PDA

View Full Version : Questions about SSA/ANA admixture among north africans



Nassbean
08-25-2020, 03:06 PM
I saw some users stating that the ANA ancestry among north africans should be interpreted as fully "ssa" which implies that I'm around 13%-17% "SSA' depending on the model. Being a bit more northern shifted than the average moroccan that would imply that the average moroccan on g25 is around 17%-22% SSA.

So should ANA admixture be interpreted as fully "SSA" genetically ? If yes why doesn't it impact our phenotypes ? Compared to modern "quadroons", north africans look way more "caucasoid/eurasian" especially riffians and kabyles


here my results :

Target: Nassbean_scaled
Distance: 2.3930% / 0.02393013
34.6 Iberia_N
25.6 MAR_EN
18.6 Levant_PPNC
11.0 Yamnaya_RUS_Samara
5.4 IRN_Ganj_Dareh_N
4.8 SSA

Target: Nassbean_scaled
Distance: 2.9504% / 0.02950412
46.0 Anatolia_Barcin_N
27.4 MAR_Iberomaurusian
11.8 Yamnaya_RUS_Samara
6.6 Yoruba
4.8 Levant_Natufian
1.8 IRN_Ganj_Dareh_N
1.6 WHG

Atlas
08-25-2020, 03:51 PM
I'm not an expert by any means, but I think the reason some users might say that has to do with this study (see pg 14 for the graph with Taforalt):
https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/2020_Lipson_Nature_ShumLaka.pdf

The reasoning may follow along the lines of a post made by a user here at the top of the page from this link:
https://anthrogenica.com/showthread.php?20374-Ashkenazi-and-Sub-Saharan-ancestry-tracing-using-segment-matches-and-G25-modelling/page16

From what I've gathered from your post, you are wondering why Moroccans often don't look like people in the Americas who may be ~20% SSA? Well, ignoring the wide range of phenotypes that people with this ancestry have, I think this is probably due to a variety of reasons and one of them probably involves the term SSA. There are various phenotypes within this region and they haven't just all existed unchanged since the dawn of time, but have changed over time due to environmental adaptation and intermixing with other groups (whether other Africans or not). I'm unsure if we know what Taforalt people looked like to make conclusions that, being roughly a quarter SSA from Taforalt (if we were to call Taforalt such), the average Moroccans you're talking about should look like someone who also has around as much SSA ancestry from west central Africa as the hypothetical mixed person from the Americas that we're imagining. The point I'm making may be better illustrated by thinking about whether you expect someone who derives half their ancestry from the ǃKung and half from the Netherlands to look "the same" as someone half Dinka and half Dutch?
However, if we are to work on the assumption that people from Taforalt looked "stereotypically SSA", then there may also be a case for selection having occurred amongst your north African ancestors wherein features dissimilar to the "stereotypical SSA" model were instead favoured.

Personally, I wouldn't worry about attaching that part of your ancestry to SSA or not. It's indigenous to Africa and that would be good enough for me :).

Hawk
08-25-2020, 04:06 PM
Because it's not. ANA is deeper than Basal Eurasian but yet on the same tree root as other Eurasian components.

It's native to North Africa, pre OOA event, yet closer to Eurasian than SSA. Add it, Iberomaurusian/Taforalt skulls by anthropologists were classified as Mechtoid, they could easily distinguish between the incoming Capsians and SSA.

sami15
08-25-2020, 04:14 PM
I'm North Moroccan (Riffian) and I have only 3% of Basal west African SSA just the same as ancient guanches (mainly Guanche11)

39161
https://anthrogenica.com/attachment.php?attachmentid=39161&d=1598372048

Nassbean
08-25-2020, 04:21 PM
I'm not an expert by any means, but I think the reason some users might say that has to do with this study (see pg 14 for the graph with Taforalt):
https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/2020_Lipson_Nature_ShumLaka.pdf

The reasoning may follow along the lines of a post made by a user here at the top of the page from this link:
https://anthrogenica.com/showthread.php?20374-Ashkenazi-and-Sub-Saharan-ancestry-tracing-using-segment-matches-and-G25-modelling/page16

From what I've gathered from your post, you are wondering why Moroccans often don't look like people in the Americas who may be ~20% SSA? Well, ignoring the wide range of phenotypes that people with this ancestry have, I think this is probably due to a variety of reasons and one of them probably involves the term SSA. There are various phenotypes within this region and they haven't just all existed unchanged since the dawn of time, but have changed over time due to environmental adaptation and intermixing with other groups (whether other Africans or not). I'm unsure if we know what Taforalt people looked like to make conclusions that, being roughly a quarter SSA from Taforalt (if we were to call Taforalt such), the average Moroccans you're talking about should look like someone who also has around as much SSA ancestry from west central Africa as the hypothetical mixed person from the Americas that we're imagining. The point I'm making may be better illustrated by thinking about whether you expect someone who derives half their ancestry from the ǃKung and half from the Netherlands to look "the same" as someone half Dinka and half Dutch?
However, if we are to work on the assumption that people from Taforalt looked "stereotypically SSA", then there may also be a case for selection having occurred amongst your north African ancestors wherein features dissimilar to the "stereotypical SSA" model were instead favoured.

Personally, I wouldn't worry about attaching that part of your ancestry to SSA or not. It's indigenous to Africa and that would be good enough for me :).


Thank you but based on what you posted it seems north africans should have more ssa (if taforalt is being modelled as 45% ANA) which seems unlikely and we actually know that iberomaurusians were mostly west eurasian and were "caucasoid" (craniometrically speaking). They didn't have the SNPs for light skin but they had the ones for straight hair (overall they were of the cro-magnoid type similar to what we find in Europe during the Paleolithic).

In terms of reconstruction this is what we have (just the skin color is inaccurate) :

https://i.imgur.com/Eub7RBU.jpg


these people had around 45% or 30% (?) of their dna being ANA

And I understand your argument but it doesn't change the fact that negroid features and dark skin alleles should impact our phenotypes no matter the existing diversity and yes I have no problem with it being african but I just want to know exactly my "SSA" admixture because between 4-6% and 13-17% there is a large gap. Anyway thank you for you intervention !

ThaYamamoto
08-25-2020, 04:25 PM
From the Shum Laka paper, if 47% of Taforalt is Ancient Ancestral North African, and if this component is SSA (not saying it is, I don't know) that's almost half of whatever Iberomaurisian you score + Yoruba/Dinka. This seems to be the case on current intra-African PCAs : 5% Yoruba/Dinka Egyptian Copt or Iraqi with 15% Yoruba/Dinka [there is one on the G25] will plot much further from the African cline then say, a North Moroccan with 5% Yoruba/Dinka but 30% Iberomaurusian. Whether or not this pull just reflects the AANA ancestry in West-Africans etc we don't know yet, but the PCA makes it clear that currently, it acts as extra SSA affinity. Phenotype discussion is a pointless endeavour.

Nassbean
08-25-2020, 04:29 PM
Because it's not. ANA is deeper than Basal Eurasian but yet on the same tree root as other Eurasian components.

It's native to North Africa, pre OOA event, yet closer to Eurasian than SSA. Add it, Iberomaurusian/Taforalt skulls by anthropologists were classified as Mechtoid, they could easily distinguish between the incoming Capsians and SSA.

Interesting but some members defended the idea that even this basal eurasian should be seen as "black" because closer to SSA pops when it comes to their phenotypes and taforalt being distinct craniometrically is maybe due to the fact that they were mostly west eurasian (dzudzuana-like)

What do you think ?

Hawk
08-25-2020, 04:34 PM
Interesting but some members defended the idea that even this basal eurasian should be seen as "black" because closer to SSA pops when it comes to their phenotypes and taforalt being distinct craniometrically is maybe due to the fact that they were mostly west eurasian (dzudzuana-like)

What do you think ?

Nah, the gracilization happened due to Dzudzuana-like. The craniometrically Mechtoids were quite disntinctive from SSA-heavy people and Dzudzuana-like people.

Because it's a deep ancestry, they just imply/assume wrongly it's SSA. I am not saying it's Eurasian, on contrary. But it's very different from SSA aswell who have various layers of different admixtures. The Shum Laka samples have some kind of archaic Y-DNA (A00 something).

Nassbean
08-25-2020, 04:37 PM
From the Shum Laka paper, if 47% of Taforalt is Ancient Ancestral North African, and if this component is SSA (not saying it is, I don't know) that's almost half of whatever Iberomaurisian you score + Yoruba/Dinka. This seems to be the case on current intra-African PCAs : 5% Yoruba/Dinka Egyptian Copt or Iraqi with 15% Yoruba/Dinka [there is one on the G25] will plot much further from the African cline then say, a North Moroccan with 5% Yoruba/Dinka but 30% Iberomaurusian. Whether or not this pull just reflects the AANA ancestry in West-Africans etc we don't know yet, but the PCA makes it clear that currently, it acts as extra SSA affinity. Phenotype discussion is a pointless endeavour.

thank you and can you post an example for this PCA because in my case on all PCAs I'm more northern shifted than egyptians like many other north moroccans but yes I agree it would clearly imply SSA affinity for this component.

And why do you think it is a pointless endeavour when it comes to phenotype ? Because it's too old to impact it ? I mean all populations with clear SSA admixture show signs of it on their phenotype

Atlas
08-25-2020, 04:41 PM
Thank you but based on what you posted it seems north africans should have more ssa (if taforalt is being modelled as 45% ANA) which seems unlikely and we actually know that iberomaurusians were mostly west eurasian and were "caucasoid" (craniometrically speaking). They didn't have the SNPs for light skin but they had the ones for straight hair (overall they were of the cro-magnoid type similar to what we find in Europe during the Paleolithic).

In terms of reconstruction this is what we have (just the skin color is inaccurate) :

https://i.imgur.com/Eub7RBU.jpg


these people had around 45% or 30% (?) of their dna being ANA

And I understand your argument but it doesn't change the fact that negroid features and dark skin alleles should impact our phenotypes no matter the existing diversity and yes I have no problem with it being african but I just want to know exactly my "SSA" admixture because between 4-6% and 13-17% there is a large gap. Anyway thank you for you intervention !

I see what you mean, and as I'm out of my depth here I think I'll sit back and read the others' posts. May I ask what you are defining as SSA? Is it all genetic components located south of the Sahara today? Is it only people with "negroid features" as you said? I think you may be overestimating the effect of maybe ~25% of an ancestral component at most in many people on phenotype regardless.

Hawk
08-25-2020, 04:53 PM
From the Dzudzuana paper.

https://www.biorxiv.org/content/biorxiv/early/2018/09/20/423079/F2.large.jpg

gihanga.rwanda
08-25-2020, 04:53 PM
To add to what Atlas was saying, I am betting that the Reich lab is right about ANA forming a clade with “East African Afro-pastoralists”, as well as Ancient West Africans and Mota (minus ghost modern), to the exclusion of Eurasians and more basal populations in Sub-Saharan Africa.

Taforalt can only be modeled as 50% ANA with the remainder of its ancestry being modeled as standard Western Eurasian, which would partly explain why Taforalt is easily distinguished from modern SSAs based on their crania or genetically.

Most if not all modern SSAs also have varying degrees of basal African, which further distinguishes them from ANA; Mota can be modeled as 29% ghost modern and the same for Yoruba and Mende at a rate of 30-35%. My prediction is that Dinka and related Nilotic groups have similar levels of ghost modern ancestry; likewise, I think it’s likely that the African ancestors of Somalis and related Cushitic groups had less ghost modern ancestry than the Dinka.

Last point, quadroons almost always look more Eurasian than African; genetically quadroon Latinos (e.g. Puerto Ricans, Dominicans) with mixed West African and South European ancestry can easily pass in the Maghreb and Egypt.

gihanga.rwanda
08-25-2020, 05:02 PM
From the Dzudzuana paper.

https://www.biorxiv.org/content/biorxiv/early/2018/09/20/423079/F2.large.jpg

The Shum Laka paper was also published by the Reich lab and is more relevant to African genetics. The Dzudzuana and Shum Laka papers don’t even contradict each other; the Dzudzuana paper proposes that Basal Eurasian experienced most of the genetic drift characteristic of non-Africans, unlike ANA. Read the supplementary documents.

Nassbean
08-25-2020, 05:05 PM
I see what you mean, and as I'm out of my depth here I think I'll sit back and read the others' posts. May I ask what you are defining as SSA? Is it all genetic components located south of the Sahara today? Is it only people with "negroid features" as you said? I think you may be overestimating the effect of maybe ~25% of an ancestral component at most in many people on phenotype regardless.

By SSA I mean every component that is not eurasian (very divergent) and indigenous to sub-saharan africa so for example I do not consider somalis or malagasy people as fully "SSA" despite both of them being in sub-saharan africa. And as far as I know people who are fully "SSA" or close to it clearly show different craniometric features (like pygmies, dinkas or yorubas) and they tend to be on the darkest spectrum when it comes to skin pigmentation.


And yes I'm maybe overestimating the effect of 20-25% SSA but people with such amount of ssa (and the rest west eurasian) clearly show negroid traits.

Hawk
08-25-2020, 05:05 PM
The Shum Laka paper was also published by the Reich lab and is more relevant to African genetics. The Dzudzuana and Shum Laka papers don’t even contradict each other; the Dzudzuana paper proposes that Basal Eurasian shares most of the genetic drift characteristic of non-Africans, unlike ANA.

The Shum Laka paper clearly states this:


We can also obtain a good fit for the Shum Laka individuals in a less parsimonious alternative model using three components, replacing the
basal West African source with a combination of ancestry from inside the
clade defined by the other West African populations and from a source
entirely outside the West African clade (near one lineage that contributes to the Taforalt individuals) (Extended Data Fig. 5, Supplementary
Information section 3). However, two-component models for the Shum
Laka individuals that have the majority source splitting closer to other
West or East Africans are rejected (Z = 7.1 and Z = 3.7, respectively).

https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/2020_Lipson_Nature_ShumLaka.pdf

Nassbean
08-25-2020, 05:14 PM
To add to what Atlas was saying, I am betting that the Reich lab is right about ANA forming a clade with “East African Afro-pastoralists”, as well as Ancient West Africans and Mota (minus ghost modern), to the exclusion of Eurasians and more basal populations in Sub-Saharan Africa.

Taforalt can only be modeled as 50% ANA with the remainder of its ancestry being modeled as standard Western Eurasian, which would partly explain why Taforalt is easily distinguished from modern SSAs based on their crania or genetically.

Most if not all modern SSAs also have varying degrees of basal African, which further distinguishes them from ANA; Mota can be modeled as 29% ghost modern and the same for Yoruba and Mende at a rate of 30-35%. My prediction is that Dinka and related Nilotic groups have similar levels of ghost modern ancestry; likewise, I think it’s likely that the African ancestors of Somalis and related Cushitic groups had less ghost modern ancestry than the Dinka.

Last point, quadroons almost always look more Eurasian than African; genetically quadroon Latinos (e.g. Puerto Ricans, Dominicans) with mixed West African and South European ancestry can easily pass in the Maghreb and Egypt.

So when it comes to calculating my amount of SSA, should I count it as SSA or only partially SSA ?

And I disagree when it comes to quadroons : they clearly show SSA features



*Image removed

3.12 Discussions pertaining to physical appearance and/or phenotype that are quantitative and accompanied with structured analysis are generally welcome in Anthrogenica. However, subjective and arbitrary discussions relating to either subject that are agreed by our administration to breach section 3.11 above are generally prohibited. What constitutes as the aforementioned, and whether further action is required outside of content modification, will be carried out at the administration's discretion.

gihanga.rwanda
08-25-2020, 05:17 PM
The Shum Laka paper clearly states this:

That quote is describing this model, which clearly aligns with my earlier statement. The Shum Laka samples can be modeled as 54% Central African hunter-gatherer related (forms a clade with South African hunter-gatherer and ghost modern, collectively basal African), 40% basal West African (a 60:40 mix between ghost North African or ANA and ghost modern), and 16% additional ANA.

At a macro-level, we’re looking at 3 populations clusters: basal Africans, less basal Africans, and non-Africans.

https://i.imgur.com/xE6FrTx.jpg

gihanga.rwanda
08-25-2020, 05:27 PM
So when it comes to calculating my amount of SSA, should I count it as SSA or only partially SSA ?

And I disagree when it comes to quadroons : they clearly show SSA features

I’d classify your Yoruba/Dinka admixture as modern SSA and your ANA as ancient African.

I don’t think it makes sense to compare first generation quadroons with North Africans who descend from ancient admixture events, but just look at most of Yannick Noah’s children (25% Cameroonian) or Ryan Giggs (25% Sierra Leonean).


*Images removed

3.12 Discussions pertaining to physical appearance and/or phenotype that are quantitative and accompanied with structured analysis are generally welcome in Anthrogenica. However, subjective and arbitrary discussions relating to either subject that are agreed by our administration to breach section 3.11 above are generally prohibited. What constitutes as the aforementioned, and whether further action is required outside of content modification, will be carried out at the administration's discretion.

Hawk
08-25-2020, 05:28 PM
That quote is describing this model, which clearly aligns with my earlier statement. The Shum Laka samples can be modeled as 54% Central African hunter-gatherer related (forms a clade with South African hunter-gatherer and ghost modern, collectively basal African), 40% basal West African (a 60:40 mix between ghost North African or ANA and ghost modern), and 16% additional ANA.

At a macro-level, we’re looking at 3 populations clusters: basal Africans, less basal Africans, and non-Africans.

https://i.imgur.com/xE6FrTx.jpg

I will not argue with you. It's written in a plain language and in line with Dzudzuana paper. It states from a source entirely outside the West African clade.

I do agree ANA is African, but they clearly had a separate line, add it that down the line they mixed with Dzudzuana-like people.

ThaYamamoto
08-25-2020, 05:33 PM
thank you and can you post an example for this PCA because in my case on all PCAs I'm more northern shifted than egyptians like many other north moroccans but yes I agree it would clearly imply SSA affinity for this component.

And why do you think it is a pointless endeavour when it comes to phenotype ? Because it's too old to impact it ? I mean all populations with clear SSA admixture show signs of it on their phenotype

Sure, I can't get my custom PCA to look like Angoliga's/Michalis Moriopoulos's so the Vahaduo set-PCA works best for me in replicating quantifying African ancestry only against non-African. Here goes:



The Morocco North average is ~7%; Egyptian average ~10-11%. Yet the Moroccan average plots closer to Africans than the Egyptian - 30%+ Iberomaurusian that the Egyptian average lacks. Look at North Moroccan user Amhas..10% SSA but 30% Iberomaurusian= ~25% pull. Again I don't know to what extent ANA is 'SSA' or all that, but it definitely effects the pull toward SSA populations in a manner which fits the 'African' component ratio. Otherwise the Egyptian average would plot ahead of the Moroccan one.

I find phenotype discussion particularly when it comes to Africans creepy - but to each his own lol, not bashing anyone it's just not for me.

edit: attached the wrong PCA

gihanga.rwanda
08-25-2020, 05:33 PM
I will not argue with you. It's written in a plain language and in line with Dzudzuana paper. It states from a source entirely outside the West African clade.

I do agree ANA is African, but they clearly had a separate line, add it that down the line they mixed with Dzudzuana-like people.

That is completely within your right.

ANA obviously falls outside the “basal West African” clade since the latter is ~30% ghost modern; this doesn’t change the point that the basal West African clade draws ~70% of its ancestry from a population closely related to ANA.

Nassbean
08-25-2020, 05:37 PM
I’d classify your Yoruba/Dinka admixture as modern SSA and your ANA as ancient African.

I think it makes sense to compare first generation quadroons with North Africans who descend from ancient admixture events, but just look at most of Yannick Noah’s children (25% Cameroonian) or Ryan Giggs (25% Sierra Leonean).

I won’t post any more pics since this is all subjective.

so according to you I'm indeed 13-17% SSA ?

and your examples are bad because these people are mostly of north-west european stock which can make them lighter compared to a middle eastern/south euro mix with 25% ssa ...just look at some quadroon dominicans or cubans they don't look as eurasian as us

Nassbean
08-25-2020, 05:42 PM
Sure, I can't get my custom PCA to look like Angoliga's/Michalis Moriopoulos's so the Vahaduo set-PCA works best for me in replicating quantifying African ancestry only against non-African. Here goes:

39165

The Morocco North average is ~7%; Egyptian average ~10-11%. Yet the Moroccan average plots closer to Africans than the Egyptian - 30%+ Iberomaurusian that the Egyptian average lacks. Look at North Moroccan user Amhas..10% SSA but 30% Iberomaurusian= ~25% pull. Again I don't know to what extent ANA is 'SSA' or all that, but it definitely effects the pull toward SSA populations in a manner which fits the 'African' component ratio. Otherwise the Egyptian average would plot ahead of the Moroccan one.

I find phenotype discussion particularly when it comes to Africans creepy - but to each his own lol, not bashing anyone it's just not for me.

Nice ! Based on this I'd say I'm more northern shifted than egyptians probably because of additionnal european ancestry (on my model interpreted as "yamnaya/steppe" ancestry) and imo if ANA was fully SSA the pull towards SSA would have been greater than this but I may be wrong on this

ThaYamamoto
08-25-2020, 05:45 PM
Nice ! Based on this I'd say I'm more northern shifted than egyptians probably because of additionnal european ancestry (on my model interpreted as "yamnaya/steppe" ancestry) and imo if ANA was fully SSA the pull towards SSA would have been greater than this but I may be wrong on this

Hey I edited the original post cuz I'd attached the wrong PCA, the new PCA still makes my point though. Yeah there's additional Yamnaya in Maghrebis etc, this PCA though, unlike the custom made PCAs on Vahaduo, is quantifying SSA ancestry against non-SSA ancestry only - or at least as best as it can within its constraints. Otherwise I'd be shooting off of it with my steppe/aboriginal indian etc ancestry lol. That's the nice thing about a well developed PCA...the cline stays nice and consistent for you to measure upon. So you are prolly northern shifted i.e. Euro shifted if that is your primary concern.

gihanga.rwanda
08-25-2020, 05:47 PM
so according to you I'm indeed 13-17% SSA ?

and your examples are bad because these people are mostly of north-west european stock which can make them lighter compared to a middle eastern/south euro mix with 25% ssa ...just look at some quadroon dominicans or cubans they don't look as eurasian as us

Based on your results from your original post, you’re ~20% indigenous African (13.7% ANA + 6.6 West African)... just leave it at that and call it a day.

You can definitely find Cubans and Puerto Ricans who are ~20% West African and look similar to coastal Maghrebis, but like I said, you’re comparing apples and oranges right now; for example, those Latinos also have notable Amerindian ancestry, which is completely absent in Africa.

Adamm
08-25-2020, 05:59 PM
Here I managed to collect 42 North African G25 samples from different tribes and areas in (mostly) Morocco and Algeria for what it's worth:

https://i.imgur.com/gkFWVjf.png

Nassbean
08-25-2020, 06:00 PM
Based on your results from your original post, you’re ~20% indigenous African (13.7% ANA + 6.6 West African)... just leave it at that and call it a day.

You can definitely find Cubans and Puerto Ricans who are ~20% West African and look similar to coastal Maghrebis, but like I said, you’re comparing apples and oranges right now; for example, those Latinos also have notable Amerindian ancestry, which is completely absent in Africa.

the first model I made is a bit more accurate so more like 15-20% "indigenous" african but we shouldn't draw such conclusions too quickly we still need consistent genetic data about them and what's your opinion about aterians ? Can they be this famous ghost population ?


And when I said "quadroons" I was specifically talking about real quadroons with 20-25% ssa and the rest south european (also I don't think 1-3% of amerindian ancestry will impact your phenotype that much).

gihanga.rwanda
08-25-2020, 06:11 PM
the first model I made is a bit more accurate so more like 15-20% "indigenous" african but we shouldn't draw such conclusions too quickly we still need consistent genetic data about them and what's your opinion about aterians ? Can they be this famous ghost population ?


And when I said "quadroons" I was specifically talking about real quadroons with 20-25% ssa and the rest south european (also I don't think 1-3% of amerindian ancestry will impact your phenotype that much).

I am not sure about the Aterians but I agree with you here, we need more Paleolithic aDNA from the northern half of Africa to draw solid conclusions. Hopefully we find new discoveries in the next 1-2 years... I think we can expect some interesting papers from the Reich lab.

ThaYamamoto
08-25-2020, 06:21 PM
Here I managed to collect 42 North African G25 samples from different tribes and areas in (mostly) Morocco and Algeria for what it's worth:

https://i.imgur.com/gkFWVjf.png

Do you have any results of anyone local from Dar El Beďda/Ain Taya? My bad if they're already on there, I wouldn't know.

Adamm
08-25-2020, 06:49 PM
Do you have any results of anyone local from Dar El Beďda/Ain Taya? My bad if they're already on there, I wouldn't know.

You mean Casablanca? I have samples from people living there but not from someone originally from it, its because majority of people living in Casablanca are outsiders and the 'original' people would be mixed into oblivion as its one big melting pot there.

ThaYamamoto
08-25-2020, 07:06 PM
You mean Casablanca? I have samples from people living there but not from someone originally from it, its because majority of people living in Casablanca are outsiders and the 'original' people would be mixed into oblivion as its one big melting pot there.

Nah nah I mean Algeria, I think near Algiers. I knew a buncha people from a place they called 'Ain Taya' they were hella proud of their district lol. Would be interesting to see their results they were pretty diverse.

Imesmouden
08-25-2020, 07:11 PM
I'm North Moroccan (Riffian) and I have only 3% of Basal west African SSA just the same as ancient guanches (mainly Guanche11)

39161
https://anthrogenica.com/attachment.php?attachmentid=39161&d=1598372048

Can you pm me the model please

Ignis90
08-25-2020, 09:05 PM
It's probably best to avoid discussing phenotypes in length.

The mix between ANA and the Eurasian ancestors of Taforalt happened at least 20000 years ago if we go by the archeological age of the Iberomaurusian culture: so no surprise they seem to have looked very different from any modern population.


We also cannot compare modern mixed populations and individuals (mostly from the "New World" ) to Old World populations. Berbers seem to have formed in prehistory (Bronze Age) or at the beginning of antiquity/early Iron Age (comparable to Europeans) with all its modern components, we have enough aDNA to already see that (including European steppes and West-Central African, although continuous geneflow cannot be totally ruled out). Furthermore, it's likely Berbers underwent major "recent" bottlenecks (as evidenced by the y-dna profile ) and drift over millenia. So I don't see why there should be any commonalities with modern populations which are recent and do not share the same ancestry at all (for exemple, West Indians have not only Native American dna, but also their African and West Eurasian ancestry is mostly unlike that of Maghrebis). It's like comparing an American with a Japanese great-great grandparent with Finns: nothing in common.

Anyway, ANA is clearly African/didn't go through the same main Out-of-Africa bottleneck ("SSA" doesn't mean much when the Sahara is wet or before the Holocene/Neolithic Revolution).
However, we still don't know how basal/non-basal ANA is in the African context. So it's hard to gauge Eurasian/OoA vs African ancestry in Maghrebis. We need more aDNA from Africa, including Aterians even though I don't think there is a strong case for Aterian continuity with later cultures of North Africa.
That being said, it won't change much from current estimates IMHO, we're talking single digits here. This is supported for instance by some Soussis, which can sometimes reach close to 50% Iberomaurusian and less than 5 % West-Central African : they don't seem to be less African-shifted in PCAs than virtually all other Maghrebis which have not only less Iberomaurusian ancestry than Soussis but also inversely more West-Central African ancestry. This tells me ANA behaves very much like modern SSA components in terms of basalness.

Nassbean
08-25-2020, 09:20 PM
It's probably best to avoid discussing phenotypes in length.

The mix between ANA and the Eurasian ancestors of Taforalt happened at least 20000 years ago if we go by the archeological age of the Iberomaurusian culture: so no surprise they seem to have looked very different from any modern population.


We also cannot compare modern mixed populations and individuals (mostly from the "New World" ) to Old World populations. Berbers seem to have formed in prehistory (Bronze Age) or at the beginning of antiquity/early Iron Age (comparable to Europeans) with all its modern components, we have enough aDNA to already see that (including European steppes and West-Central African, although continuous geneflow cannot be totally ruled out). Furthermore, it's likely Berbers underwent major "recent" bottlenecks (as evidenced by the y-dna profile ) and drift over millenia. So I don't see why there should be any commonalities with modern populations which are recent and do not share the same ancestry at all (for exemple, West Indians have not only Native American dna, but also their African and West Eurasian ancestry is mostly unlike that of Maghrebis). It's like comparing an American with a Japanese great-great grandparent with Finns: nothing in common.

Anyway, ANA is clearly African/didn't go through the same main Out-of-Africa bottleneck ("SSA" doesn't mean much when the Sahara is wet or before the Holocene/Neolithic Revolution).
However, we still don't know how basal/non-basal ANA is in the African context. So it's hard to gauge Eurasian/OoA vs African ancestry in Maghrebis. We need more aDNA from Africa, including Aterians even though I don't think there is a strong case for Aterian continuity with later cultures of North Africa.
That being said, it won't change much from current estimates IMHO, we're talking single digits here. This is supported for instance by some Soussis, which can sometimes reach close to 50% Iberomaurusian and less than 5 % West-Central African : they don't seem to be less African-shifted in PCAs than virtually all other Maghrebis which have not only less Iberomaurusian ancestry than Soussis but also inversely more West-Central African ancestry. This tells me ANA behaves very much like modern SSA components in terms of basalness.

I see but then I suppose you would have no problem with someone saying north africans are around 20% ssa whether it's recent or prehistoric ?

Ignis90
08-25-2020, 10:54 PM
I WOULD have a problem with the term "SSA" since we're talking about divergent streams of ancestries which are often more divergent than between Eurasian streams. And ANA being indigenous to North Africa in the past, it is does not make sense to call it SSA.

"African/non-OoA" is more accurate IMO. We can probably sum up Berbers/Maghrebis generally as a "periphery West Eurasian population with strong local African paleolithic continuity and Holocene African contributions".

"SSA" might be more useful when talking about Haratins for instance in a modern Maghrebi context.

sami15
08-26-2020, 12:30 AM
...........

sami15
08-26-2020, 12:31 AM
Can you pm me the model please

what model you want?

sami15
08-26-2020, 12:31 AM
Can you pm me the model please

my pm does not work

sami15
08-26-2020, 12:32 AM
Can you pm me the model please

for what reason you want that model you asking for?

Michalis Moriopoulos
08-26-2020, 04:38 AM
So should ANA admixture be interpreted as fully "SSA" genetically ?

We don't really know what ANA is yet, so the prudent thing to do for now is to withhold judgment about how to characterize it. All lineages deeper than Basal Eurasian are described by Harvard as "Deep [African]" lineages. That seems like a fair descriptor for now. For a long time SSA served as a wastebasket taxon for all lineages deeper than the Eurasian bottleneck because modern sub-Saharan African populations were the only representatives of pre-Eurasian bottleneck populations that we had around. But once ancient DNA became viable, it was just going to be a matter of time before we discovered lineages that were intermediate between modern SSAs and Eurasians.

We also don't know how or when sub-Saharan African structure took on its current form. For all we know SSAs are the result of various very deep African lineages (what one might call Paleo-African) coalescing together before mixing with the "shallower lineages" (we might call them Neo-African) which presumably gave rise to ANA and Basal Eurasian. There are many models now and nobody knows for sure what's right. So watch this space.

Michalis Moriopoulos
08-26-2020, 04:47 AM
You mean Casablanca? I have samples from people living there but not from someone originally from it, its because majority of people living in Casablanca are outsiders and the 'original' people would be mixed into oblivion as its one big melting pot there.

Incidentally, I've learned recently that the Moroccan samples labelled "MCA" in the G25 were collected in Casablanca.

Thanks to Amhas' guidance, the other abbreviations for Moroccan_North and Moroccan_South make sense, too. Most of the Moroccan South samples are clearly from the Southeast (Errachidia and Ouarzazate), which explains why there are so many haratin-type people in the set.

Michalis Moriopoulos
08-26-2020, 05:01 AM
they don't seem to be less African-shifted in PCAs than virtually all other Maghrebis which have not only less Iberomaurusian ancestry than Soussis but also inversely more West-Central African ancestry. This tells me ANA behaves very much like modern SSA components in terms of basalness.

Yes. One look at where Iberomaurusians cluster in Global PCAs can tell us that a pure ANA component would cluster with SSAs. I suppose once Dzudzuana is released, someone can make "ghost" ANA coordinates and we can see for ourselves.

Then again just because the deep African lineages all cluster together on the typical Global PCA isn't all that informative because Eurasian drift absolutely dominates the variation on global PCAs, obscuring the huge differences that exist between San, pygmies, and Tropical Africans. They might all cluster in the same corner when put in a PCA with thousands of Eurasians but they are definitely not the same.

Tangentially, I've wondered where a Basal Eurasian would cluster on a Global PCA. That's a real brainteaser considering where the Basal-rich populations plot in relation to Villabruna and company.

Mansamusa
08-26-2020, 06:43 AM
We don't really know what ANA is yet, so the prudent thing to do for now is to withhold judgment about how to characterize it. All lineages deeper than Basal Eurasian are described by Harvard as "Deep [African]" lineages. That seems like a fair descriptor for now. For a long time SSA served as a wastebasket taxon for all lineages deeper than the Eurasian bottleneck because modern sub-Saharan African populations were the only representatives of pre-Eurasian bottleneck populations that we had around. But once ancient DNA became viable, it was just going to be a matter of time before we discovered lineages that were intermediate between modern SSAs and Eurasians.

We also don't know how or when sub-Saharan African structure took on its current form. For all we know SSAs are the result of various very deep African lineages (what one might call Paleo-African) coalescing together before mixing with the "shallower lineages" (we might call them Neo-African) which presumably gave rise to ANA and Basal Eurasian. There are many models now and nobody knows for sure what's right. So watch this space. I don't remember them ever using the word African to describe this ancestry or even acknowledging that it could possibly be African. Am I mistaken?

Imesmouden
08-26-2020, 11:12 AM
for what reason you want that model you asking for?

i want to see my results with this model

you can post it just here if your pm doesn't works

sami15
08-26-2020, 01:30 PM
Incidentally, I've learned recently that the Moroccan samples labelled "MCA" in the G25 were collected in Casablanca.

Thanks to Amhas' guidance, the other abbreviations for Moroccan_North and Moroccan_South make sense, too. Most of the Moroccan South samples are clearly from the Southeast (Errachidia and Ouarzazate), which explains why there are so many haratin-type people in the set.

But haratins are berber speaking people by the way and they speak southern berber language AKA tachelhit
check your inbox...I sent to you my G25 coordinates
sorry for the delay its because my PM does not work before

sami15
08-26-2020, 01:30 PM
i want to see my results with this model

you can post it just here if your pm doesn't works

ok check your inbox

Michalis Moriopoulos
08-27-2020, 04:01 AM
I don't remember them ever using the word African to describe this ancestry or even acknowledging that it could possibly be African. Am I mistaken?

The authors don't emphasize it but these passages make it pretty clear Lazaridis and company believe ANA is a lineage indigenous to Africa:

"Europeans are differentiated by an excess of up to ~20% Villabruna-related ancestry relative to non-European populations and also by a relative lack of extra ‘Deep’ ancestry compared to the Near East and North Africa, a type of ancestry that may only partially be explained by the Basal Eurasian ancestry of ancient West Eurasian populations and must also trace to Africa (Extended Data Fig. 7)." - Dzudzuana Preprint

"The other “Deep” lineage found in Taforalt (Fig. 2) experienced only 0.008 units of genetic drift with non-Africans (Fig. 2) and could be plausibly interpreted as having deep presence in (North) Africa. Note that Taforalt and the Neolithic of the Maghreb are well below the regression line (Extended Data Fig. 8) and thus lack more genetic drift with Ust’Ishim than is predicted by their level of archaic ancestry; this is expected if they trace their ancestry from a lineage that is even more deeply diverged than the Basal Eurasians." - Dzudzuana Preprint Supplement


check your inbox...I sent to you my G25 coordinates
sorry for the delay its because my PM does not work before

No problem, man, thanks for your contribution. The more Berbers the merrier!


But haratins are berber speaking people by the way and they speak southern berber language AKA tachelhit

I'm sure most do, but what about in places like Errachidia? Three of the Errachidia samples in the G25 score ~35-50% SSA so I can only presume these are haratin. Judging by this map (https://upload.wikimedia.org/wikipedia/commons/7/74/Tachelhit.png) people in that area speak Central Atlas Tamazight, Zenata dialects, and/or Arabic.

In any case, it's cool to see G25 support a strong haratin presence in the Draa-Tafilalet area given how elevated SSA ancestry is in the Ouarzazate samples (half of the series shows strong SSA admixture). This is what you'd expect from that region given what I've read and been told about its history. A couple of the Tiznit samples are also haratin-like.

Hawk
08-27-2020, 06:16 AM
based on the morphological analyses of the Ibero-Maurusians, ANA was morphologically distinct from sub-saharan africans (Cranio-Dentally) because if it was so it would have been obvious in the Iberomaurusians but it isn't , (Ibero-Maurusians resemble late Pleistocene Europeans)
Its clear that this ANA had its own phenotype, and genetically is not as deep as sub-saharan africans its intermediate..
ANA can be summed in this " a population pre-OOA ,indigenous to North africa".
There's no need to force the term "SSA" on it , since it was physically and genetically different..while understanding that it wasn't an Eurasian ancestry.

Taforalt is 45% ANA and 55% Dzudzuana
But Taforalt in PCA don't act the same as a usual mixed 45% SSA and 55% Eurasian populations , its more Eurasian shifted.

Ddudzuana itself has Basal Eurasian in it probably. I highly doubt Taforalt has so much Dzdudzuana, it's probably the shared Basal Eurasian which masks in. But yes, i agree with you. Watching the skeleton of the Paleolithic/Mesolithic hunter gatherer from Cairo museum, the guy most likely was ~90% ANA, he looks distinct in morphology. Extremelly robust, especially the mandible.

sami15
08-27-2020, 02:34 PM
I'm sure most do, but what about in places like Errachidia? Three of the Errachidia samples in the G25 score ~35-50% SSA so I can only presume these are haratin. Judging by this map (https://upload.wikimedia.org/wikipedia/commons/7/74/Tachelhit.png) people in that area speak Central Atlas Tamazight, Zenata dialects, and/or Arabic.

In any case, it's cool to see G25 support a strong haratin presence in the Draa-Tafilalet area given how elevated SSA ancestry is in the Ouarzazate samples (half of the series shows strong SSA admixture). This is what you'd expect from that region given what I've read and been told about its history. A couple of the Tiznit samples are also haratin-like.

yeah, around Errachidia they speak not only Central Atlas Tamazight, Zenata dialects, and/or Arabic but also Tachelhit, it depends where samples are taking from, those near warzazate City are Tachelhit speaking people as for example, but this area in roman era was known by the presence of westerner aethiopian peoples called Daratiae Aethiopian according to Pliny The Elder,

we don't know yet, if these dark-berber/arabic people are a mixture of SSA from islamic era or from roman age, or both
There is also another historical explanation The word haratin is derived from the word Ahardane in berber language, the singular of it is aharid or aharar, and its origin is a Latin Roman word ardere, which means burned ... its equivalent to the word, Aethiopian, which is a Greek word meaning burnt face ... We find the word ardere/Darae also in the ancient Roman maps of the Draa Valley in southern Morocco (where a lot of SSA admixed poplation are found), and it was also used in Same context and connotation.

Culturally, Linguistically and ethnically, those SSA admixed population in Morocco in Our modern-days, don't consider themselves different from other berbers or arabic-speaking people, its because they claim themselves as natives, they act as they're indigenous, hence the word haratin is not really used as a term for any distinct community in Morocco.
it's pretty difficult to find what a unadmixed Haratin was like in Genetics or in modern life society.

gihanga.rwanda
08-27-2020, 04:14 PM
actually "qpAdm can also successfully
model Taforalt as a mixture of 73% Dzudzuana and 27% Mbuti"
which is actually closer to admixture
Which means if anything ANA is the one eating there extra Eurasian ancestry , this explains why in PCA taforalt is more Eurasian than an 45% african admixed populations.

Taforalt clusters just north of the Afar/Agaw, who can be modeled as approximately 50:50 SSA/West Eurasian, in every global PCA I’ve seen since the original paper; this suggests that they have comparable levels of African ancestry. In fact, Taforalt tends to cluster between Habeshas and the most SSA-admixed Maghrebis.

See for yourself...

https://science.sciencemag.org/content/sci/early/2018/03/14/science.aar8380/F2.large.jpg

https://3.bp.blogspot.com/-K3cd4ZD_xFE/WsnqwT96xsI/AAAAAAAAD0g/UsFtAwY8ZnYvrg95NxsBSjWoJJM1P4RjwCLcBGAs/s1600/PCAsupp.png

Mbuti is clearly not an appropriate reference population for the African ancestry in Taforalt, since the former has elevated basal African ancestry that isn’t indigenous to North Africa.

drobbah
08-27-2020, 04:28 PM
Taforalt clusters just north of the Afar/Agaw, who can be modeled as approximately 50:50 SSA/West Eurasian, in every global PCA I’ve seen since the original paper; this suggests that they have comparable levels of African ancestry. In fact, Taforalt tends to cluster between Habeshas and the most SSA-admixed Maghrebis.

See for yourself...

https://science.sciencemag.org/content/sci/early/2018/03/14/science.aar8380/F2.large.jpg

https://3.bp.blogspot.com/-K3cd4ZD_xFE/WsnqwT96xsI/AAAAAAAAD0g/UsFtAwY8ZnYvrg95NxsBSjWoJJM1P4RjwCLcBGAs/s1600/PCAsupp.png

Mbuti is clearly not an appropriate reference population for the African ancestry in Taforalt, since the former has elevated basal African ancestry that isn’t indigenous to North Africa.

They cluster north of the Eritreans too and those are the most Eurasian Horners (60% Natufian/Iran_N ancestry).Taforalt and MAR_EN at minimum based on how they behave on pca plots are around the 66%-70% range

gihanga.rwanda
08-27-2020, 04:41 PM
Is the Reich lab planning to release a second Dzudzuana paper or I am I making that up?

I want to see more research on the purported gene flow from North Africa into Natufians. In the original paper that models Taforalt as 45% ANA, Natufians can be modeled as nearly 30% Taforalt, which would make them about 15% ANA; it’s a shame that they didn’t include Natufians in any of the models in the Shum Laka paper. Time will tell.

Caius Agrippa
08-27-2020, 07:15 PM
I don't know if ANA is SSA or not, but being around 7-10% ''true'' SSA and ~3% ANA I still plot considerably closer to mainstream West Eurasians than all North Africans on any PCA plots, even the ones scoring only 2-3% West African. I'm actually not much distant from Canarians and some late medieval Iberians with 20% North African ancestry, while North Maghrebians generally plot south even of Egypt, a place that has more recent legit SSA ancestry. Iberomaurusian/Taforalt in great quantities definitely can shift someone toward Sub-Sahara. I also think the thread starter overstimate the impact of African DNA on phenotypes due to lack of familiarity with it. It's not like someone that is 20-25% SSA will look like Michael Jordan or ''black''. One of my grandparents was around 25% SSA and ''passed'' as white, that's why I score any SSA to begin with.

gihanga.rwanda
08-27-2020, 07:41 PM
Where's the north? Taforalt in the original Loosdrecht paper is more Eurasian shifted than Afar , also Taforalt in admixture is 65% Natufian / not 50%
Afar are about 55% Eurasian(Levant_Natufian+IranN ancestry) which means PCA is compatible with Admixture

here's a PCA of africans and ancient North africans , Taforalt is more Eurasian shifted than EarlyPastoral Neolithic who were more eurasian than modern somalis.
https://i.ibb.co/JQ79Nmt/newplot-10.png

as for the Mbuti your comment doesn't make much sense "has elevated basal African ancestry" ..Mbuti is an ancestry Basal to all Eurasians and its used to distinguish between Deep ancestry and Eurasian populations like Dzudzuana It is appropriate to be used and its actually compatible with admixture , maybe you mean the ANA (ANA is hypothetical ancestor)
This is the definition of Mbuti " The proportion of ‘Mbuti’ ancestry represents the total of ‘Deep’ ancestry from lineages that split prior to the split of Ust’Ishim, Tianyuan, and West Eurasians and can include both ‘Basal Eurasian’ andother (e.g., Sub-Saharan African) ancestry."
Which means it can even distinguish between elevated Basal eurasian in Dzudzuana and in Taforalt not just Sub-saharan african ancestry.
In the original Study you can see that Ethiopean Jews got higher Mbuti than Taforalt , this all tells us that ANA is not similar to any sub-saharan ancestry but holds a unique position.

I am really not understanding your train of thought.

1. I never said that Taforalt was 50% SSA, I was referring to the Afar and Agaw in my previous post. That being said, the Shum Laka paper was able to successfully model Taforalt as 54% ANA and Agaw as 59% East African agro-pastoralist + Mota. We’ll need more aDNA from the region to sort out the actual admixture proportions in these populations.

2. Taforalt is more “northern” than Afar or Agaw in the sense that it’s more Eurasian than these extant groups. I don’t think anyone in this thread has made the case that Taforalt is more African than these extant groups from the HoA. That being said, Taforalt clusters close enough to these groups to suggest it also derived from an ancient admixture event(s) between an indigenous African population and a Western Eurasian population. Taforalt behaves like an admixed SSA/Western Eurasian population, which suggests that ANA is related to some degree to other ancient populations that contributed to SSAs. I don’t know why we have to make this so complicated.

3. Taforalt is more than 10,000 years older than the Natufians. It no longer makes any sense to model Taforalt as part Natufian especially when the latter likely has significant North African ancestry. Hopefully we can get our hands on Dzudzuana, which we can then use to measure Western Eurasian ancestry in Africa.

4. I don’t have a problem with using Mbuti as a “trace dye” for African or basal Eurasian ancestry, but I definitely don’t think it can be used to measure actual admixture. The fact that Agaw shows more “Mbuti” is a non-issue for me for the reasons stated above.

5. What the Shum Laka paper demonstrated is that ANA is unlikely to be all that unique to Africa. This is not a new proposal and it’s been suggested for years but modern SSAs likely derive varying degrees of ancestry from both extremely basal branches of humanity and less basal branches closely related to the population that ultimately led to non-Africans. If the Shum Laka model stands firm, the main distinguisher between ANA and let’s say basal West African will be the basal African ancestry in the latter. That being said, we’ll need more aDNA to draw ironclad conclusions.

Nassbean
08-27-2020, 09:11 PM
I don't know if ANA is SSA or not, but being around 7-10% ''true'' SSA and ~3% ANA I still plot considerably closer to mainstream West Eurasians than all North Africans on any PCA plots, even the ones scoring only 2-3% West African. I'm actually not much distant from Canarians and some late medieval Iberians with 20% North African ancestry, while North Maghrebians generally plot south even of Egypt, a place that has more recent legit SSA ancestry. Iberomaurusian/Taforalt in great quantities definitely can shift someone toward Sub-Sahara. I also think the thread starter overstimate the impact of African DNA on phenotypes due to lack of familiarity with it. It's not like someone that is 20-25% SSA will look like Michael Jordan or ''black''. One of my grandparents was around 25% SSA and ''passed'' as white, that's why I score any SSA to begin with.

You plot next to which population exactly ? And it doesn't make sense because I'm well inside the west eurasian cluster and more northern shifted than egyptians (like many other coastal berbers). As for phenotypes the carribean area for example is full of quadroons (with low amerindian) and they still look clearly ssa influenced

Caius Agrippa
08-27-2020, 09:24 PM
You plot next to which population exactly ? And it doesn't make sense because I'm well inside the west eurasian cluster and more northern shifted than egyptians (like many other coastal berbers). As for phenotypes the carribean area for example is full of quadroons (with low amerindian) and they still look clearly ssa influenced

I plot near these guys:

Iberia_Southeast_c.10-16CE,0.0807196,0.1425973,0.0213387,-0.0267552,0.0343398,-0.0141537,-0.0099878,0.0001922,0.0278322,0.02609,0.0011097,0. 0037467,-0.0040013,-0.0112048,0.0084032,0.0006298,0.0013799,-0.0068833,-0.0100453,0.0010317,-0.001466,-0.0077489,0.0030196,-0.0049402,0.0003692

In a 3D PCA I'm always near these samples. Northern Moroccans, Algerians and Tunisians are considerably distant even of Egypt and Jordan. That's why I think Iberomaurusian shifts someone heavily to Africa. Look at some Canarians, they already begin to plot outside the Iberian cluster because of that.

I don't think phenotype comparisons are fruitful, but in my opinion people who are 75% European, 25% African don't look very ''black'' on average, I live in Brazil, I know what I'm talking about, I have daily experience with all types of admixture. I'm not saying ''quadroons'' look white on average, at this point of recent admixture it's alleatory and they can end up looking like anything (though most quadroons obviously look much more European than African) but the comparison with Berbers is absurd considering that the components and type of admixture are really different.

Nassbean
08-27-2020, 09:48 PM
I plot near these guys:

Iberia_Southeast_c.10-16CE,0.0807196,0.1425973,0.0213387,-0.0267552,0.0343398,-0.0141537,-0.0099878,0.0001922,0.0278322,0.02609,0.0011097,0. 0037467,-0.0040013,-0.0112048,0.0084032,0.0006298,0.0013799,-0.0068833,-0.0100453,0.0010317,-0.001466,-0.0077489,0.0030196,-0.0049402,0.0003692

In a 3D PCA I'm always near these samples. Northern Moroccans, Algerians and Tunisians are considerably distant even of Egypt and Jordan. That's why I think Iberomaurusian shifts someone heavily to Africa. Look at some Canarians, they already begin to plot outside the Iberian cluster because of that.

I don't think phenotype comparisons are fruitful, but in my opinion people who are 75% European, 25% African don't look very ''black'' on average, I live in Brazil, I know what I'm talking about, I have daily experience with all types of admixture. I'm not saying ''quadroons'' look white on average, at this point of recent admixture it's alleatory and they can end up looking like anything (though most quadroons obviously look much more European than African) but the comparison with Berbers is absurd considering that the components and type of admixture are really different.

That's basically plotting close to canarians and canarians aren't that far from us (north moroccans). Also yes maghrebis are distinct from egypt and jordan because we have way more EEF and yamnaya ancestry.

Here my position :

https://i.imgur.com/jFzbjcu.png?1

drobbah
08-27-2020, 09:50 PM
You plot next to which population exactly ? And it doesn't make sense because I'm well inside the west eurasian cluster and more northern shifted than egyptians (like many other coastal berbers). As for phenotypes the carribean area for example is full of quadroons (with low amerindian) and they still look clearly ssa influenced
Coastal Maghrebis aren't more Eurasian than Egyptians unless they are like that Morisco sample that Micheal posted in a pca.We had this argument before in your prior account here that got banned

ThaYamamoto
08-27-2020, 09:54 PM
That's basically plotting close to canarians and canarians aren't that far from us (north moroccans). Also yes maghrebis are distinct from egypt and jordan because we have way more EEF and yamnaya ancestry.

Here my position :

I can add you to the non-african/african cline PCA if you like, see if you're closer to tropical Africans than an Egyptian (provided you have coords?). Universal PCAs for this kinda thing is defunct don't you think?

Caius Agrippa
08-27-2020, 09:55 PM
That's basically plotting close to canarians and canarians aren't that far from us (north moroccans). Also yes maghrebis are distinct from egypt and jordan because we have way more EEF and yamnaya ancestry.

Here my position :

https://i.imgur.com/jFzbjcu.png?1

Canarians are significantly more Eurasian than any North Africans, they are 15-30% North African depending on the island, the remaining admixture is European. These plots of K15 aren't that useful anymore, I plot in between Sicilians, Tuscans and Sardinians in them haha.

Nassbean
08-27-2020, 10:07 PM
Coastal Maghrebis aren't more Eurasian than Egyptians unless they are like that Morisco sample that Micheal posted in a pca.We had this argument before in your prior account here that got banned

I'm more eurasian shifted than egyptians and you saw it yourself (I'm not from a morisco background)

Nassbean
08-27-2020, 10:11 PM
Canarians are significantly more Eurasian than any North Africans, they are 15-30% North African depending on the island, the remaining admixture is European. These plots of K15 aren't that useful anymore, I plot in between Sicilians, Tuscans and Sardinians in them haha.

I never said that they were as eurasian as us ...?? It's not the first time I see a latino being so offensive with me don't worry I'm not saying that I'm close to iberians

Nassbean
08-27-2020, 10:13 PM
Based on the recent answers it seems I'm around 13-17% indigenous african but only 4-6% SSA

Caius Agrippa
08-27-2020, 10:14 PM
I never said that they were as eurasian as us ...?? It's not the first time I see a latino being so offensive with me don't worry I'm not saying that I'm close to iberians

I'm not being offensive with you man. Don't get what you're trying to say.

drobbah
08-27-2020, 10:31 PM
I'm more eurasian shifted than egyptians and you saw it yourself (I'm not from a morisco background)
That K15 isn't a real pca plot and coastal maghrebis aren't more Eurasian than the average coptic or muslim Egyptians.Every pca plot I've seen shows the Egyptians plotting near the Arabians and Southern Levantines.Which makes sense considering muslim egyptians are on average 11% African (Dinka + West African) while Maghrebis range between 15-30% African (West African + ANA).

Nassbean
08-27-2020, 10:37 PM
That K15 isn't a real pca plot and coastal maghrebis aren't more Eurasian than the average coptic or muslim Egyptians.Every pca plot I've seen shows the Egyptians plotting near the Arabians and Southern Levantines.Which makes sense considering muslim egyptians are on average 11% African (Dinka + West African) while Maghrebis range between 15-30% African (West African + ANA).

Not only the K15 PCA ..take any PCA you want and I'll be more northern shifted than the average egyptian (I can send you my coordinates). Also you overestimate the SSA of maghrebis it can easily be below 15% for riffians and kabyles.

Caius Agrippa
08-27-2020, 10:48 PM
A 3D PCA is more accurate than those 2D PCAs and here you can clearly see where the populations are located in a more ''worldwide'' scale.

Look at the positions of Egyptians and Maghrebis:

https://i.postimg.cc/28kXM8CM/Captura-de-Tela-107.png

Nassbean
08-27-2020, 11:02 PM
Most Maghrebians (North and south) are more SSA than Muslim Egyptians , even if we exclude the african ancestry found in Iberomaurusians...but your ssa is lower than other Moroccans.

Honestly that's not what I saw most riffians have even less ssa than me and when we say "egyptians" are we talking about the ones from the Delta ? Because I really doubt upper egyptians have less SSA admixture than us. Anyway it doesn't change the fact that I'm indeed more northern shifted than the average egyptians

Nassbean
08-27-2020, 11:05 PM
A 3D PCA is more accurate than those 2D PCAs and here you can clearly see where the populations are located in a more ''worldwide'' scale.

Look at the positions of Egyptians and Maghrebis:




I invite you to reread what I said I was specifically talking about me

Moldovlah
08-27-2020, 11:06 PM
Coastal Maghrebis aren't more Eurasian than Egyptians unless they are like that Morisco sample that Micheal posted in a pca.We had this argument before in your prior account here that got banned

African admixture is similar in both peoples, the north coastal Moroccans are trace higher in modern West African lineage than Egyptians you might trace it to their influence in Sahel but is not that much difference is very close.

drobbah
08-27-2020, 11:16 PM
African admixture is similar in both peoples, the north coastal Moroccans are trace higher in modern West African lineage than Egyptians you might trace it to their influence in Sahel but is not that much difference is very close.
Egyptians are 11% African vs 15% in coastal Maghrebis.The source of their African is different as well, with Nilo-Saharan ancestry in Egyptians and ANA in Maghrebis but both have Yoruba ancestry due to the Islamic slave trade.Never said there was a huge gap but to pretend the people of the Nile Valley and the coastal Maghreb are carbon copies of each other or to believe Nass's claims of Egyptians being more African than coastal Maghrebis when there's nothing to back such claims is just plain wrong.

Nassbean
08-27-2020, 11:28 PM
Egyptians are 11% African vs 15% in coastal Maghrebis.The source of their African is different as well, with Nilo-Saharan ancestry in Egyptians and ANA in Maghrebis but both have Yoruba ancestry due to the Islamic slave trade.Never said there was a huge gap but to pretend the people of the Nile Valley and the coastal Maghreb are carbon copies of each other or to believe Nass's claims of Egyptians being more African than coastal Maghrebis when there's nothing to back such claims is just plain wrong.

So you're implying that ANA is fully SSA ? If yes what are your evidence for this ? You're putting nilotic/yoruba ancestry at the same level as ANA ...You still can't explain why I'm more northern shifted than egyptians.

Moldovlah
08-27-2020, 11:30 PM
Egyptians are 11% African vs 15% in coastal Maghrebis.The source of their African is different as well, with Nilo-Saharan ancestry in Egyptians and ANA in Maghrebis but both have Yoruba ancestry due to the Islamic slave trade.Never said there was a huge gap but to pretend the people of the Nile Valley and the coastal Maghreb are carbon copies of each other or to believe Nass's claims of Egyptians being more African than coastal Maghrebis when there's nothing to back such claims is just plain wrong.

I don't think they same people, they are not, but their SSA in a model was similar. I think the elevated Natufian admixture in Egyptians opposed to the ANA as you saying in the coastal Moroccans is also part African and the Egyptians have mediated this admixture down the Nile valley and potentially beyond that more than any other population in Africa.

I had some models before, I closed the window now tho but maybe after I post it and if there is some mistake in my model you point the errors or accuracy if necessary.

Moldovlah
08-27-2020, 11:35 PM
So you're implying that ANA is fully SSA ? If yes what are your evidence for this ? You're putting nilotic/yoruba ancestry at the same level as ANA ...You still can't explain why I'm more northern shifted than egyptians.

Shifted is not important. I'm not trying argue. Just if I take a half Finnish half African American person (I have a sample actually like this) they might plot more north of you but still be more SSA. Plot is art project, is not any real science or history involved. It is dependent on different variables.

ThaYamamoto
08-27-2020, 11:59 PM
This results is of Cairo egyptians who are mixed south/north+ high slave trade influenced, native population of the delta have lower ssa , and even towns in northern half of south egypt have lower ssa than cairo egyptians.

Do you know if the exact regions/ethnic groups of the slaves brought into Egypt have been recorded?

pegasus
08-28-2020, 12:02 AM
A 3D PCA is more accurate than those 2D PCAs and here you can clearly see where the populations are located in a more ''worldwide'' scale.

Look at the positions of Egyptians and Maghrebis:

https://i.postimg.cc/28kXM8CM/Captura-de-Tela-107.png


OP's claims were ridiculous and petty, its pretty evident all populations with significant Iberomaurusian ancestry will pull markedly towards SSA related populations.

Adamm
08-28-2020, 12:13 AM
Egyptians are 11% African vs 15% in coastal Maghrebis.The source of their African is different as well, with Nilo-Saharan ancestry in Egyptians and ANA in Maghrebis but both have Yoruba ancestry due to the Islamic slave trade.Never said there was a huge gap but to pretend the people of the Nile Valley and the coastal Maghreb are carbon copies of each other or to believe Nass's claims of Egyptians being more African than coastal Maghrebis when there's nothing to back such claims is just plain wrong.


but both have Yoruba ancestry due to the Islamic slave trade

On what kind of scale? And how can we identify this Yoruba ancestry?

ThaYamamoto
08-28-2020, 12:19 AM
in cairo mainly west africans were brought, in South egypt were mostly Sudanic nilotic slaves and settled in towns later , according to historians some towns became fully intermixed due to there high numbers.

Interesting, do we know of specific ethnic groups or regional descriptors?

drobbah
08-28-2020, 12:49 AM
the sudanic nilotic were mainly from Kordofan and Darfur and were sold by the Arab sudanic slave traders who hunted dinka and shilluk..
also from regions surrounding chad slaves from there were sold by the Libyan bedouins and brought to cities in middle egypt.
Both those ancient egyptian samples had minor Dinka ancestry

drobbah
08-28-2020, 01:00 AM
So What?
Be respectful bud.

You are making it seem like the Nilotic ancestry found in modern Egyptians was because of the Islamic slave trade when both the Hellenistic sample and the Late Period one shows that minor Ancient Saharan (Dinka/Sudanese are a good proxy for this) ancestry has been in Egypt for extremely long time and probably prior to the formation of the Egyptian state.Others have let me know that the sample is of low quality but these results don't seem out of the ordinary.

Target: EGY_Late_Period:JK2134
Distance: 3.2750% / 0.03274991
51.0 Levant_PPNC
16.8 Levant_Natufian
12.4 TUR_Barcin_C
12.2 IRN_Ganj_Dareh_N
4.0 TUR_Barcin_N
3.2 Dinka
0.4 Yamnaya_RUS_Samara

Target: EGY_Hellenistic:JK2888
Distance: 4.4448% / 0.04444801
43.6 Levant_Natufian
17.8 IRN_Ganj_Dareh_N
17.0 TUR_Barcin_N
11.4 Levant_PPNC
8.0 TUR_Barcin_C
2.2 Dinka

drobbah
08-28-2020, 01:20 AM
How can we assume these Northern samples from the Hyksos era (Semitic Levantine influx) encompasses the entire genetic history of Egypt.We both don't know how Neolthic Egypt looked genetically or if the Upper Egyptians always had higher levels of Ancient Saharan ancestry perhaps due to contact with the Cushitic populations that lived in Upper Nubia (C-group culture) and the Eastern Desert (Medjays) prior to the Sudanic migrations (Nubians).

Brandon S. Pilcher
08-28-2020, 01:56 AM
Since the topic came up again...

It might be too early to say anything conclusive without a larger variety of aDNA samples from Egypt and northern Sudan, but I believe at the very least a large chunk of earlier Egypto-Nubian ancestry is going to be native to northeastern Africa itself. That component probably will have some OOA genetic affinity since Northeast Africa would have been the jumping-off point for OOA, but it'd still be native African rather than Eurasian in origin. In that respect, it'd be similar to ANA or "Basal Eurasian" (the latter also present in Natufian, Neolithic Levantine, and EEF populations, of course). Not all indigenous African ancestry is going to look West/Central Africa or "SSA".

Humans Trekked Out of Africa Via Egypt, Study Suggests (https://www.livescience.com/51005-humans-migrated-out-of-egypt.html)

To see which route the ancestors of all humans outside of Africa might have taken, the researchers sequenced the genomes of 225 people from northeast Africa — 100 Egyptians and 125 Ethiopians. They then compared this data with DNA from East Asians, South Asians and Europeans — specifically, Han Chinese, Gujarati Indians and Tuscan Italians, respectively. They also compared this data with DNA from modern West Africans from south of the Sahara, which should generally reflect the ancient sub-Saharan gene pool.

The scientists noted that both modern Egyptians and Ethiopians have recently experienced migrations from outside Africa, and the interbreeding that resulted might increase their genetic similarity with those migratory people. To account for this, the researchers removed any genetic sequences that might have come from these recent migrations.

If the southern route was the main path out of Africa, Ethiopians should be more genetically similar to Eurasians. Instead, the researchers found that Egyptians were more genetically similar to Eurasians, suggesting the northern route was the predominant way out of Africa. The researchers estimated that Eurasians genetically diverged from Egyptians 55,000 years ago, Ethiopians 65,000 years ago and West Africans 75,000 years ago.

Tsakhur
08-28-2020, 03:20 AM
Be respectful bud.

You are making it seem like the Nilotic ancestry found in modern Egyptians was because of the Islamic slave trade when both the Hellenistic sample and the Late Period one shows that minor Ancient Saharan (Dinka/Sudanese are a good proxy for this) ancestry has been in Egypt for extremely long time and probably prior to the formation of the Egyptian state.Others have let me know that the sample is of low quality but these results don't seem out of the ordinary.

Target: EGY_Late_Period:JK2134
Distance: 3.2750% / 0.03274991
51.0 Levant_PPNC
16.8 Levant_Natufian
12.4 TUR_Barcin_C
12.2 IRN_Ganj_Dareh_N
4.0 TUR_Barcin_N
3.2 Dinka
0.4 Yamnaya_RUS_Samara

Target: EGY_Hellenistic:JK2888
Distance: 4.4448% / 0.04444801
43.6 Levant_Natufian
17.8 IRN_Ganj_Dareh_N
17.0 TUR_Barcin_N
11.4 Levant_PPNC
8.0 TUR_Barcin_C
2.2 Dinka

I can tried to make them a bit more African than that which is around 4.4%. What's odd is the Hellenistic individual seems to show minor West African affinity (Esan_Nigeria) at 1.2%, and its fascinating that both individuals need minor Steppe admixture to improve the fits.

Target: EGY_Late_Period:JK2134
Distance: 3.1484% / 0.03148366
39.2 Levant_PPNC
28.2 Levant_ISR_C
16.2 IRN_Seh_Gabi_C
8.4 Levant_Natufian
4.4 Dinka
3.6 Yamnaya_RUS_Samara

Target: EGY_Hellenistic:JK2888
Distance: 4.4345% / 0.04434529
53.0 Levant_ISR_C
23.8 Levant_Natufian
16.0 IRN_Seh_Gabi_C
3.2 Dinka
2.8 Yamnaya_RUS_Samara
1.2 Esan_Nigeria


Even adding Tarofalt doesn't impact the amount of SSA ancestry in the two samples:

Target: EGY_Late_Period:JK2134
Distance: 3.1155% / 0.03115477
38.8 Levant_ISR_C
32.6 Levant_PPNC
14.4 IRN_Seh_Gabi_C
4.2 Dinka
4.2 Levant_Natufian
3.6 Yamnaya_RUS_Samara
2.2 MAR_Taforalt

Target: EGY_Hellenistic:JK2888
Distance: 4.4300% / 0.04429967
54.8 Levant_ISR_C
21.6 Levant_Natufian
15.6 IRN_Seh_Gabi_C
3.4 Dinka
2.8 Yamnaya_RUS_Samara
1.0 Esan_Nigeria
0.8 MAR_Taforalt

Hawk
08-28-2020, 05:29 AM
Peer reviewed papers declare completely something else, people here want to believe something else.

From Dzudzuana paper.

https://www.biorxiv.org/content/biorxiv/early/2018/09/20/423079/F2.large.jpg

From Shum Laka paper.



We can also obtain a good fit for the Shum Laka individuals in a less parsimonious alternative model using three components, replacing the
basal West African source with a combination of ancestry from inside the
clade defined by the other West African populations and from a source
entirely outside the West African clade (near one lineage that contributes to the Taforalt individuals)
(Extended Data Fig. 5, Supplementary
Information section 3). However, two-component models for the Shum
Laka individuals that have the majority source splitting closer to other
West or East Africans are rejected (Z = 7.1 and Z = 3.7, respectively).

https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/2020_Lipson_Nature_ShumLaka.pdf


Do you guys understand what entirely outside the clade means or what?

Tsakhur
08-28-2020, 05:40 AM
Egyptians are 11% African vs 15% in coastal Maghrebis.The source of their African is different as well, with Nilo-Saharan ancestry in Egyptians and ANA in Maghrebis but both have Yoruba ancestry due to the Islamic slave trade.Never said there was a huge gap but to pretend the people of the Nile Valley and the coastal Maghreb are carbon copies of each other or to believe Nass's claims of Egyptians being more African than coastal Maghrebis when there's nothing to back such claims is just plain wrong.

Here are top four most African-shifted Egyptians from G25 (not counting Tarofalt as I am still unsure what type of "African" it really is). Four out of five individuals show minor Steppe affinity (Yamnaya) from 1.4-5.4% percent. Makes me wonder if the Steppe comes from the Hyksos and other foreign intrusion into ancient Egypt.

19% SSA (mainly Dinka with some West African (Esan) and mimor Pygmy (Bedzan))
Target: Egyptian:7AJ129
Distance: 3.1192% / 0.03119246
47.2 Levant_ISR_C
17.4 IRN_Seh_Gabi_C
13.8 Dinka
9.6 Levant_Natufian
5.4 Yamnaya_RUS_Samara
4.2 Esan_Nigeria
1.2 MAR_Taforalt
1.0 Bedzan
0.2 WHG

14.6% SSA (primarily Dinka with some West African (Esan) and also minor Pygmy (Bedzan))
Target: Egyptian:Egypt8AT113
Distance: 2.6081% / 0.02608071
35.4 Levant_ISR_C
19.6 IRN_Hasanlu_IA
10.0 Levant_Natufian
9.0 Levant_PPNC
8.0 Dinka
7.4 IRN_Seh_Gabi_C
4.0 Esan_Nigeria
2.6 Bedzan
2.6 MAR_Taforalt
1.4 Yamnaya_RUS_Samara

14.2% SSA (predominantly Dinka with negligible West African)
Target: Egyptian:Egypt9AQ177
Distance: 1.9150% / 0.01914994
45.4 Levant_ISR_C
25.4 IRN_Hasanlu_IA
13.4 Dinka
6.4 IRN_Seh_Gabi_C
6.2 MAR_Taforalt
1.4 WHG
1.0 Levant_Natufian
0.8 Esan_Nigeria

14.2% SSA (only Dinka)
Target: Egyptian:4AJ137
Distance: 3.0081% / 0.03008061
40.2 Levant_ISR_C
19.8 IRN_Seh_Gabi_C
14.8 Levant_PPNC
14.2 Dinka
5.6 MAR_Taforalt
5.2 Yamnaya_RUS_Samara
0.2 TUR_Barcin_N

14% SSA (predominantly Dinka with some West African)
Target: Egyptian:2AJ137
Distance: 2.0542% / 0.02054221
57.8 Levant_ISR_C
12.4 IRN_Seh_Gabi_C
10.0 Dinka
7.0 Levant_Natufian
5.4 Yamnaya_RUS_Samara
4.0 Esan_Nigeria
3.4 MAR_Taforalt

...............

Average Egyptian G25 result: 13% African (Dinka+some West African). Around 6% Steppe (probably from Semitic Levantine migrants into Egypt).

Target: Egyptian
Distance: 1.0988% / 0.01098760
54.4 Levant_ISR_C
15.2 IRN_Seh_Gabi_C
10.4 Dinka
6.0 Yamnaya_RUS_Samara
5.4 MAR_Taforalt
3.2 Levant_Natufian
2.8 IRN_Hasanlu_IA
2.6 Esan_Nigeria

Tsakhur
08-28-2020, 07:21 AM
They cluster north of the Eritreans too and those are the most Eurasian Horners (60% Natufian/Iran_N ancestry).Taforalt and MAR_EN at minimum based on how they behave on pca plots are around the 66%-70% range

Here is the most West Eurasian-shifted Eritrean individual who plot slightly more north than the average Tarofalt and this MAR_EN individual (MAR_EN:IAM.5)

Eritrean:Eritrean_2,-0.175288,0.114755,-0.027153,-0.094316,0.001231,-0.039602,-0.00893,0.000923,0.094899,-0.061231,0.002436,-0.01079,0.025272,0.00812,0.020222,-0.001193,0.007041,-0.003421,0.008673,0.002876,0.006489,0.006554,-0.006286,-0.005181,-0.002275

MAR_Taforalt,-0.189857,0.0814452,-0.0242866,-0.085595,0.027636,-0.0552202,-0.0705968,0.0184146,0.155397,0.003499,0.0209156,-0.0318316,0.0747168,-0.0513334,0.0711988,-0.0363032,0.0052676,-0.066106,-0.1424162,0.0389938,-0.0376836,-0.1255322,0.0730118,-0.0137606,0.0164534

MAR_EN:IAM.5,-0.180979,0.091398,-0.018856,-0.090117,0.030159,-0.05773,-0.076848,0.021461,0.154825,-0.001458,0.024521,-0.025178,0.078047,-0.047755,0.07356,-0.034738,0.007171,-0.060811,-0.147695,0.044771,-0.041053,-0.121427,0.083439,-0.007109,0.027782

Target: Eritrean:Eritrean_2: around 64% Eurasian and 36% African.
Distance: 4.2933% / 0.04293325
44.8 Levant_Natufian
32.4 Dinka
16.4 Levant_JOR_EBA
3.4 ETH_4500BP
3.0 GEO_CHG

So yep agreed. Tarofalt and MAR_EN is probably in that range.

beyoku
08-28-2020, 11:59 AM
I can tried to make them a bit more African than that which is around 4.4%. What's odd is the Hellenistic individual seems to show minor West African affinity (Esan_Nigeria) at 1.2%, and its fascinating that both individuals need minor Steppe admixture to improve the fits.

Target: EGY_Late_Period:JK2134
Distance: 3.1484% / 0.03148366
39.2 Levant_PPNC
28.2 Levant_ISR_C
16.2 IRN_Seh_Gabi_C
8.4 Levant_Natufian
4.4 Dinka
3.6 Yamnaya_RUS_Samara

Target: EGY_Hellenistic:JK2888
Distance: 4.4345% / 0.04434529
53.0 Levant_ISR_C
23.8 Levant_Natufian
16.0 IRN_Seh_Gabi_C
3.2 Dinka
2.8 Yamnaya_RUS_Samara
1.2 Esan_Nigeria


Even adding Tarofalt doesn't impact the amount of SSA ancestry in the two samples:

Target: EGY_Late_Period:JK2134
Distance: 3.1155% / 0.03115477
38.8 Levant_ISR_C
32.6 Levant_PPNC
14.4 IRN_Seh_Gabi_C
4.2 Dinka
4.2 Levant_Natufian
3.6 Yamnaya_RUS_Samara
2.2 MAR_Taforalt

Target: EGY_Hellenistic:JK2888
Distance: 4.4300% / 0.04429967
54.8 Levant_ISR_C
21.6 Levant_Natufian
15.6 IRN_Seh_Gabi_C
3.4 Dinka
2.8 Yamnaya_RUS_Samara
1.0 Esan_Nigeria
0.8 MAR_Taforalt

Where do modern /bronze aged Levantines, Egyptians, Mandinka, Dinka and Horners fit into an analysis like this when it comes to their MAR_Taforalt levels?

Tsakhur
08-28-2020, 03:56 PM
can you explain to me this model ? why PPNB and Levant EBA and Yemenite mahra show Dinka admixture ?
https://i.ibb.co/4KR3mY8/Debunker.jpg

I am not really sure as I am not that knowledgeable on Middle Eastern genetics but I think it's means the aformentioned groups might have ancient African/African-like ancestry.

maroco
08-28-2020, 04:08 PM
Well according to my k5 admixture I’m about 24 percent ssa
39211

maroco
08-28-2020, 05:01 PM
nope they don't have any ssa ..And that's the real problem , the person who is making G25 models should be fully aware of the genetic component of this populations
what you did in your models is called overfitting we know that this ancient near easterners had high Natufian ancestry when you over-fit
with many mixed references the population became starved and don't get his actual ancient Natufian or IranN ancestry, this overfitting force the remaining natufian or iran like components to go dinka or yamnaya when its not.
ancient and modern near easterners must get a proper models with basic ancestries..because overfitting gives them misleading results.
like this model , this is a correct one , since this populations were rich in Natufian ancestry they are getting there actual ancestry while not been starved the yamnaya now only appears in the mixed Egyptian-Levantine son from Beirut but not from his mother or abusir mummies( compatible with the original Admixture from both studies)
the dinka disappeared from PPNB and Yemenite mahra who were getting about 4-2% in the wrong model..the gap in ssa from earlier model and this one between egyptians and PPNB is identical about 2%
https://i.ibb.co/71PDGVs/image.png
Your distances are terrible

Tsakhur
08-28-2020, 05:11 PM
nope they don't have any ssa ..And that's the real problem , the person who is making G25 models should be fully aware of the genetic component of this populations
what you did in your models is called overfitting we know that this ancient near easterners had high Natufian ancestry when you over-fit
with many mixed references the population became starved and don't get his actual ancient Natufian or IranN ancestry, this overfitting force the remaining natufian or iran like components to go dinka or yamnaya when its not.
ancient and modern near easterners must get a proper models with basic ancestries..because overfitting gives them misleading results.
like this model , this is a correct one , since this populations were rich in Natufian ancestry they are getting there actual ancestry while not been starved the yamnaya now only appears in the mixed Egyptian-Levantine son from Beirut but not from his mother or abusir mummies( compatible with the original Admixture from both studies)
the dinka disappeared from PPNB and Yemenite mahra who were getting about 4-2% in the wrong model..the gap in ssa from earlier model and this one between egyptians and PPNB is identical about 2%
https://i.ibb.co/71PDGVs/image.png


You mean my models for Hellenistic and Late Period individuals were overfitted? Nope, you are wrong. If it overfitted, it has to be something like less than 0.50 like 0.15 or 0.02, not 3 to 4 which is what I got for my modelling.

And no, I disagree. The more population sources you utilized, the more accurate your model would be, as the distances becomes less which means you are in the right direction.

ThaYamamoto
08-28-2020, 05:27 PM
Your distances are terrible


You mean my models for Hellenistic and Late Period individuals were overfitted? Nope, you are wrong. If it overfitted, it has to be something like less than 0.50 like 0.15 or 0.02, not 3 to 4 which is what I got for my modelling.

And no, I disagree. The more population sources you utilized, the more accurate your model would be, as the distances becomes less which means you are in the right direction.

Don't care for Egypt=/=SSA discussions interested as everyone else is when the studies are finally published. But you're both wrong, the euclidean distances here mean nothing as each component will compensate for another on the PCA. The amount of shared allele frequencies between populations is staggering, so you're going off of unique/rare allele frequencies that make it easy to fit anything , anywhere. Closer fit doesn't mean in the right direction at all. This is why qpadm stats are taken seriously but only with the appropriate outgroups etc opposed to G25. Cross validation error increases exponentially after K4/5 so more pops=better distance=greatest conclusion isn't right.

Tsakhur
08-28-2020, 05:56 PM
Don't care for Egypt=/=SSA discussions interested as everyone else is when the studies are finally published. But you're both wrong, the euclidean distances here mean nothing as each component will compensate for another on the PCA. Please use G25 in the manner in which it is to be used. The amount of shared allele frequencies between populations is staggering, so you're going off of unique/rare allele frequencies that make it easy to fit anything , anywhere. Closer fit doesn't mean in the right direction at all. This is why qpadm stats are taken seriously but only with the appropriate outgroups etc opposed to G25. Cross validation error increases exponentially after K4/5 so why you think more pops=better distance=greatest conclusion is beyond me.

Tbh, I don't really care about this Egypt=/=African subject either. I was just playing to see how much SSA I can get for the two ancient Egyptian individuals but it just for fun, nothing else.

What should G25 be used for then?

And thanks for correcting me on my understanding of the subject of "fitting distance", I never know about this type of stuff before. These are all new information for me to learn.

ThaYamamoto
08-28-2020, 06:03 PM
Tbh, I don't really care about this Egypt=/=African subject either. I was just playing to see how much SSA I can get for the two ancient Egyptian individuals but it just for fun, nothing else.

What should G25 be used for then?

And thanks for correcting me on my understanding of the subject of "fitting distance", I never know about this type of stuff before. These are all new information for me to learn.

No problem I'm no expert I gotta say and there's others who'd break it down far better than me and prolly point out a lot of errors that I've made. Sorry if I came across harsh or anything ;)

Tsakhur
08-28-2020, 06:08 PM
No problem I'm no expert I gotta say and there's others who'd break it down far better than me and prolly point out a lot of errors that I've made. Sorry if I came across harsh or anything ;)

You do sound a bit harsh at first but it's ok.

Moderator
08-28-2020, 10:00 PM
Troll post removed. Members who send troll posts will be imposed sanctions based on the terms of service of our forum.

beyoku
08-28-2020, 10:06 PM
To those running the numbers.... So is Abusir Less African than modern Egypt? Less SSA and Less North African? Are they less SSA and North African than their Near Eastern contemporaries?

Moldovlah
08-28-2020, 10:07 PM
My run, almost 10% SSA. Mother can be as high 18%, Father 3% in various admixture runs.

39224

Adamm
08-28-2020, 10:35 PM
This thread has been going off topic for a while, it was about SSA in North Africans (Maghrebis) and not about Egyptians or about some personal runs.

Moldovlah
08-28-2020, 10:36 PM
This thread has been going off topic for a while, it was about SSA in North Africans (Maghrebis) and not about Egyptians or about some personal runs.

Exactly, I am North African.

Brwn_trd
08-29-2020, 01:29 AM
This book here might be of interest to some of you. It's archeology, but has a good discussion the origins of iberomaurusians in a short paragraph at the end of the last chapter. Section 18.7. Bottom line is, we don't know enough but might be interesting to those who want to know more about these people.

It's quite comprehensive material on the Taforalt site and is current.

Cemeteries and Sedentism in the Later Stone Age of NW Africa: Excavations at Grotte des Pigeons, Taforalt, Morocco (https://books.ub.uni-heidelberg.de/propylaeum/catalog/book/734)

Mnemonics
08-29-2020, 03:44 AM
To those running the numbers.... So is Abusir Less African than modern Egypt? Less SSA and Less North African? Are they less SSA and North African than their Near Eastern contemporaries?

Yes, JK2911 which is the sample with the highest coverage and without significant contamination seems to be no more African/Basal than BedouinB and Yemenite Jews, it has roughly 12 percent of some form of ancestry that exists somewhere on the cline between Dinka-like Africans and Eurasians (Although it is almost certainly much closer to Dinka.), which is significantly less than the 30-25 percent found in the Mozabite and Saharawi samples in the Reichlabs dataset.