement by or mass breeding from totally ethnically seperate invaders.

Salden said...

Also, the Ancient Egyptian sample leaning so strongly towards Neolithic Levant either rules out it being a Hyskos migrant descendent or that the Hyskos were significantly different as an ethnic group from the Lower Egyptian population.

Salden said...

It also leans towards a Bedouin sample.

Romulus said...

Holy shit they are most similar to EEF!

Matt said...

@ Davidski, semi on topic, I discussed some D stats with @ Kurd on the last thread. To measure African admixture between ancient and modern populations, which he suggested could be the reason for f/D stat vs Fst differences which couldn't be explained by drift (where I was inclined instead that there was some phenomenon biasing formal stats against more heterozygous populations).

These are here: https://pastebin.com/vZBFe3PA

I'd appreciate if you could run these, if possible. If not n/p.

(I'm kind of expecting them not to be significant, as for one, if they're significant, seems kind of strange they've not yet popped up in last year's papers...).

Simon_W said...

@ Romulus

"Holy shit they are most similar to EEF!"

Where exactly? In the admixture analysis they show more than 50% Natufian-like (brown) ancestry and also some Iran-related (light blue) ancestry, while the Anatolia_N-related (blue) ancestry is only a fraction of their genomes. In the admixture analysis they look most similar to Levant_BA, is my impression.

Shaikorth said...

In the outgroup f3 test their closest relative by a small margin is Levant_N, though EEF-type Anatolia_N follows closely. Those f3 stats also have modern Egyptians noticeably closer to EEF and modern South Europe than to the Levant, ancient or modern.

Anthro Survey said...

@Simon_W:
Gee thanks, you stole my comment. :D
(Unless he included meant Levant_Neo by his EEF category?)

@topic

Anyway, from what it appears, Egyptians received additional SSA asmixture, but also some CHG-rich admixture from the Near East in post-Roman times. That mathematically explains their position on the PCA and also apparent from this as well as other admixture results I've seen.

You know, the admixture analysis was not so satisfactory in this study. The red mainly corresponds to "WestAfrican-rich" ancestry of some sort and it seems that the presumably ancient East african DNA Ancient Egyptians carried(which formal stats confirmed here)got "buried" in the brown.

Awale featured a much better admixture study of modern populations with Egyptians sporting Yoruba-like and Dinka-like components. The Copts just had a Dinka-signal, while modern Egyptians had both. I'm sure you've seen it, but I can find and post the link here if anyone wants.

Anthro Survey said...

I'm a bit disappointed they didn't test for any steppe-related ancestry in this study given that these samples all post-date the Caananite samples and given that modern Lebanese supposedly got their steppe ancestry after that time. It's highly likely that this introgression occurred in Egypt, too.

Anthro Survey said...

@Rob:

It was pretty easy to guess they would be clustering in that general area.

The thing that struck me was how impacted they were by CHG-rich Messopotamian DNA---almost like Bronze Age Levant.

Formal stats also show they had some ~6% SSA in contrast to the Natufians, so the Upper Egypt AA urhmeimat theory is not dead.

Roy King said...

I think that this paper and the Canaanite paper shed a lot of light on the archaeology of the region. The ancient Egyptian samples are just north of the Levant Neolithic, slightly further north are the EBA Jordanians and even further north are the Sidonians. This gradient reflects the degree that Iranian Chalcolithic moved south to admix with the Levant Neolithic. From the unpublished work on the Chalcolithic Israel samples, these Chalcolithic samples will fall approximately where the ancient Egyptians lie.

http://italicroots.lefora.com said...

Apparently the additional Iran neolitich admixture in modern Egyptian was male mediated, unlike the African one. It looks like Arabs and maybe Turks did to Egypt what Spaniards did to Mexico.

Matt said...

Egyptians modern and ancient projected on a modern West Eurasian PCA, from Supplementary Figure 3 - http://i.imgur.com/x8T9hud.png / http://i.imgur.com/Ufg5CEE.png, with Lazaridis 2016 equivalent by side.

Really unclear to me from that how much extra Iran_Chl ancestry modern Egyptians should have compared to the ancient samples. Seems like you would think from that PCA, almost none.

Also would say, the formal models in Lazaridis 2016 looked to be underestimating Levant_N or Anatolia_N like ancestry in Iran_Chl and Levant_BA, comparing to PCA - http://i.imgur.com/OYmYLBH.png (from Extended Data Figure 5). Based on the PCA, Jordan_BA would probably more like 29% Iran_Chl than 44%.

Unless of course some effect here is present where African ancestry is confounding position on West Eurasia PCA (e.g. more Iran_Chl ancestry pushing "east" is offset by African ancestry pushing "west"?).

Anthro Survey said...

@Roy:
My thoughts exactly. But when do you think Iran_Chalco-like DNA began arriving to Egypt?
How about Lybia and the Maghreb?

@italicroots
Arabs? Possibly. Turks? lol Everyone knows that the Ottoman conquest was not characterized by large numbers of nomads invading and settling the area. The Ottomans were a highly urbanized, sedentary state. It was much more comparable to Roman conquest and perhaps to European conquest of Africa and SE Asia whereby a limited numbers of governors from Bosnia/Albania/Rumelia and/or Anatolia were sent to administer.

Romulus said...

Lots of interesting mtDNA here. H13 and lots of U:

U
U1a1
U1a1a3
U3b
U3b
U5a
U6a
U6a2
U6a3
U7
U7
U7
U8b1a1
U8b1a1
U8b1b1

Davidski said...

I wonder where that U5a came from?

Matt said...

Re: Iran_Chl, the ADMIXTURE from the supplement allows us to do a rough little time series comparison of the Neolithic->Metal Ages->recent in Iran vs the Levant:

http://i.imgur.com/NbSpoHe.png

At the start of the sequence, in the early Neolithic, the Levant is pretty much equal parts brown component (Natufian) and blue (Anatolian). Iran_N is cyan (Iran) with a minority pink (South Asian).

Moving to the metal ages, changes in components are seen with the introduction of cyan to the Levant and brown and blue components to Iran_Chl.

At this stage, it seems like the shift in Iran is slightly greater than in the Levant; just comparing the amount of extra brown+blue in Iran_Chl (about 30%) vs cyan in Levant_BA (about 20%).

I would speculate that this is perhaps because geographically and genetically intermediate Mesopotamian cultures may have been who were actually expanding into both and may have been either a) closer to Levant than Iran or b) made more impact in Iran than the Levant.

Move to the modern age, and you have a more even picture, where the modern Levant is narrowly more related to Levant_N (looks about 60% of ancestry), while Iran is narrowly more related to Iran_N (again about 60-70% ancestry).

(Change since the Metal Ages seems maybe more pronounced in the Levant than Iran, but changes between Neolithic to Metal Ages maybe more pronounced in Iran?).

Modern Egyptians seem to have more Iran_N ancestry than ancients, but the Ancient Egyptian samples also had less than Levant BA (and less Anatolian), and modern Egyptians have less Iran_N ancestry than the modern Levant (even proportionately after their greater SSA / Yoruba-like ancestry is considered).

MaxT said...

We would need various samples from Early Dynastic Period to Middle Kingdom to make any collective conclusion about ancient Egyptians.

Early Dynastic Period - 3100–2686 BC
Old Kingdom - 2686–2181 BC
1st Intermediate - Period 2181–2055 BC
Middle Kingdom - 2055–1650 BC
2nd Intermediate Period - 1650–1550 BC
New Kingdom - 1550–1069 BC
3rd Intermediate Period - 1069–664 BC
Late Period - 664–332 BC

New Kingdom appears to be closest to Bedouin Arabs based on the PCA from this study.

MaxT said...


2nd Intermediate Period - 1650–1550 BC

"The Second Intermediate Period marks a period when Ancient Egypt fell into disarray for a second time, between the end of the Middle Kingdom and the start of the New Kingdom."

https://en.wikipedia.org/wiki/Second_Intermediate_Period_of_Egypt

Map shows mass migrations from Near East just before appearance of New Kingdome.

New Kingdom begins after Second Intermediate period.

MaxT said...

Just before New Kingdome.

"It is best known as the period when the Hyksos made their appearance in Egypt and whose reign comprised the Fifteenth dynasty."

"The Hyksos were a people of mixed origins from Western Asia, who settled in the eastern Nile Delta, some time before 1650 BC. The arrival of the Hyksos led to the end of the Thirteenth Dynasty of Egypt and initiated the Second Intermediate Period of Egypt. In the context of Ancient Egypt, the term "Asiatic" – which is often used for the Hyksos"

Davidski said...

@Matt

https://drive.google.com/file/d/0B9o3EYTdM8lQOUs1bDhNS2ZXaEE/view?usp=sharing

Knight said...

@Salden These sample aren't for Northern Egyptians. It is for Middle Egyptians from Abusir El-Meleq south of Fayum in Middle Egypt ( near to Southern Egypt ). I would like to add that the majority of Copts live in Middle and Upper Egypt and Many of the Copts in the North are just immigrants from Middle Egypt. The Hyksos stuff is pretty funny, Hyksos gained control only on the North-Eastern-most part of Egypt which is very very far from South of Fayum ( the area from which the samples are taken ), they never reached Fayum or even near. Also the Hyksos were totally kicked out and massacred by Ancient Egyptians and lastly these samples are taken from Egyptian MUMMIES. It is pretty obvious now that Ancient Egyptians were just Eurasians.

a said...

Im so excited for these new results;surely by now they have sequenced one of the most famous Egyptian dynasties. Of interest since a sample was taken way back in 1968. How long do we have to wait to find out this dynasty genetic identity-why did it take soo long.

https://jomarchant.com/390/tracking-down-a-long-lost-piece-of-king-tut

What’s less well known, however, is that this isn’t the whole King Tut: a tiny piece of his mummy was removed in 1968 and flown to Liverpool. When researching my book about Tutankhamun’s mummy, The Shadow King, I couldn’t resist tracking it down.

The Eighteenth Dynasty of Egypt (notated Dynasty XVIII, alternatively 18th Dynasty or Dynasty 18) is classified as the first Dynasty of the Ancient Egyptian New Kingdom period, lasting from 1549/1550 BC to 1292 BC. It boasts several of Egypt's most famous pharaohs, including Tutankhamun, whose tomb was found by Howard Carter in 1922. This dynasty is also known as the Thutmosid Dynasty for the four pharaohs named Thutmosis.

Famous pharaohs of Dynasty XVIII include Hatshepsut (c. 1479 BC–1458 BC), longest-reigning woman-pharaoh of an indigenous dynasty, and Akhenaten (c. 1353–1336 BC), the "heretic pharaoh", with his queen, Nefertiti.

Dynasty XVIII is the first of the three dynasties of the Egyptian New Kingdom, the period in which ancient Egypt reached the peak of its power.





The 1922 discovery by Howard Carter and Lord Carnarvon[7][8] of Tutankhamun's nearly intact tomb received worldwide press coverage. It sparked a renewed public interest in ancient Egypt, for which Tutankhamun's mask, now in the Egyptian Museum, remains the popular symbol. Exhibits of artifacts from his tomb have toured the world. In February 2010, the results of DNA tests confirmed that he was the son of the mummy found in the tomb KV55, believed by some to be Akhenaten.....



His ydna line & autosomal origins confirm...............

MaxT said...

@Knight

Thats what i could find about them and Hoyksos seem to have ruled Egypt for a while aka 15th Dynasty.

"The Hyksos practiced horse burials, and their chief deity, their native storm god, Baal, became associated with the Egyptian storm and desert god, Set."

I agree that they were predominantly Eurasian but we also need to look into what earlier dynasties were like before New Kingdom.

MaxT said...

Horse were introduced to Egypt just before New Kingdom.

"Horses were introduced into Egypt during the Second Intermediate Period (about 1700-1550 BC). The earliest remains of horses are a few bones from Avaris and the skeleton of a horse found at Buhen. The Buhen remains date to the early Second Intermediate Period, but this date is disputed. In the wars between the Theban 17th Dynasty and the Hyksos both sides used horses. In later times, the kingdom of Kush in the Sudan was famous for its horses, perhaps from good grazing grounds in areas of Upper Nubia: in the Victory Stela of king Piy, special mention is made of the royal attention to horses."

"In the New Kingdom horses were animals of the military elite and the ruling class. In general Egyptians did not ride on horses but used them for chariots. Two horses are the rule. Horseshoes were not used. Egyptian horses, which were probably almost identical to those in the Near East, are rather small by comparison with modern horses, and attested in different colours (brown, reddish etc.)"

http://www.ucl.ac.uk/museums-static/digitalegypt/foodproduction/horse.html

Knight said...

@ MaxT These samples are taken from south of Fayum in Middle Egypt. The Hyksos stuff is pretty funny, Hyksos barely gained control only on the very North-Eastern-most part of Egypt which is very very far from Fayum ( the area from which the samples are taken ), they never reached Fayum or even near. Also the Hyksos were totally kicked out and massacred by Ancient Egyptians and lastly these samples are taken from Ancient Egyptian MUMMIES. It is pretty obvious now that Ancient Egyptians were just Eurasians. There is no any evidence of mass immigration to Ancient Egypt which could significantly change the population, so please stop the non-sense. For instance, Iraq was conquered by many nations such as Mongols but that couldn't change the population because there is also no evidence of any mass migration. It is pretty obvious now that Ancient Egyptians were just Eurasians. I have met so many Afro-Centrists on the Internet hiding behind fake accounts and identities and trying to publish fake and wrong information which serve their agenda although they obviously has nothing to do with Ancient Egypt.

Anthro Survey said...

@Davidski: I had a few d-stats in mind and was wondering if you could run a few please? If not, no worries.

@Knight: "There is no any evidence of mass immigration to Ancient Egypt which could significantly change the population, so please stop the non-sense."

lol.... What?! You're looking at it, kiddo! I agree that we don't want to rush to judgment and attribute the CHG/Iran shift in Egyptians to the Hyksos, but it's pretty clear that the western fertile experienced this kind of shift during chalco and bronze age times. They are not undisturbed Natufians.

I agree that various Afrocentrist(we wuz kangz) and Eurocentrist(we wuz aryanz) agendas are out there trying to claim Egypt and other MENA cultures as theirs and whatnot. Pride is healthy and great but does not give one the right to throw objectivity out the door. I think it was pretty clear/intuitive that Egyptians were of the Fertile Crescent genetico-cultural contiuum even before this paper. :D

Davidski said...

Hyksos can't explain the significant Iran_ChL shift in the Levant as far back as the Chalcolithic as per that recent Harvard abstract. The same shift probably affected Egypt at around the same time.

Which D-stats do you want?

Ryan said...

How the hell did they get to Egypt in such large numbers though? Kind of some geographic barriers there.

Anthro Survey said...

@Davidski

First off, if anyone already requested this and you have them, go ahead and send me the file/link.

If not, I think what I want would be like this:

Natufian X Mbuti Chimp
X=Levant Neo, Levant BA
Mbuti---also try other SSA groups from E or W Africa. Obv, Horners will not work because they have a ton of (confounding) Natufian-like ancestry.

Also
-Levant_Neo Levant_BA SSAgroup Chimp
-Iran_Neo Iran_Chalco SSAgroup Chimp
-Anatolia_Neo Anatolia Chalco SSAgroup Chimp

Basically, I want to get a rough idea of when SSA admixture in the region began. This ties in with the AA urheimat hypothesis in north Sudan.

MaxT said...

@Knight

What ruling elite do these mummies belong to? They don't seem to mention it.

Based on Hyksos dynasty maps i have come across, Hyksos kings seem to have ruled much of Egypt, including the Fayum region. They were kicked out much later but Hyksos kings ruled for a long time.

I'm more interested in Hyksos introducing horses to Egypt, their horse burial practice is pretty interesting as well.

Davidski said...

Make a text file and upload it somewhere. But you have to use the correct names, like not Levant BA but Jordan_EBA, and not Iran_Chalco but Iran_Chalcolithic.

Anthro Survey said...

@Davidski:
Forgot to add---- For the SSA groups, try Dinka if it's available. I remember Copts had that signal, but lacked "Yoruba"(unlike regular Egyptians). I sae this in one of Awale's blog entries featuring a K-admixture study.

Anthro Survey said...

Sounds good. I'll do what Matt did, then.

Gihanga Rwanda said...

Let's ignore for a minute that this site, Abousir el Malaq, doesn't happen to be one of the few Hyksos sites in Egypt; it's important to note that the Hyksos were preceded to Egypt by the influx of the Aamu/Asiatics from the Southern Levant. We need both older samples and genomes from Upper Egypt for a more comprehensive picture of ancient Egypt; imho, it's pretty intuitive that Egypt belonged to two distinct genetic/cultural continuums: Levant/Fertile Cresent/SW Asia and Nile Valley/EasternSahara/East Africa.

Gihanga Rwanda said...

@ Davidski and Anthro Survey

In addition to Dinka, could we also include Mota in the analysis?

JohnP said...

About the article:
Very consistent. Going Westwards trough North Africa, I expect less CHG and gradually appearing more WHG when closing to Iberia, similar to Anatolia_N or Europe_EN.

@Gihanga Rwanda
If anything, before the Hyksos, Egypt was a very closed society (specially because they were barters and didn't used precious metals as currency, so trade was extremely limited, so was travel and migration) and any influx of people wouldn't change them dramatically.
If they changed, instead of being very Levant_BA, they would have a little less CHG and that's it.
There's no presence any non-Caucasian admixture either, removing the possibility of a more African past people, which is very consistent historically.
If you're an Afrocentrist, and I hope you're not, Dark Skin™ is not a trait unique to Africans and light-skinned Caucasians can get very dark when tanned too.

Gihanga Rwanda said...

@ JohnP The study actually detected SSA ancestry (~10%) in these three individuals, which is comparable to contemporary Copts, who possess East African ancestry but lack the more recent Yoruba-like ancestry found in their Muslim neighbors.

Anthro Survey said...

@Gihanga
Claro que sí! I might not be as savvy as some of you other folks at this, but something like that I wouldn't overlook. ;-) I have to get to my desktop first, tho.

@JohnP
Of course the notion of Old Dynasties having a high SSA content is ridiculous but I don't believe it was zero. There is probably something to AA expansion theory from north sudan and this would have brought some SSA to Egypt, NA and Levant/Messopotamia. Still, it would be mostly back-migrated Natufian-like ancestry.

You're correct-----Maghrebis DO have a greater share of Loshbur-like ancestry and lower Iran/CHG than folks east of Cyrenaica. This is apparent in one Admixture study from Lazaridis as well as in Eurogenes K13. But it's certainly not as high as in Iberia or Sardinia/EEF. My guess? Pre-Neolithic Iberomaurusian culture were largely WHG who came via Sicily or Iberia.

Rob said...

Yes I'm looking forward to seeing the results of N African palaeolithic
I suspect IM would be natufian like
The WHG - like signal would come from Mesolithic Capsian industries

JohnP said...

@Gihanga Rwanda
There's no outer influx, which means that the SSA presence is just residuals from the Natufians or something like that.

I would like to know when Bedouins received their East African admixture. They were clearly a Levant_BA-like population who contacted East Africans at the very south of Arabia at some point. Normal migrations from there would help spreading East African genes to the rest of the Crescent.
Then, the Islamic Invasions would change dramatically many things, boosting their admixture in local populations and including the importation of Yoruba slaves into the Middle East and North Africa.

Salden said...

Oh, is there a North African focused study coming up?

Knight said...

@ MaxT LOOOOOL, That can be only based on your Afro-Centric agenda. It seems that you know NOTHING AT ALL about the Egyptian history my amateur Afro-Centrist friend, i really can't stop laughing, please stop your pure nonsense. Hyksos NEVER EVER reached Fayum or even near, they BARELY gained control on the very North-eastern-most of the country and even in this small area which is very very far from Fayum, they were tiny minority and they weren't welcomed at all and there were so many revolutions against them. Actually they didn't stay for long time as you said, it was only 90 years, then they were TOTALLY KICKED OUT and MASSACRED by the EGYPTIAN pharaoh Ahmose I who was son of Seqenenre Tao and Brother of King Kamose. Again my Afro-Centrist friend, these samples are taken from Ancient EGYPTIAN (((MUMMIES))). It is pretty obvious now that Ancient Egyptians were just Eurasians, there is no any evidence of mass immigration to Ancient Egypt which could significantly change the population, so please stop the non-sense. As i said before Iraq for example was conquered by many nations such as Mongols but that couldn't change the population because there is also no evidence of any mass migration. I really hope that Afro-Centrists stop trolling and stop their pure non-sense. They are desperate that they hide themselves behind fake accounts and make up funny pathetic stories and lies about themselves and their life. Some of them even pretend to be Western European as if this gives them more credibility to their pack of Afro-Centric lies they try to sell. Based on the Afro-Centric Logic, Modern Syrians are Turks because they eat Shawarma which is Turkish LOL.

Lenny Dykstra said...

This isn't surprising.

The Roman gladiator from `York, 3DRIF-26, had dental isotope testing placing him in either Sinai or Egypt and his closest Oracle matches turned out to be Egyptian Copts: http://www.anthrogenica.com/showthread.php?7496-coptic-oracle-results&p=165139&viewfull=1#post165139

Modern Copts and ME Christians in general show much lower levels of SSA than their Muslim counterparts. It's become clear SSA admixture in the Near East dates laregly to the Islamic slave trade.

Matt said...

@ Davidski, thanks. So stats seem to show:

* Modern Levant - Lebanese and Druze - are both significantly shifted towards African populations, relative to Jordan_EBA (significantly negative; both Lebanese and Druze by about the same amount).

* Modern Armenian is not significantly shifted towards Africans relative to Armenia Chalcolithic or MLBA (negative, less than -2 or -3).

* Modern Iran is not significantly shifted towards Africans relative to Iran_Chalcolithic (zero).

* Modern Germany is not significantly shifted towards Africans relative to Bell_Beaker_Germany (slightly the opposite if anything; positive, less than 2 or 3).

(http://i.imgur.com/Fn553Aq.png)

So African ancestry may explain why Levant does not have the highest f3(African, Modern, Ancient) stats where Ancient is an ancient Levantine population, despite lowest Fsts.

But looks like it would not explain for Armenia, Iran or Germany why the highest f3 is not found for the region, when lowest Fst is.

If possible, few more stats to test similar scenario of ancient vs modern for a few more populations: https://pastebin.com/RD5vC1Pf (to check if the recent populations are shifted towards East Asians, South Asians, Natufians or BedouinB relative to the ancient).

Samuel Andrews said...

The newly published ancient Egyptian and Canaanite genomes challenge an African origin for Afro Asiatic languages. Because think about both are ancient Afro Asiatic speakers and have no or tiny traces of African ancestry.

But then again aren't there Afro Asiatic speakers with no traces of Eurasian ancestry? Maybe there's common ancestry Egyptians-Canaanites and Afro Asiatic speakers share which is being absorbed by Eurasian and African categories because we don't have the appropriate ancient DNA.

Samuel Andrews said...

Does anyone know how to explain the extra dose of IranNeo stuff in modern SouthWest Asians? JordanEBA, Egyptians, and the Roman-era Middle Easterner from Britain lack it. Even European Jews have this extra layer of IranNeo stuff. But interestingly Yemanite_Jews lack it which indicates they immigrated to Arabia before an influx of new admixture from the northern part of the Middle East.

nee4speed111 said...

I'm willing to give the coptic genomes I have, if someone wants to do some sort of D-stat or other test with them

Nirjhar007 said...

So lets see if something similar happens for ASI in the subcontinent ...

Kurti said...

As I thought, Ancient Egyptians look like a Levant_BA or Levant_ late Neolithic population rather than Levant_Neolithic/Natufian. So ancient EGyptians might indeed be a Bronze Age arrival in the region, explains their more modern Middle Eastern like DNA in comparison to other North Africans who look like Levant_Neolithic without much Iran_Neo/CHL ancestry.

Kurti said...

@Anthony
"The thing that struck me was how impacted they were by CHG-rich Messopotamian DNA---almost like Bronze Age Levant. "

It's Iran_CHL like. As I have been proposing for a very long time. the Near East could be divided into three major fractions a Iran_Herder like that stretched all the way into Caucasus, Mesopotamia. a Anatolian_Farmer like that stretched from North Levant all the way into Anatolia and a Levant_Farmer group that started around South Levant (Jordan) all the way into North_Africa and Arabia.

Where do you think the Egyptians got their Pyriamids from? From the same region where they got their Iran_CHL ancestry from. Mesopotamia! Look at those Pyramids they are just modified Mesopotamian Ziggurats.

Unknown said...


Is the light red component in Iranians the ASI/Ust'-Ishim admixture? If not then I guess purple is ASI. If the purple component is ASI then what is light red component and green components in Iran Neolithic? Maybe light red is ANE but no idea what are these small green components.
http://i.imgur.com/NbSpoHe.png

Kurti said...

@Unknown

Iran_Neo predates ASI. this light red is shared anceestry. If anything the red light is Iran_Neo ancestry typical today for modern South_Central Asians.

The whole graphic with it's components are made around ancient Egyptian genomes to explain their ancestry. Thats why Natufian looks like it has 15% Anatolian_Neo ancestry. Do you believe that can be true?

Obviously it is just additional shared ancestry with Anatolian_Neo where Anatolian_Neo is on the receiving end.

Unknown said...

ASI component more than likely predates Iran Neo as Ust'-Ishim and even Oase1 seem to be ASIs with a WHG/West Eurasian shift. Although the Basal Eurasian part of Iran EN amay predate Oase1 but probably not as BEs must have changed since Paleolithic too as did UP Europeans although Paniya isn't much different than Ust'-Ishim except for rheir Iran EN admixture. Any idea what those green components could be? Anatolia Neolihic is pretty much LevantNeo with WHG, Iranian and probably CHG admixture that acquired drift and LevantNeo is Natufian with Iranian, Anatolian and probably some WHG admixture. Anatolia Neolithic very likely has much more than just 15% of Natufian ancestry but those components are overlapping.
Modern South Asians probably have more Iran EN than Iran EN had that light red.

Unknown said...


>Does anyone know how to explain the extra dose of IranNeo stuff in modern SouthWest Asians? JordanEBA, Egyptians, and the Roman-era Middle Easterner from Britain lack it. Even European Jews have this extra layer of IranNeo stuff. But interestingly Yemanite_Jews lack it which indicates they immigrated to Arabia before an influx of new admixture from the northern part of the Middle East.

Immigration and trade from South Asia into the Persian Gulf existed even before the oil boom. Sumerians traded with India for example.

Grey said...

Davidski said...
"Hyksos can't explain the significant Iran_ChL shift in the Levant as far back as the Chalcolithic as per that recent Harvard abstract. The same shift probably affected Egypt at around the same time."

domino effect?
- hyksos originally near steppe
- steppe push hyksos
- hyksos push people south of them
- people pushed by hyksos effect Egypt first
- actual hyksos arrive later

MaxT said...

@Knight

You sound like a nut-case. My expertise mostly lies in Indo-European expansions, and Hykosis use of horses and horse burial is similar to what was practiced on the steppes and i assumed they were first Indo-Europeans to arrive in Egypt. This looks unlikly for now based on these mummies since they do not have steppe component.

I have to agree with @Anthro Survey on old Dynasties.

MaxT said...

@Grey

My first thoughts as well, but these mummies do not have steppe-input. It's possible that one of those Near Eastern population could have brought horses through trade from steppe people?

Grey said...

Anthro Survey said...

"Of course the notion of Old Dynasties having a high SSA content is ridiculous but I don't believe it was zero"

the Cleopatra thing definitely and also generally yes but

1) ancient egypt was around a long time
2) they had a lot of dynasties over that time
3) my guess is a lot of dynasties started out as mercenary rebellions
4) nubia was one of the places they recruited mercenaries from

so i wouldn't be surprised if there was at least one nubian dynasty in the list somewhere during the two+ thousand years but not necessarily one having a big genetic impact.

Grey said...

MaxT

if it was a rolling domino effect then maybe each population had some elements of the other populations so say it was (temporary labels)
- steppe pushing
- hyksos pushing
- southern
- into egypt

then maybe
- "southern" were mostly southern + some hyksos + a few steppe
- "hyksos" were mostly hyksos + some steppe

?

Singh said...

@Kurti

ASI is synonyms with ASE/ENA, Iran_Neolithic is not older than those those components. We can see that in Lazaridis et al (2016) chart.

ASI split just before East Eurasians and Australasians split from each other. ASI shows affinity to both groups but closer to mainland groups like East Eurasians.

On Eurogenes Basal K7, this is how ANI and ASI break-down.

Gond (ASI shifted-tribe)
East Eurasian 34%
Ancient North Eurasian 22%
Basal Rich 18%
Southeast Asian 13%
Oceanian 10%

Spreadsheet has various South Asian populations from Gujarati, Bengali, Kalash, Pathan etc https://docs.google.com/spreadsheets/d/1tFAa7oxWpcNN-OdMMjBdb4NeWKG7EkpKMzZJVW2_MME/edit?pref=2&pli=1#gid=289082796

http://eurogenes.blogspot.com/2016/07/sneak-peek-basal-eurasian-k7.html

Over all, ANE and ASE are dominant ancestral component in South Asians, followed by Basal. We still need ancient DNA from South Asia though.

Unknown said...

Singh Gond likely has recent mongoloid east asian admixture.
Could you or someone else here run fineStructure or Gedmatch and formal stats if you know how on Paniya, Pullinyar, Munda, Mala and Iran Neolithic along with some modern Pakistanis/North Indians and Ust'-Ishim, Oase1, Kostenki, Vestonice and Goyet because Aurignacians/Gravettians and South Asians seem to be connected.

Singh said...

@Unknown
No, it's appears to be same with all populations in both ANI-shifted and ASI-shifed. That's how that componenet breaks down.

Spreadsheet has populations you mentioned, including Chenchu and Chamars. Some bloggers have also pointed this out. East Eurasian-like component, followed by Southeast Asian and Oceanian. Since this whole packpage as "ASI" appears unique to South Asians, it's likely very ancient and affinity dates before the split.

MaxT said...

@Grey

Even if that was the case we would still see small % of steppe-input in these mummies. They appear to be from same time period as Hyksos. I'll just go with idea that these mummies are not Hyksos for now.

MaxT said...

@Unknown
"Ust'-Ishim, Oase1, Kostenki, Vestonice and Goyet because Aurignacians/Gravettians and South Asians seem to be connected."

Learn some basic shit. That connection is ancient before "Crown Eurasians" (WHG/ANE/ASE) split up, they share more alleles with eachother than they do with 'Basal Eurasian'. Again, this connection you are seeing dates before Crown Eurasian split into WHG/ANE/ASE.

The genetic history of Ice Age Europe' from Fu et al 2016

"A previous genetic analysis of early modern humans in Europe using data from the ~37,000-year-old Kostenki14 suggested that the population to which Kostenki14 belonged harboured within it the three major lineages that exist in mixed form in Europe today4, 15: (1) a lineage related to all later pre-Neolithic Europeans, (2) a ‘Basal Eurasian’ lineage that split from the ancestors of Europeans and east Asians before they separated from each other; and (3) a lineage related to the ~24,000-year-old Mal’ta1 from Siberia. With our more extensive sampling of Ice Age Europe, we find no support for this. When we test whether the ~45,000-year-old Ust’-Ishim—an early Eurasian without any evidence of Basal Eurasian ancestry—shares more alleles with one test individual or another by computing statistics of the form D(Test1, Test2; Ust’-Ishim, Mbuti), we find that the statistic is consistent with zero when the Test populations are any pre-Neolithic Europeans or present-day east Asians3, 13. This would not be expected if some of the pre-Neolithic Europeans, including Kostenki14, had Basal Eurasian ancestry."

"Han Chinese share more alleles with two Villabruna Cluster individuals (Loschbour and LaBrana1) than they do with Kostenki14, as reflected in significantly negative statistics of the form D(Kostenki14, Loschbour/LaBrana1; Han, Mbuti)4. This statistic was originally interpreted as evidence of Basal Eurasian ancestry in Kostenki14. However, because this statistic is consistent with zero when Han is replaced with Ust’-Ishim, these findings cannot be driven by Basal Eurasian ancestry (as we discuss earlier), and must instead be driven by gene flow between populations related to east Asians and the ancestors of some Europeans."

Unknown said...

Ice Age article didn't test the relation between South Asian tribals and UP Europeans but D-stats are often wrong.

Unknown said...


>East Eurasian-like component, followed by Southeast Asian and Oceanian. Since this whole packpage as "ASI" appears unique to South Asians, it's likely very ancient and affinity dates before the split.

This is not quite what a pure ASI was as modern ASI heavy populations often have recent EA/SE admixture in addition to Iran N and Steppe. Although ASI could ahave picked ANE-like admixture already in paleolithic. I think Ust'-Ishim can be modeled as Oase1+MA1.

MaxT said...

@Unknown

It's irrelevant, crown eurasians split long ago. Upper Paeolithic Europeans simply reflect that ancient affinity, nothing more.

Singh said...

@Unknown
Pure ASI is basically Ancient South Eurasian, from this branch East Eurasian, Southeast, Australasian also descend as shown in Lazaridis. That is how ASI ancestry breaks down in K7, showing ancestry relations to those groups.

ASI breaks down in similar fashion in population who do not have recent East or Oceania ancestry, like Chamars, Chenchu, Velamas, Sindhis, Gujaratis etc as shown in K7 spreadsheet.

Unknown said...

Yes australoid admixture in them is probably from a shared split but it doesn't preclude potential later(mesolithic for example) Gravettian or Aurignacian admixture into Iran N and ASI. WHG seem to have it._Gene_flow_from_Paleo-ASI_into_Paleo-Europeans pre-LGM is also possible.

http://www.anthrogenica.com/archive/index.php/t-9163.html

Iran_Neolithic
"Israel_Natufian" 62.05
"AG3-MA1" 22.8
"Ust_Ishim1" 14.8
"Yoruba" 0.35
"Villabruna" 0
Fit 0.05

Villabruna
19.6% Ust_Ishim
20.5% MA1
59.8% Iberia_EN

or

7.7% Ust_Ishim
17.9% MA1
72.9% Iberia_HG
1.5% Han


BedouinB
"Vestonice16" 53.85
"Basal" 29.55
"AfontovaGora3" 11.3
"Esan_Nigeria" 5.25
"MA1" 0.05
"GoyetQ116-1" 0
"Kostenki14" 0
"ElMiron" 0
"Villabruna" 0

Unknown said...


Singh Ust'Ishim did have an east asian-like component but not as big as many south Indian tribals who have relatively recent east asian ancestry of both Tibetan and other types. I think Oase1 shows less EA and more WHG-like ancestry on admixture than Ishim despite being older.

https://genetiker.wordpress.com/2014/11/01/analyses-of-the-ust-ishim-genome/

Matt said...

@ Davidski, thanks for the stats, interesting stuff.

Looking at the chimpanzee outgroup stats as a completely neutral outgroup:

http://i.imgur.com/a5yJHU2.png

In the presence of a neutral outgroup, the strongest ancient-to-modern shift in that set is actually of modern Levant towards Asia (particularly ASI) compared to the ancient Levant.

Note that this doesn't imply that the modern Levant is closer to Asia than modern Iran. It means that the shift from ancient to modern was greater. IRC Levant_BA was more isolated from Asia than Iran_Chl, which we'd see from a set of D (Levant_BA, Iran_Chl, Asian, Chimp) stats, and also D(Druze/Lebanese, Iranian, Asian, Chimp).

(If poss. Davidski, could you run these stats to check that?: https://pastebin.com/XRFA9B8D).

The shift towards Africa by the modern Levant compared to the ancients is actually lesser than its Asian shift. Part of this is probably a change to a more Iran_Chl like ancestry.

We also see a shift by Iranians to Asian outgroups (most strongly East Asian), and a shift by Armenians towards BedouinB. Levant_BA ancestry into Iran (decreasing relatedness to Asia compared to Iran_Chl ancestry) probably offsets geneflow from Asia to some degree.

At the same time, Armenian, Iranian, Druze and Lebanese are all shifted away from Natufian.

Looking at the Luo outgroup stats to compare Asian / Natufian / Bedouin vs African shift:
http://i.imgur.com/KZIm8ZI.png

Now in these stats, the shift by Iranians away from Luo towards Han, compared to their ancient proxy, is greater than in the Levant equivalents. This is probably due to the greater African admixture in the Levant.

So hmm... I'm more convinced that flow from Asia and Africa plays some role in offsetting the ancient vs modern stats, at least in the Near East.

Surprised there's not much / any discussion of this in the scientific lit so far though, or why there even are simple models like Levant_BA+Steppe_EMBA as a model for Lebanese, in play.

Unknown said...


Matt asian shift in modern Levant is caused by CHG, Iranian, ANE, and in case of Iran mongolian admixtures. As you can see on this diagram modern Levantines have a bit of ASI too. So this makes sense.

http://img.4plebs.org/boards/pol/image/1398/88/1398888033132.png

Anthro Survey said...

Rob, I think you mixed up Iberomaurusian and Capsian.

Capsian: more eastern nucleus of culture, post-dates IM, contemporaneous w/Levant_Neo, skull shapes categorized as "proto-med".

Anthro Survey said...

Kurti:

Good point about zigurrats. There was probably some early, primitive precursor structure that differentiated into Chogna-Zanbil type in the east and into a pyramidal form in the west.

What about the rectilinear motifs we readily associate w/Messopotamian urban landscape? Do you think this was carried into Egypt on one end and into Sogdia on the other by Iran_Chalco package? Or more of a cultural diffusion thing in this case? Or independent developments?

Kurti said...

@Unknown

"ASI component more than likely predates Iran Neo as Ust'-Ishim and even Oase1 seem to be ASIs with a WHG/West Eurasian shift. "

ASI which is modeled after MODERN South Asians
can not predate Ushtishm, Mal'ta etc. ASI harbours ancestry that derives from a ancient group which shared ancestry with Ushtishm and Mal'ta.

Kurti said...

@Singh

I repeat ASI is modeled after modern South Asian population it can't predate Iran_Neolithic which is basically Iran_Mesolithic derived.

If we find ancient mesolithic South Asian that will come out as Proto ASI like than we can discuss geneflow otherwise claiming Iran_Neo_Meso has ASI ancestry is merely speculation.

Kurti said...

"Anatolia Neolihic is pretty much LevantNeo with WHG, Iranian and probably CHG admixture that acquired drift and LevantNeo is Natufian with Iranian, Anatolian and probably some WHG admixture. Anatolia Neolithic very likely has much more than just 15% of Natufian ancestry but those components are overlapping."


Again the chart shows Natufian being Anatolian_Neo admixed because Levant_Neo and ANatolian_Neo are used as proxy groups. It is impossible for MESOLITHIC Natufians to be 15% Anatolian_Neo derived. Based on this logic I am trying to explain that Iran_Neo is not slightly ASI admixed but the populations on which ASI has been modeled, have some Iran_Neo derived ancestry. And we know this is the case, even the most ASI South Asian DO have some Iran_Neo ancestry.

Singh said...

@Kurti

Iran_Neolithic does not* have ASI related ancestry, nor does Mesolithic Iran_Hotu. That is what i have been trying to tell him. ASI is very East-shifted population, in other words ENA/ASE.

ASI is basically Ancient South Eurasian-related and is older than ANE and WHG component in Eurasia. We can see this in branching charts in Lazaridis (2016) study and older Reich (2009) study.

Seinundzeit said...

Just putting a few quick thoughts out there...



If it does turn out that "ASI" is really just some form of ENA ancestry, then I highly doubt that Alborzian Mesolithic + Zargosian Neolithic peoples had any ASI.

Honestly, I just don't see any kind of ENA ancestry as having such deep penetration into prehistoric West Eurasia.

Then again, there is talk of all West Eurasians having some very serous/heavy ENA admixture (even ancient Near Eastern populations like Anatolia_Neolithic and CHG).

So, perhaps east-west contacts have always been a thing, and perhaps they have always involved substantial genetic entanglement.

Still, until we see more work in this vein, I think it would more parsimonious to assume that ancient populations on the Iranian plateau lacked ASI (again, if ASI = some sort of ENA).

Anyway, if we go by Fst-based modelling, there is substantial Ust-Ishim-related ancestry in South Indian populations, which could be a hint of things to come.

The Y-DNA data is already suggestive of the possibility that people like the Paniya might be "unique" in a general Eurasian context.

Now, if it turns out that ASI isn't merely ENA, but rather a poorly understood ghost population with complex affinities, like perhaps some sort of Paleolithic/Mesolithic Central Asian stream of populations, then I can definitely see Iran_Meso/Neo as having "ASI".

The good thing though is that we'll find out soon enough, as it seems that the Rakhgarhi paper will include Mesolithic aDNA from India. This should finally allow us to see things in proper perspective.

Singh said...

@Seinundzeit

I agree but we do know it's not related to West Eurasians from various studies, starting with Reich (2009) and to Lazaridis (2016), it's related to Ancient South Eurasian. "ASI" component found in South Asians does break down into everything found in Eastern Eurasia in K7, in other words ASE/ENA, rest of non-ASI ancestry in South Asians is mostly ANE and Basal.

Matt said...

@ Davidski, cheers.

So from those

* Using the neutral outgroup, Iran_Chalcolithic is indeed closer to Asian populations than Levant_BA, and Iran_Neolithic is closer to Asian populations than Levant_Neolithic

but

* Iran_Chalcolithic is also closer to African populations than Levant_BA, and Iran_Neolithic is also closer to African populations than Levant_Neolithic

* The magnitude of the signal is such that the differences in relatedness to Africans and Asians are greater for the Chalcolithic/BA populations than the Neolithics.

If the Bronze/Chalcolithic samples were just mixtures of the Neolithics, you'd expect the differences to go down, and be sort of intermediate. But instead they are greater.

* Levant_BA and Levant_Neolithic are, like expected, much closer to BedouinB and Natufian.

http://i.imgur.com/woNuEau.png

* Finally, comparing Iran_Chalcolithic to Armenia_Chalcolithic; Armenia is closer to Natufian / BedouinB, and to Asians (particularly North Asia); sharing with Africans is similar between Iran_Chalcolithic and Armenia_Chalcolithic is the same (neither is more or less similar).

http://i.imgur.com/2xtfijD.png
http://i.imgur.com/k6MjR6L.png

@ Davidski, to check the modern equivalents against the Metal Age, could you please run these D-stats for me if possible: https://pastebin.com/b1jrQNM9

Grey said...

MaxT

"we would still see small % of steppe-input in these mummies. They appear to be from same time period as Hyksos. I'll just go with idea that these mummies are not Hyksos for now."

Yes I agree, when I say

"southern" were mostly southern + some hyksos + a few steppe"

i mean allied groups rather than ancestry - like in later times German tribes in Dacian confederations or Celtic tribes in German confederations or Saxon mercenaries in Byzantium after the Norman conquest where you wouldn't necessarily expect to find Saxon dna in the Byzantine emperors.

Even if true it's only a minor thing in this context. My reason for being OCD about it is

"t's possible that one of those Near Eastern population could have brought horses through trade from steppe people?"

i don't think the presence of horses is the key factor - from being focused on military history I think the key factor is time in the saddle from a very early age - hence why cavalry mercenaries actually from the steppe (or other transhumance regions) were a thing until a few hundred years ago. If correct the only alternative outside regions where providing food required spending 10 hours in the saddle every day would be the aristo model where the surplus from a village of peasants subsidizes one guy to spend all his time in the saddle.

Only a minor quibble in the context of this discussion but if correct i think it explains why so much of history revolves around small mounted aristocracies.

Unknown said...


>If we find ancient mesolithic South Asian that will come out as Proto ASI like than we can discuss geneflow otherwise claiming Iran_Neo_Meso has ASI ancestry is merely speculation.

We might have had already found paleolithic ASIs as both Ust'-Ishim and Oase1 cluster right next to South Asian tribals and in fact both show a bit of Near Eastern admixture just like modern Paniya, although they also show a bit of European admixture too. Iran Neo can be modeled as Ust'-Ishim admixed.

>I agree but we do know it's not related to West Eurasians from various studies, starting with Reich (2009) and to Lazaridis (2016), it's related to Ancient South Eurasian. "ASI" component found in South Asians does break down into everything found in Eastern Eurasia in K7, in other words ASE/ENA, rest of non-ASI ancestry in South Asians is mostly ANE and Basal.

Well Ust'-Ishim and Oase1 weren't pure ASE/Papuan. Ust'-Ishim was less mixed than modern SA tribals but Oase1 was even more mixed which may look like that due his elevated Neanderthal admixture.
Paleolithic genomes tend to be very mixed due to incomplete differentiation. I think South Asians may have admixtures from other types of australoid ENAs than just Onge, Ust'-Ishim and Oase.

What's interesting is that Oase1 can be most closely modeled as Onge+Puerto Rican. Which suggests Oase already had West Eurasian components including Basal Eurasian and WHG. He also shows SSA admixture which is absent in modern tribal Adivasis and Ust'-Ishim.

Unknown said...


>So from those

* Using the neutral outgroup, Iran_Chalcolithic is indeed closer to Asian populations than Levant_BA, and Iran_Neolithic is closer to Asian populations than Levant_Neolithic

but

* Iran_Chalcolithic is also closer to African populations than Levant_BA, and Iran_Neolithic is also closer to African populations than Levant_Neolithic

* The magnitude of the signal is such that the differences in relatedness to Africans and Asians are greater for the Chalcolithic/BA populations than the Neolithics.

If the Bronze/Chalcolithic samples were just mixtures of the Neolithics, you'd expect the differences to go down, and be sort of intermediate. But instead they are greater.

* Levant_BA and Levant_Neolithic are, like expected, much closer to BedouinB and Natufian.

http://i.imgur.com/woNuEau.png

* Finally, comparing Iran_Chalcolithic to Armenia_Chalcolithic; Armenia is closer to Natufian / BedouinB, and to Asians (particularly North Asia); sharing with Africans is similar between Iran_Chalcolithic and Armenia_Chalcolithic is the same (neither is more or less similar).

http://i.imgur.com/2xtfijD.png
http://i.imgur.com/k6MjR6L.png

@ Davidski, to check the modern equivalents against the Metal Age, could you please run these D-stats for me if possible: https://pastebin.com/b1jrQNM9

Matt could you compare who had greater relatedness to Asians and Africans between Natufians and Levant N?
Also could you compare relatedness to Asians(East, Siberian, SE and South) and Africans between Natufians/Levant, El Miron/Goyet, WHG, MA1/AG2/AG3 and Aurignacians/Gravettians. Among WHG test KO1 because out of WHG he seems to have the lowest ENA admixture and Loschbour because he seems to have the highest ENA admixture.

Unknown said...


>Based on this logic I am trying to explain that Iran_Neo is not slightly ASI admixed but the populations on which ASI has been modeled, have some Iran_Neo derived ancestry. And we know this is the case, even the most ASI South Asian DO have some Iran_Neo ancestry.

Indian tribals are indeed Iran Neo admixed(not sure about Kusunda from Nepal who plots near Oase1) but this doesn't preclude Iran Neo from being ASI admixed either. They have a very strong ENA signal. Someone should make an ASI Gedmatch calculator based on Ust'Ishim and Oase1 or maybe one calculator based on just Ust'Ishim and the other based on Onge, Papuan, Ust'-Ishim, Oase1 and maybe even Melanesian and Negrito. Modern South Asians seem to have admixture from all four of them and probably a bit of Negrito and other unsampled but similar lineages.

astenb said...

I swear yall have 100's of threads to talk about Asians and the Steppes. Why here? Does all Eurasia talk always have to drown out comments regarding ancient African genomes?

Singh said...

@Unknown
"Paleolithic genomes tend to be very mixed due to incomplete differentiation."

Exactly, Paleolithic are too ancient and show affinity towards all Eurasians. That is why Lazaridis et al (2016) shows Ust'-Ishim branching off before the formation of ASE ANE WHG, Figure S11.3 in supplementary information. Yes, Ust'Ishim is indeed closer to ASE groups but branched off right before formation of ASE. In case of ASI in South Asians, it is nothing more than ASE/ENA and it breaks down as such.

Reich et al. (2009), "ASI, Proto-East-Asian and Onge split around 1,700 gens ago"

How ASI breaks down shows what Reich already found in 2009 and at K7 ASI break down as East Asia, Southeast Asia, Australasia in South Asians.

Unknown said...


>Exactly, Paleolithic are too ancient and show affinity towards all Eurasians. That is why Lazaridis et al (2016) shows Ust'-Ishim branching off before the formation of ASE ANE WHG, Figure S11.3 in supplementary information. Yes, Ust'Ishim is indeed closer to ASE groups but branched off right before formation of ASE. In case of ASI in South Asians, it is nothing more than ASE/ENA and it breaks down as such.
Reich et al. (2009), "ASI, Proto-East-Asian and Onge split around 1,700 gens ago"
How ASI breaks down shows what Reich already found in 2009 and at K7 ASI break down as East Asia, Southeast Asia, Australasia in South Asians.

They weren't fully differentiated but that doesn't mean they weren't differentiated at all.

When you break down Tianyuan, Oase and Ishim you can see they aren't identical. Besides ASI probably changed its allele frequencies since paleolithic and modern tribes have additional admixtures.

https://genetiker.files.wordpress.com/2016/08/all-13-14-1.png

http://www.eupedia.com/forum/archive/index.php/t-30619.html

http://www.anthrogenica.com/archive/index.php/t-9163.html

http://www.eupedia.com/forum/archive/index.php/t-30571.html

http://www.anthrogenica.com/archive/index.php/t-9874.html

Matt said...

@ Unknown, you'd have to get Davidski to run the stats. I think you probably want:


Levant_N,Israel_Natufian,Esan_Nigeria,Chimp
Levant_N,Israel_Natufian,Han,Chimp
Levant_N,Israel_Natufian,Kharia,Chimp
Levant_N,Israel_Natufian,Luo,Chimp
Levant_N,Israel_Natufian,Mala,Chimp
Levant_N,Israel_Natufian,Nganasan,Chimp
Levant_N,Israel_Natufian,Yoruba,Chimp

You'd have to get Davidski to run these off for you. He's usually pretty obliging if he notices the post.

You might also be interested in
(Iran_Hotu,Israel_Natufian,Esan_Nigeria,Chimp
Iran_Hotu,Israel_Natufian,Han,Chimp
Iran_Hotu,Israel_Natufian,Kharia,Chimp
Iran_Hotu,Israel_Natufian,Luo,Chimp
Iran_Hotu,Israel_Natufian,Mala,Chimp
Iran_Hotu,Israel_Natufian,Nganasan,Chimp
Iran_Hotu,Israel_Natufian,Yoruba,Chimp

and

Iran_Neolithic,Iran_Hotu,Esan_Nigeria,Chimp
Iran_Neolithic,Iran_Hotu,Han,Chimp
Iran_Neolithic,Iran_Hotu,Kharia,Chimp
Iran_Neolithic,Iran_Hotu,Luo,Chimp
Iran_Neolithic,Iran_Hotu,Mala,Chimp
Iran_Neolithic,Iran_Hotu,Nganasan,Chimp
Iran_Neolithic,Iran_Hotu,Yoruba,Chimp)

If you've got a bunch, I'd advise put them in a PasteBin, like https://pastebin.com/b1jrQNM9 from a few posts up.

batman said...

Then the Glacial Maximum -
25-12.000 yrs ago - decimated the human populations of Europe and northern Asia to a terminal minimum.

During the last part of this unparralelled cold-peaks on the northern hemisphere 2/3 of all the land-animals died out - completly, known as a "mass-extiction event".

The few mammalspecies that survived the peak of ice-time where also very small and few, known as "refugia".

It seems that the basal Eurasian genome went through a very critical phase at the end og ice-time - as all known settlements from paleolithic Eurasia gets abandonned - no later than 13.000 BP.

First by 12.000 BP does the climate start to improve - allowing a handfull of Eurasian mwn to syart multiplying - defining all the y-lines that spread throughout the depopulated continents of Europe and Northern Asia.

Which means that the y-lines from paleolithic Eurasia do not have any post-glacial, present descendants. Exept for one or two.

So far only one refugia is known to have existed, as well as reproducing ancestral lines to ptesent Eurasians...

Unknown said...


Levant Neolithic is very African and Han shifted compared to Natufians. Levant Neo is comparable to CHG whereas among tested Natufians are second most differentiated from Yoruba after WHG.
http://img.4plebs.org/boards/pol/image/1496/55/1496555680686.png

europeuplement said...

Can be that Yoruba has a little mixture coming from Levant dating from Chalcolithique (Levant_N + Iran_N or CHG) which migrated by the Sahara (Pastoral Period of Sahara). Hamitic migrations.

The DNA of Fulani would be very interesting.

Unknown said...


>Can be that Yoruba has a little mixture coming from Levant dating from Chalcolithique (Levant_N + Iran_N or CHG) which migrated by the Sahara (Pastoral Period of Sahara). Hamitic migrations.
The DNA of Fulani would be very interesting.

Yoruba don't have any detectable Near Eastern admixture. They are 100% negroids as far as we know. And even the Horners have Natufian-like admixture with little to no EEF/CHG/IranN whereas Neolithic Levantines had EEF and Iranian(this is likely the reason why they are closer) admixtures.