"MIDDLE_EASTERN", a Natufian zombie?

[Fire Haired14]

With only ADMIXTURE and modern DNA genome-bloggers 4-7 years ago isolated components mostly made up of CHG or Iran Neolithic Natufian or EEF or WHG alleles. That's impressive and can be used to trace such ancestry in modern populations.

In my opinon Eurogenes isolated a component which tracks Natufian-like ancestry accuratly. Specfically these Eurogenes tests did.

Eurogenes EUtest: MIDDLE_EASTERN
Eurogenes K9: Southwest Asian.
Eurogenes K10: Southwest Asian.
Eurogenes K11: Southwest Asian.
Eurogenes K12: Southwest Asian.
Eurogenes Hunter gatherer_Farmer: Middle Eastern Herder

All except the EUtest do not provide spreadsheets with results for modern populations. Below are ancient and modern EUtest MIDDLE_EASTERN scores.....

Eurogenes EUtest: MIDDLE_EASTERN

Ancient Pops
Natufian(# M041601) scores 69%, the other 31% is in WEST_MED and EAST_MED.
LevantNeolithic(# M115616) scores 45%, the other 55% is in WEST_MED and EAST_MED.
BarcinNeolithic(# M936428) scores only 6%.
Middle Neolithic Europe scores 0-4%.
European HGs, Iran Neolithic, and CHG score 0%.
BarcinChalcolithic, Armenia Metal Ages score 0%. IranChalcolithic scores 10%, JordanBronze age scores 44%.

Modern Pops.
Middle East: Bedouin at 58% and Ethiopians at 42% and NorthWest Africa at 30-40%. It ranges from 20-30% in the Levant. Assyrians score 13%, Iranians and Kurds score about 10%. Turkish score 8%, Caucasians score 1-3%.
Europe: Peaks in South Italy at 12%, ranges from 6-8% in the rest of Italy. Sardinia scores 0%. I assume the Southern Balkans scores similarly to Italy but no results for the Southern Balkans are aviable. Spain, France, and South Slavs score about 2%. The rest of Europe receives negligible scores(0-1%).

The correlation is pretty good. In most ADMIXTURE tests Neolithic Europeans score big in the "East Med" or "SouthWest Asian" components because it carries a lot of EEF-type stuff but many Eurogenes tests were able to isolate a Southwest Asian component that Neolithic Europeans score low in.

 

[Blanco]

Middle Eastern uses some Israeli Bedouins as reference group, instead of Natufian.

I know the public gedmatch of an Israeli Bedouin which scores 100% Middle Eastern on most calculators.

Z886762 Bedouin from Israel

By the way the reason why Ancient Natufians and Neolithic Levantines seem district from modern Near Eastern populations is because they lack the Iran Neolithic and Sub Saharan admixture which has increased in the Modern Middle East.
There's a tendency as a geneflow from Central-South Asia pushed the Natufians and Neolithic Anatolians towards Arabia, North Africa and the Mediterranean regions.
(Maybe it was because of the long history of Persian/Scythian domination)

 

[Fire Haired14]

Middle Eastern uses some Israeli Bedouins as reference group, instead of Natufian.

Did the creator of the test tell you this? Anyways just because the EUtest's MIDDLE_EASTERN component carries Bedouin-specfic stuff doesn't mean it can't also be mostly Natufianish. The same is true for West_Med. It might carry Sardinian-specfic stuff but is still mostly Neolithic Anatolian.

I know the public gedmatch of an Israeli Bedouin which scores 100% Middle Eastern on most calculators.

The Middle Eastern/SouthWest Asian/Red Sea/East Med components in other tests don't isolate a Natufian-specfic ancestry as well as the "MIDDLE_EASTERN" component in the EUtest.

Z886762 Bedouin from Israel[/QUOTE]

Here's what he scores in Eurogenes EUtest.

WEST_MED 3.39
EAST_MED 30.90
WEST_ASIAN -
MIDDLE_EASTERN 63.94
SOUTH_ASIAN -
EAST_AFRICAN 1.27
EAST_ASIAN 0.50

That's what other Bedouin score. Natufians score slightly higher 69% MIDDLE_Eastern. Of course MIDDLE_EASTERN is admixed but there's a good chance it is mostly Natufianish.

 

[Angela]

There's a good chance...might carry...and on and on. This is not scientific or mathematical analysis it's just more speculation.

You want to know how much "Natufian" is in certain groups? Use the Natufian genome at least.

Don't people have anything better to do than play around with numbers which aren't going to be accurate?

 

[Fire Haired14]

There's a good chance...might carry...and on and on. This is not scientific or mathematical analysis it's just more speculation.

It's better than nothing.

Don't people have anything better to do than play around with numbers which aren't going to be accurate?

Bet you a million dollars these numbers are accurate; Natufinish ancestry peaks in Arabia, East Africa, and North Africa. If I didn't "play with numbers" I wouldn't have found a pretty reliable detector of Natufianish ancestry. Don't know why you a moderator are always so hostile to posters.....

 

[Blanco]

Eurogenes calculators were created for people of mainly Northern European descent, the less European like ancestry someone has the more rubbish the calculator gets.
I very much doubt these calculators would give accurate answers to estimate each person's ancestry, especially when it comes to ancient genome testing.
We should realize most of these calculators were made in a room by DNA bloggers who often has their own bias and lack scientific objectivity.
I remember how some blogger (i don't wanna reveal his identity nor company) got so angry that his own ethnicity scored East Asian and blamed the inaccuracy of the calculator. The same way i guess calculators which were made by Turks or any other ethnicity will always be favorable to those certain ethnic groups over anything else.

 

[Fire Haired14]

Eurogenes calculators were created for people of mainly Northern European descent, the less European like ancestry someone has the more rubbish the calculator gets.

His tests focus on how NorthWest and NorthEast Europe are distinct. That's why he created the only ADMIXTURE tests with "North Sea" and "Baltic" and "East Euro." His focus on Europe doesn't result in poor quality tests for Middle Easterners at all. There's no reason to believe that's the case.

I very much doubt these calculators would give accurate answers to estimate each person's ancestry, especially when it comes to ancient genome testing.

We should realize most of these calculators were made in a room by DNA bloggers who often has their own bias and lack scientific objectivity.

All of those bloggers come out with tests which basically say the same thing about West Eurasian diversity; North Europe(Lithuania), West Med(Sardinia), West Asia(Georgia), SouthWest Asia(Bedouin or Saudi Arabia). Some even broke up "West Asia" into Gedorsia(Balochi) and Caucasus(Georgia) components. ADMIXTURE tests from literature come out with the same results. These components aren't the result of biases they're the result of the ADMIXTURE tool isolating the main types of ancestry in West Eurasia.

Those ADMIXTURE West Eurasian components, which were created before we had more than 1-3 ancient genomes, have been proven to be accurate by ancient genomes. North Europe is made up of mostly WHG and ANE alleles, West Med mostly Anatolia Neolithic, West Asia mostly CHG and Iran Neo, and SouthWest Asia mostly Levant Neolithic/Natufian.

EUtest's "MIDDLE_EASTERN" component is detecting a type of ancestry unique to Arabia and North Africa which has high affinity to Natufians. It isn't the result of David Wesoloski's bias.

 

[Fire Haired14]

0 out of 2 members found this post helpful.

Don't people have anything better to do than play around with numbers which aren't going to be accurate?

Is this how Eupedia reacts to people who want to have a discussion about genetics in the genetics section? You guys need to cheer up.

 

[LeBrok]

Is this how Eupedia reacts to people who want to have a discussion about genetics in the genetics section? You guys need to cheer up.

Just 2 members don't like it. The rest of people think about what you said. Isn't it good?

 

[Fire Haired14]

All the grouchy naysayers

Most ADMIXTURE tests at GEDmatch include a component which peaks in Arabia, East Africa, and North Africa. These components are definitely tracking Natufian-like ancestry and hence are good ways to track the distribution of Natufian-like ancestry in modern populations. Natuyfians score extremly high in "SouthWest Asia" components just like how Mesolithic Europeans score 100% in "North Europe" components, like how CHG scores almost 100% in "West Asia" components, and like how Anatolia Neolithic scores high in "West Med" components.

There shouldn't even be a debate here. If an ancient genome scores close to 100% in a single component it's most closely related to whatever modern population has the most of that component and modern scores in that component can probably track the distribution of ancestry from that ancient genome's population.

Apparently ancient Egyptians score high in a Natufian-like component, see here.

[Ancient Egyptians] cluster w. Neol & Br Age Levant. STRUCTURE: important Natufian component, some Anatolian, Iran Neol

The Jordan Bronze age genomes and single Middle Eastern genome from Iron age England are probably mostly Natufian as well. The modern population most similar to them are Yemanite_Jews. The Levant ancestry of other dispersal Jews looks more like modern Levanties than the Bronze age Jordan genomes, maybe Yemanite Jews traveled to Yeman at a time when the Levant was more Natufian-like and less Iran Neolithic-like.

 

[LeBrok]

There shouldn't even be a debate here. If an ancient genome scores close to 100% in a single component it's most closely related to whatever modern population has the most of that component and modern scores in that component can probably track the distribution of ancestry from that ancient genome's population.
.

Not necessarily. It might also mean that ancient Natufians and Saudis for example had same ancestors, therefore Saudis are not necessarily Natufian's descendents.

For example, SHG scores 90% on NE Euro. The most of all. But they are not ancestors of Slavic people which are high in NE Euro. The true ancestor who gave it to both was Villabruna dude. The first true WHG. Well, the biggest donor lets say, because some could have come from ANE type ancestor from central Asia. On top of it, some additional portion of NE Euro in SHG, could have been created in SHG themselves. But this is getting too complicated.

 

[Angela]

It's better than nothing.



Bet you a million dollars these numbers are accurate; Natufinish ancestry peaks in Arabia, East Africa, and North Africa. If I didn't "play with numbers" I wouldn't have found a pretty reliable detector of Natufianish ancestry. Don't know why you a moderator are always so hostile to posters.....

The question is not whether it's better than nothing. The question is whether it's the best we can do, and the answer to that is no, it's not.

It's a wonder how you so consistently tell us things we already know and have discussed, and then expect us to be amazed at your brilliance.

Yes, a southwest Asian component probably has a lot of "Natufian" like alleles, whether the component is from your "hero" Eurogenes, or from any of the other calculator creators, all the way back to Dienekes. Yes, the Bronze Age Levant samples probably have a good amount of it, as do the Bedouin and Yemeni, and yes it probably went into Africa. WE KNOW ALL OF THIS. We've discussed it extensively.

THAT WASN'T MY POINT. My point was that these modern "clusters" have a lot more than "Natufian" in them, and in addition they're highly drifted. That's why it's not a good idea to use modern clusters created by Admixture as proxies for ancient population groups unless you're aware, and make your readers aware, of the limitations of doing that, and that the percentages should only be used to provide hints, not as some hard and fast "fact".

For goodness' sakes, do I really have to explain why it's better to use ancient genomes for something like this?

It's not enough to understand math, you know. In both a career and life in general you have to be able to understand language and be able to read and understand what you've read. That means stop and think about what you've read. Stop being careless and sloppy when reaching conclusions. You seem to have difficulty with this and it's going to cause you big problems if you don't work on your skills in those areas.

 

[Fire Haired14]

Angela you clearly have an attitude issue that you need to work on. You maliciously attack every thread I make. I guarantee you almost all the claims I make are correct. You get upset not because my arguments are poor but because my evidence doesn't pass every little picky qualification you think evidence must pass. You then explode, attack me as if I insulted you or said something offensive. I just post to discuss ideas there's no reason to get upset.

The question is whether it's the best we can do, and the answer to that is no.

That isn't the question!! You don't decide what the question is. I posted this thread in order to show what I think is the most Natufian-like component and which I think can give us a general idea about the distribution of Natufian-like ancestry.

It's a wonder how you so consistently tell us things we already know and have discussed, and then expect us to be amazed at your brilliance.

Yes, a southwest Asian component probably has a lot of "Natufian" like alleles, whether the component is from your "hero" Eurogenes, or from any of the other calculator creators, all the way back to Dienekes. Yes, the Bronze Age Levant samples probably have a good amount of it, as do the Bedouin and Yemeni, and yes it probably went into Africa. WE KNOW ALL OF THIS. We've discussed it extensively.

I didn't create this thread to show that Bronze age Levanties, Arabs, East AFricans have Natufian-like ancestry. I created it to to show what I think is the most Natufian-like component and which I think can give us a general idea about the distribution of Natufian-like ancestry.

THAT WASN'T MY POINT. My point was that these modern "clusters" have a lot more than "Natufian" in them, and in addition they're highly drifted.

I'm fully aware of that. Nonetheless, like how North Europe components track Euro_HG ancestry, SouthWest Asia components can track Natufian-like ancestry.

That's why it's not a good idea to use modern clusters created by Admixture unless you're aware, and make your readers aware, of their limitations, and that the percentages should only be used to provide hints, not as some hard and fast "fact".

For goodness' sakes, do I really have to explain why it's better to use ancient genomes for something like this?

You don't have to because I never claimed it's better to use modern genomes. I don't think EUtest is the best way to track Natufian ancestry, I just think it is a way.

 

[Angela]

I repeat: we already know and have discussed what you seem to have mistakenly thought was some astonishing new insight, so I didn't see the point of your thread. There's nothing new there.

It's also not the best way to find "Natufian" like ancestry, which is what I was trying to point out, and as you yourself finally admit.

Please go back and read my posts, and especially number 12, and try to understand them.

I don't see anything else of substance to which to respond.

 

[MarkoZ]

Abstract population clusters based on modern genomes just look like an additonal layer of complexity without having any real explanatory power. I don't really get the purpose unless there's a complete lack of ancient samples.

 

[Fire Haired14]

I repeat: we already know and have discussed what you seem to have mistakenly thought was some astonishing new insight, so I didn't see the point of your thread. There's nothing new there.

As I said earlier I know many(not most) here understand Red Sea/SouthWest Asian components carry a lot of Natufian-like stuff. I posted this thread because I think EUtest's "MIDDLE_EASTERN" component is the most Natufian-like component and a good tracker of Natufian-like ancestry. So no I am not restating something which has already been discussed. I'm bringing forth new interesting information; the most Natufian-like component.

The above mentioned purpose of this thread is worth discussing for multiple reasons. Here are a few...

1: Some posters, including yourself, have claimed the SouthWest Asian scores in Southern Europe are from EEF not non-EEF Middle Eastern ancestry. Because Barcin_Neolithic scores high in SouthWest Asian components in other calculators that is valid. But this EUtest is maybe the only calculator in which a Natufian scores over 50% in a SouthWest Asian component which Barcin_Neolithic scores only 6% in. Therefore the 6-12% scores in modern SouthEast Europe can probably only be explained by some Levant Neolithic/Natufian-like ancestry.

2: Posters here have discussed post-Neolithic CHG influx into Southern Europe but few have discussed post Neolithic Natufian(ish) influx into Southern Europe. EUtest suggests the extra "CHG" came with Natufian(ish) stuff. Some posters have suggested BarcinChalcolithic, who had lots of CHG, is a good proxy for the people who brought extra CHG into Southern Europe. But BarcinChalcolithic scores 0% MIDDLE_EASTERN while Sicilians for example score 12%. EUtest indicates a more Levant/Arabian/Southwestern-like population penetrated Southern Europe after the Neolithic.

3: Levant_Neolithic has been suggested to be a good proxy for West Eurasian ancestry in Ethiopia. Ethiopians though score 0% in WEST_MED and EAST_MED while Levant_Neolithic scores a combined ~50%. 100% of Ethiopians West Eurasian score goes to MIDDLE_EASTERN. This indicates Ethiopians' West Eurasian ancestry is even more Natufian-like than Levant_Neolithic. But of course we can't be very confident because EUtest is based on modern populations.

4: Cypriot scores a healthy amount of MIDDLE_EASTERN. Turkey scores some as well. BarcinChalcolithic though scores 0%. This indicates people from the southeast migrated into Anatolia and Cypriot.

It's also not the best way to find "Natufian" like ancestry, which is what I was trying to point out, and as you yourself finally admit.

I never said it is the best way to find Natufian-like ancestry!!! Never! Those are words you put in my mouth. You just made up an argument I was never apart of then declared yourself the victor.

Look, I understand I didn't write a liability statement clarifying that I understand ADMIXTURE components are created from modern DNA, that MIDDLE_EASTERN isn't pure Natufian. If one doesn't already know I understand those things I can understand why they would get a little irritated by my thread. But now that you know I understand those things you have no reason to be upset.

 

[Fire Haired14]

Abstract population clusters based on modern genomes just look like an additonal layer of complexity without having any real explanatory power.

Even when ADMIXTURE is created with modern genomes it can detect ancient forms of ancestry.

The West Eurasian components in ADMIXTURE tests using modern genomes mostly derive from single Mesolithic/Neolithic West Eurasian populations.
Component, its main Mesolithic ancestor.
Caucasus; Mesolithic Caucasus.
Gedorsia; Neolithic Iran.
West Med; Neolithic Anatolia.
North Europe; Mesolithic Europe.
SouthWest Asia; Mesolithic SouthWest Asia.

I don't really get the purpose unless there's a complete lack of ancient samples.

It can detect relatively recent ancestor components. For example it can detect an Ashkenasi Jewish component. It can probably detect a Slavic component. It's definitly useful.