First Genomes from Ancient Egypt

[Angela]

Now, Levant Neolithic and BA samples, plus modern Lebanese, Palestinian, Bedouin and Egyptian.

M115616	I0867		M291439	I1706		Modern			Modern			Modern			Modern
Levant Neolithic		Levant BA		Lebanese		Palestinian		Bedouin			Egyptian
Run time	9.93		Run time	13		Run time			Run time			Run time			Run time
S-Indian	-		S-Indian	0.26		S-Indian	1		S-Indian	1		S-Indian	0		S-Indian	1
Baloch	-		Baloch	3.57		Baloch	11		Baloch	7		Baloch	5		Baloch	3
Caucasian	25.97		Caucasian	37.26		Caucasian	41		Caucasian	39		Caucasian	21		Caucasian	28
NE-Euro	-		NE-Euro	-		NE-Euro	3		NE-Euro	1		NE-Euro	2		NE-Euro	1
SE-Asian	0.07		SE-Asian	0.62		SE-Asian	1		SE-Asian	0		SE-Asian	0		SE-Asian	0
Siberian	-		Siberian	-		Siberian	1		Siberian	1		Siberian	0		Siberian	0
NE-Asian	0.06		NE-Asian	0.43		NE-Asian	1		NE-Asian	0		NE-Asian	0		NE-Asian	0
Papuan	-		Papuan	-		Papuan	0		Papuan	0		Papuan	0		Papuan	0
American	-		American	0.34		American	0		American	0		American	0		American	0
Beringian	-		Beringian	-		Beringian	0		Beringian	0		Beringian	0		Beringian	0
Mediterranean	32.53		Mediterranean	12.01		Mediterranean	13		Mediterranean	12		Mediterranean	7		Mediterranean	17
SW-Asian	39.86		SW-Asian	44.73		SW-Asian	23		SW-Asian	31		SW-Asian	56		SW-Asian	33
San	-		San	-		San	0		San	0		San	0		San	0
E-African	1.52		E-African	-		E-African	3		E-African	5		E-African	5		E-African	12
Pygmy	-		Pygmy	0.4		Pygmy	0		Pygmy	0		Pygmy	0		Pygmy	0
W-African	-		W-African	0.38		W-African	1		W-African	1		W-African	3		W-African	6

It's good to remind people that Levant Neolithic is not the same as Natufian. The transition brought more "Caucasus" and Med and decreased the SWAsian. I think that's admixture with Anatolian Neolithic, which shows up in the modeling.

Then in the Levant Bronze, Caucasian went up, which is in line with what these papers have been talking about, Baloch appears, but SW Asian also went up. Perhaps there was more movement from Arabia north? Virtually no SSA in the Levant Bronze Age, however, so whatever brought the additional SW Asian didn't carry it. Both the East African and West African came after that.

For SWAsian, it's 23 to 31, Lebanese to Palestinian. Are these Christian Lebanese? If they aren't, the differences might be larger.

Well, you can see not only the increase in the East African in modern Egyptians, but also the increase in West African.


Amazing how low the Med component drops. Anatolia Neolithic was 46% Med. So, I guess since Med is the Sardinian cluster, it's mostly Anatolian Neolithic plus WHG?

 

[LeBrok]

It's good to remind people that Levant Neolithic is not the same as Natufian. The transition brought more "Caucasus" and Med and decreased the SWAsian. I think that's admixture with Anatolian Neolithic, which shows up in the modeling.

Then in the Levant Bronze, Caucasian went up, which is in line with what these papers have been talking about, Baloch appears, but SW Asian also went up. Perhaps there was more movement from Arabia north? Virtually no SSA in the Levant Bronze Age, however, so whatever brought the additional SW Asian didn't carry it. Both the East African and West African came after that.

For SWAsian, it's 23 to 31, Lebanese to Palestinian. Are these Christian Lebanese? If they aren't, the differences might be larger.

Well, you can see not only the increase in the East African in modern Egyptians, but also the increase in West African.


Amazing how low the Med component drops. Anatolia Neolithic was 46% Med. So, I guess since Med is the Sardinian cluster, it's mostly Anatolian Neolithic plus WHG?

Palestinians are almost true BA Levant, plus 5 percent of baloch and 6 percent of SSA. Amazing. I'm betting that ancient Jews will turn very similar and most likely Phoenicians too.

Here is Anatolian and EEF, BA Italian and Sardinian

M54279	I0746		M405327	I1506 NE1		Remedello Average		Modern from Harappa table
Anatolian EF		Hungary, Polgár-Ferenci-hát	7.2kya		Bronze Age (Neolithic Genome)	Sardinian
Run time	10.2		Run time	19.95		Run time			Run time
S-Indian	-		S-Indian	-		S-Indian	-		S-Indian
Baloch	-		Baloch	-		Baloch	-		Baloch
Caucasian	35.9		Caucasian	28.27		Caucasian	11.03		Caucasian	20
NE-Euro	3.91		NE-Euro	12.13		NE-Euro	21.25		NE-Euro	13
SE-Asian	-		SE-Asian	-		SE-Asian	0.61		SE-Asian
Siberian	-		Siberian	-		Siberian	-		Siberian
NE-Asian	-		NE-Asian	-		NE-Asian	-		NE-Asian
Papuan	-		Papuan	-		Papuan	-		Papuan
American	-		American	-		American	-		American
Beringian	-		Beringian	-		Beringian	-		Beringian
Mediterranean	46.12		Mediterranean	45.75		Mediterranean	60.61		Mediterranean	60
SW-Asian	14.03		SW-Asian	13.45		SW-Asian	5.50		SW-Asian	7
San	-		San	-		San	-		San
E-African	-		E-African	-		E-African	-		E-African
Pygmy	-		Pygmy	0.05		Pygmy	0.08		Pygmy
W-African	-		W-African	0.35		W-African	0.92		W-African

 

[bicicleur 2]

Palestinians are almost true BA Levant, plus 5 percent of baloch and 6 percent of SSA. Amazing. I'm betting that ancient Jews will turn very similar and most likely Phoenicians too.

so the Palestinians are not the Filistines, the Sea People that settled in Gaza

maybe the Jews and the Phoenicians have common ancestors, but that is then from before they entered in history
Phoenicians are from the Lebanese coast, often a place for refugees from inland, trying to escape from the domination of the Hittites or the Egyptians
and the origin of the Jews are marginal herders in the hills in the interface between the Southern Levant and the Negev desert who expanded into the vacuum created by the Egyptians when they abandonned the Levant

 

[harena]

Anatolian Neolithic was surprisingly more distinct than just WHG admixture. IIRC Anatolian Neolithic was 10% WHG, 20 Natufian and the rest (70%) their own stock, Anatolian farmers.

Well, they are related by ancient relatives belonging to 3 major admixtures/groups, though in different proportions, and they have drifted way through thousands of years of separation. Here are their genomes in HarappaWorld GedMatch run.

M041601	Merged		M54279	I0746
Natufian			Anatolian EF
Run time	6.39		Run time	10.20
S-Indian	-		S-Indian	-
Baloch	-		Baloch	-
Caucasian	13.98		Caucasian	35.90
NE-Euro	-		NE-Euro	3.91
SE-Asian	-		SE-Asian	-
Siberian	-		Siberian	-
NE-Asian	-		NE-Asian	-
Papuan	0.68		Papuan	-
American	-		American	-
Beringian	-		Beringian	-
Mediterranean	27.39		Mediterranean	46.12
SW-Asian	53.62		SW-Asian	14.03
San	-		San	-
E-African	4.33		E-African	-
Pygmy	-		Pygmy	-
W-African	-		W-African	-

They have mostly similar admixtures (living relatively close by), but in so different proportions that it makes them very distinct.

I'm not too sure about backward modeling very ancient samples like that, some of those components are partly derived from Natufian/Anatolian Farmers not the other way around.
IIRC Lazaridis modeled Natufians roughly as 50% Basal Eurasian and 50% WHG; Levant Neolithic, Anatolian farmers, EEF and Copper Age Iberia are all on the same cline, just look at a PCA.

 

[Fire Haired14]

Post at my blog about the new ancient Egyptian mtDNA results: First look at ancient Egyptian mtDNA

mtDNA doesn't get anymore SouthWest Asian than the ancient Egyptian's mtDNA. There isn't a lot of East African mtDNA in my database. The stuff I've read about East African mtDNA indicates their West Eurasian mtDNA shares a lot of similarities.

R0a, HV1, T1a, J2a2, N1, M1a might all ultimately derive from Natufian-like people. N1, T1a, J2a link Neolithic Anatolians with Natufian-rich people.

Think about this. Andronovo, Bronze age British, and these ancient Egyptians were roughly contemporary to each other. Each carries roughly the same frequency of T1a and I as the ancient Egyptians. That demonstrates the widespread distribution of Neolithic West Asian ancestry.

 

[bicicleur 2]

the calls for Y-DNA of the 3 males :

https://genetiker.wordpress.com/2017/06/01/y-snp-calls-from-ancient-egypt/

 

[Fire Haired14]

This paper's ADMIXTURE analysis is maybe the most interesting aspect of the paper. It's 100% consistent with results I and others have gotten using D-stats provided by David Wesoloski at Eurogenes. Recall that the ADMIXTURE isolated Natufian, AnatoliaNeolthic, CHG, and Europe HG centered components.

Here are some interesting details the ADMIXTURE analysis shows...

-Saami's European-side has more Euro_HG(WHG, EHG) ancestry than any Europeans including Lithuanians and Finns. D-stats indicate they have 10-15%(exact percentages might be wrong) Scandinavian Hunter Gatherer ancestry which was rich in EHG and also had its own unique alleles. D-stats and this paper's ADMIXTURE give Saami roughly 40% EuroHG ancestry, 20% CHG ancestry, 15% AnatoliaNeo ancestry, and 25% something East Asian-like.

-Southern Italians score as much in the Natufian component as Iranians which indicates their Near Eastern ancestors lived south and west of Iran.

-BedouinB might be mostly Natufian(ish). They score 70% in the Natufian component, 20% in the CHG component, and 20% in the AnatoliaN component.

-YemaniteJew, ancient Egyptians, and Jordan_EBA are all really similar to each other. They score 50% in the Natufian component, 20-30% in CHG, and 20-30% in AnatoliaN. All modern SouthWest Asians score an extra dose in CHG(30-40%) and significantly less in Natufian(35-40%).

-The Near Eastern ancestor of North African Jews and European Jews was probably similar to the average modern Levantie. The Near Eastern ancestor of Yemanite Jews was probably similar to Jordan_EBA and ancient Egyptians.

 

[MarkoZ]

The site where the samples come from have no record of Hyksos or Greek settlements + they are burried in typical Egyptian fashion. There is absolutely no way that they were mixed with Hyksos. They are Iran_Neo admixed because the ancient Egyptian derive most likely from late-Neolithic or Bronze Age Levant. This would explain why ancient Egytpain language is linguistically closer to Semitic than Berber. It is because it left the South Levant as a secondary wave of Afro_Asiatic speakers. After the Proto Berbers left earlier.

However the predynastic sites that look like the best candidates for the Protoegyptians (the Naqada horizon) are rather concentrated in Upper Egypt. I guess only ancient DNA will tell, but in my mind that makes an origin from the Levant unlikely.

 

[bicicleur 2]

However the predynastic sites that look like the best candidates for the Protoegyptians (the Naqada horizon) are rather concentrated in Upper Egypt. I guess only ancient DNA will tell, but in my mind that makes an origin from the Levant unlikely.

prior to Naqada was the 8.2 ka climate event, which attracted all kind of herders and farmers from SW Asia to the Nile Delta and beyond, into Northern Africa
after the 8.2 ka event, the Sahara became 'green again' till 5.9 ka when the Sahara desert expanded again, which drove many herders & farmers back into the Nile Valley
the Protoegyptians probable were in Northern Africa since 8.2 ka and their origin was SW Asia
of course, later chalcolithic influxes are also possible and likely

 

[MarkoZ]

prior to Naqada was the 8.2 ka climate event, which attracted all kind of herders and farmers from SW Asia to the Nile Delta and beyond, into Northern Africa
after the 8.2 ka event, the Sahara became 'green again' till 5.9 ka when the Sahara desert expanded again, which drove many herders & farmers back into the Nile Valley
the Protoegyptians probable were in Northern Africa since 8.2 ka and their origin was SW Asia
of course, later chalcolithic influxes are also possible and likely

While I think that's possible, the most commonly proposed origin for Afrasian seems to be in the pre-neolithic Eastern Sahara and the Horn of Africa.

 

[davef]

Because Natufians, who makeup most of that SW Asian component in the paper, might be closely related to Anatolia Neolithic.

Sorry, but if you go by Hauteville's charts posted in the first page of this thread, you'll find that Natufians aren't anywhere near Anatolian Neolithic farmers. They are way way different. They lack that "dark blue" component. Natufians were pretty much Bedouins if I recall. Leagues away from Anatolian.

 

[bicicleur 2]

While I think that's possible, the most commonly proposed origin for Afrasian seems to be in the pre-neolithic Eastern Sahara and the Horn of Africa.

that is what I tought also, untill the Y-DNA of the Natufians was published about a year ago
since then, I'm pretty convinced that the Afroasiatic languages originated in the Levant along with haplo E1b1b1 and also spread along with this clade out of the Levant

as for the 8.2 ka event : before 8.2 ka the 'green Sahara' was full of HG, during the 8.2 ka event, which lasted a few centuries, the Sahara was empty, and after the event, the 'green Sahara' was full of herders

whatever happened in the Nile delta at that time, nobody knows because everything is burried under very deep layers of sediments

 

[Aaron1981]

From the graphic posted I don't see that at all, unless you're coining your own definitions for population groups. Northeast Africans are Horners.

The only "Africans" the ancient Egyptians from this period plot close to are North Africans, who are mostly Levantines with SSA, more than half of it coming in the last 2000 years, presumably through the Arab slave trade, plus whatever traces of the prior population(s) are left. None of that is a surprise.

Anyone who has been paying attention to the papers knows that two major population flows moved out of the Near East, the western farmers, and then a few thousand years later a population related to the Iranian farmers. Both spread over vast distances and mixed with earlier populations where ever they went. The first group spread all along the southern coast of the Mediterranean, among many other places, and even deep into Africa. We can see the traces of the second major Near East group in northern Africa too, but in much smaller percentages.

Basically, as LeBrok pointed out, the Bronze Age may be the last really major population upheaval in western Eurasia. As I stated in another thread, the second gene flow out of the Near East, which was less consequential for most of Europe was like a pincer movement into Europe, I believe, with part of it going over the Caucasus and onto the steppe, and part of it going into southeastern and southern Europe, as well as all over the Near East, and some of it even reaching North Africa.

It looks like a modified version of the old Dienekes theory of the Womb of Nations to me, but as you have to consider also the western farmers, it's not just the Caucasus area, but the Anatolia/Levant region as well.

While related, the admixture in northern Europe is from mesolithic age CHG and doesn't appear male mediated, the admixture event in the western Middle East is Iran_CHL/Iran_NEO, which is a distant, younger relative. If you try to model north Europeans as Iran_CHL/NEO, it won't jive.

Let's face it, until J1, J2, G2...etc start showing up in droves on the neolithic steppe, I will still consider CHG admixture in northern Europe as female driven. We know R1 was originally WHG/EHG along with I.

 

[MarkoZ]

While related, the admixture in northern Europe is from mesolithic age CHG and doesn't appear male mediated, the admixture event in the western Middle East is Iran_CHL/Iran_NEO, which is a distant, younger relative. If you try to model north Europeans as Iran_CHL/NEO, it won't jive.

Let's face it, until J1, J2, G2...etc start showing up in droves on the neolithic steppe, I will still consider CHG admixture in northern Europe as female driven. We know R1 was originally WHG/EHG along with I.

You might want to back that up, because Iran_ChL/EHG is exactly the model Lazaridis proposed for Yamnaya.

R1 wasn't originally WHG-EHG for god's sake. Stop being silly.

 

[MarkoZ]

since then, I'm pretty convinced that the Afroasiatic languages originated in the Levant along with haplo E1b1b1 and also spread along with this clade out of the Levant

I guess more samples are needed. Though E-M35 should be around thousands of years before Natufian.

 

[kingjohn]

no mtdna h3 was found in the ancient egyptions
it looks like it originated or in iberia or in morocco/ algeria area ...

 

[LeBrok]

I'm not too sure about backward modeling very ancient samples like that, some of those components are partly derived from Natufian/Anatolian Farmers not the other way around.

I wish someone has finally made a calculator, strictly based on these ancient samples of 3 farmer groups (or better the h-gs they came from) and 3 h-gs groups. Sort of gold standard, based on samples from 10 kya.

IIRC Lazaridis modeled Natufians roughly as 50% Basal Eurasian and 50% WHG;

Relly?! Could you link me to this fact as I don't seem to remember it. One impossible thing to overcome is that genesis of Natufians starts at the same time as genesis of WHG about 14 kya, and they plot way away on PCA. THis is the first that I hear that WHG was an ancestor of Natufian, and in 50% level. There is some degree of "immediate" common ancestry, but not 50% and not directly WHG!

Levant Neolithic, Anatolian farmers, EEF and Copper Age Iberia are all on the same cline, just look at a PCA.

Sure they are on the same cline, as I said that they have big degree of same ancestry, but you wouldn't ever say that Iberia Copper is the same as Levant Neolithic. At least half of their genome is different from each other and drifted 5 thousand years apart, so even having closely related genes they developed different alleles, mutations.

 

[Angela]

Palestinians are almost true BA Levant, plus 5 percent of baloch and 6 percent of SSA. Amazing. I'm betting that ancient Jews will turn very similar and most likely Phoenicians too.

Here is Anatolian and EEF, BA Italian and Sardinian

M54279	I0746		M405327	I1506 NE1		Remedello Average		Modern from Harappa table
Anatolian EF		Hungary, Polgár-Ferenci-hát	7.2kya		Bronze Age (Neolithic Genome)	Sardinian
Run time	10.2		Run time	19.95		Run time			Run time
S-Indian	-		S-Indian	-		S-Indian	-		S-Indian
Baloch	-		Baloch	-		Baloch	-		Baloch
Caucasian	35.9		Caucasian	28.27		Caucasian	11.03		Caucasian	20
NE-Euro	3.91		NE-Euro	12.13		NE-Euro	21.25		NE-Euro	13
SE-Asian	-		SE-Asian	-		SE-Asian	0.61		SE-Asian
Siberian	-		Siberian	-		Siberian	-		Siberian
NE-Asian	-		NE-Asian	-		NE-Asian	-		NE-Asian
Papuan	-		Papuan	-		Papuan	-		Papuan
American	-		American	-		American	-		American
Beringian	-		Beringian	-		Beringian	-		Beringian
Mediterranean	46.12		Mediterranean	45.75		Mediterranean	60.61		Mediterranean	60
SW-Asian	14.03		SW-Asian	13.45		SW-Asian	5.50		SW-Asian	7
San	-		San	-		San	-		San
E-African	-		E-African	-		E-African	-		E-African
Pygmy	-		Pygmy	0.05		Pygmy	0.08		Pygmy
W-African	-		W-African	0.35		W-African	0.92		W-African

Remedello isn't Bronze Age. It's basically just Middle Neolithic, although they already had copper.

In your prior post, interesting that the 6 point increase in SSA, plus a few other minor changes, lowered the SW Asian in Palestinians by 8 points.

 

[bicicleur 2]

afaik the Natufians were not 100 % proper Natufian, they had some EEF admixture in them
I would also be surprised that they were 100 % E1b1, all samples we have are from 1 single site and about the same age.
I wouldn't be surprised to find some haplo G2 (but not G2a2) in Natufians, and some H2.

 

[Angela]

While related, the admixture in northern Europe is from mesolithic age CHG and doesn't appear male mediated, the admixture event in the western Middle East is Iran_CHL/Iran_NEO, which is a distant, younger relative. If you try to model north Europeans as Iran_CHL/NEO, it won't jive.

Let's face it, until J1, J2, G2...etc start showing up in droves on the neolithic steppe, I will still consider CHG admixture in northern Europe as female driven. We know R1 was originally WHG/EHG along with I.

I don't know if it was female or male driven. Does it change the fact that the gene flow occurred? Does it somehow not count if it was female driven? I think not.

I don't have an agenda or a paper or a book or thousands of prior posts to defend. I'm just following the data. It's possible there was bride exchange at the edges into a very lightly populated steppe. However, a lot, if not most, of the mtdna looks standard northern. How could that amount of "southern" mtdna have led to people who were roughly half "southern"? Maybe a modified version of Maciamo's theory is correct. I always thought it was a possibility. I'm content to wait and see what the dna shows.

As for the mixing agent being "Caucasus", color me skeptical that this population survived in unadmixed form thousands of years after those ancient samples. Plus, as Marko pointed out, the Reich Lab has modeled the mixing agent as something resembling Iran Chl. As Lazaridis intelligently pointed out, they may find a population which fits better, and it may not have actually come from Iran.

Honestly, it's as if there's a phobia with connecting anything with Iran. I suppose I don't completely get the subtext.