PDA

View Full Version : 4500-year-old Ethiopian - Y-DNA E1b1, mtDNA L3, and considerable Eurasian admixture



Passa
10-08-2015, 07:12 PM
http://www.sciencemag.org/content/early/2015/10/07/science.aad2879.abstract

Characterizing genetic diversity in Africa is a crucial step for most analyses reconstructing the evolutionary history of anatomically modern humans. However, historic migrations from Eurasia into Africa have affected many contemporary populations, confounding inferences. Here, we present a 12.5x coverage ancient genome of an Ethiopian male (‘Mota’) who lived approximately 4,500 years ago. We use this genome to demonstrate that the Eurasian backflow into Africa came from a population closely related to Early Neolithic farmers, who had colonized Europe 4,000 years earlier. The extent of this backflow was much greater than previously reported, reaching all the way to Central, West and Southern Africa, affecting even populations such as Yoruba and Mbuti, previously thought to be relatively unadmixed, who harbor 6-7% Eurasian ancestry.

MfA
10-08-2015, 07:16 PM
http://abload.de/img/cqn1swivaaekm-j1kypql.jpg

http://abload.de/img/cjwbp0_wiaecdqa1rooh.jpg

MfA
10-08-2015, 07:18 PM
Principal component analysis shows that Ari and Sandawe are the
closest contemporary populations to Mota

http://abload.de/img/jp2launcher_2015_10_0bgpv7.png


Mota was placed close to the Ethiopian samples (Fig. 1, A), in between the clusters formed by the
Ari and the Sandawe (but very close to an Ari individual that stands out from the rest of that
group). The Ari can be split into two castes, Ari Cultivator and Ari Blacksmith, which share a
common origin within the last 4,500 years (62). Since data on a larger number of SNPs are
available for Ethiopian populations (4), we repeated the PCA using this higher quality dataset
(processed as in SM S6), which gave us 484,161 usable SNPs that could be called in Mota. Once
again, Mota fell in between the Ari and the Sandawe cluster (Fig. S5).



S9. f4 ratio analysis shows that Mota has no component of West
Eurasian admixture


The absence of a West Eurasian component in Mota supports the dating of the
backflow into Africa, which, at ~3.5kya, is younger than our ancient genome (dated to 4.5 kya)

Given that Mota predates the backflow, it potentially provides a better unadmixed African
reference than contemporary Yoruba. Thus, we recomputed the extent of the West Eurasian
component in contemporary African populations using Mota, λMota,Druze, instead of Yoruba in our f4
ratio. By using this better reference, we estimated West Eurasian admixture to be significantly
larger than previously estimated, with an additional 6-9% of the genome of contemporary African
populations being of Eurasian origin (Fig. S6, and Table S5). Importantly, this analysis shows that
the West Eurasian component can be found also in West Africa (Fig. S6), albeit at lower levels
than in Eastern Africa. Importantly, a sizeable West Eurasian component is also found in the
Yoruba and Mbuti, which are often used a representative of an unadmixed African population.



S10. Admixture f3 statistics show that the West Eurasian component
originated from a population similar to the early Neolithic farmers





S11. D statistics and f4 ratios show that Mota has no discernible
Neanderthal component, as opposed to modern-day Africans.

J Man
10-08-2015, 07:27 PM
E1b1 in Africa coming from Near Eastern Neolithic farmers (West Eurasians)?

Shaikorth
10-08-2015, 07:51 PM
E1b1 in Africa coming from Near Eastern Neolithic farmers (West Eurasians)?

No because Mota does not have West Eurasian (or Neanderthal or Denisovan) admixture. Modern East Africans more apparently do, but the haplogroup seems native in light of this.

Passa
10-08-2015, 08:04 PM
His Y-DNA was probably E-M329, a 30-40 ky old lineage indigenous to the Horn of Africa.

The authors date the Eurasian admixture at less than 4500 years ago, but I think that this is fairly unbelievable, the first reason being that there are not demonstrably recent lineages shared between all Sub-Saharan Africans, except obviously haplogroup E.

Lank
10-08-2015, 08:06 PM
Very, very intriguing results! This goes to show, yet again, that without ancient DNA we will only have a very limited understanding of human history. Now the question is, how do we make sense of the widespread, apparently recent Eurasian admixture in SSA outside East Africa?

In light of Mota's autosomal similarity with Omotics, it seems likely that his E1b1 lineage will belong to E-M329, a distant (40 ky old) relative of West-Central African E-M2 that is highly frequent in modern Omotic populations.

Oh, and my dad belongs to mtDNA L3x2a, like Mota. :)



The authors date the Eurasian admixture at less than 4500 years ago, but I think that this is fairly unbelievable, the first reason being that there are not demonstrably recent lineages shared between all Sub-Saharan Africans, except obviously haplogroup E.
Yeah, I thought so too. Within East Africa it's not too surprising, but it's really difficult to make sense of the more widespread Eurasian signal.

ADW_1981
10-08-2015, 08:06 PM
I wonder if any of this has anything to do with R1b-V88, despite this particular guy not carrying that signature. I recall the recent analysis on the split between the St Maarten V88 and Sardinians was around 6500 BC. R1b-V88 has also been found in Sardinia, as well as an Iberian farmer carrying a SNP also in the same branch. The timeline fits well with the Neolithic wet Saharan period too.

Lank
10-08-2015, 08:14 PM
The table of Eurasian admixture estimates using Yoruba and Mota as the reference. Note how admixture levels only decrease in some Ethiopian populations (Wolayta, Afar, Somali, Oromo, Tigray, Amhara, Jews) using Mota as a reference instead of Yoruba. In all other African populations, Eurasian admixture levels rise using Mota.

http://i60.tinypic.com/2elcj7n.png


I think ancient DNA from other parts of Africa is needed to really make sense of these results, although it's great that we finally have our first sample. Exciting times ahead!

Shaikorth
10-08-2015, 08:18 PM
The table of Eurasian admixture estimates using Yoruba and Mota as the reference. Note how admixture levels only decrease in some Ethiopian populations (Wolayta, Afar, Somali, Oromo, Tigray, Amhara, Jews) using Mota as a reference instead of Yoruba. In all other African populations, Eurasian admixture levels rise using Mota, rather significantly in some cases.

http://i60.tinypic.com/2elcj7n.png

Yeah, would have been nice to see Ju'hoan reference and Mbuti reference tests for comparison since they are more distant from Eurasians than Yoruba.

Lank
10-08-2015, 08:30 PM
Mota has a notable 'Eurasian' (Proto-Eurasian?) pull in the PCA, but nothing in the formal stats...

http://i58.tinypic.com/2emei6o.png

Passa
10-08-2015, 08:38 PM
Mota has a notable 'Eurasian' (Proto-Eurasian?) pull in the PCA, but nothing in the formal stats...

http://i58.tinypic.com/2emei6o.png

If he had been included in the Hodgson et al. study he would have had a lot of the "Ethiopic" component, which was close both to the Nilo-Saharan and Ethio-Somali one (the latter said to be of Eurasian origin by the authors).

MfA
10-08-2015, 08:44 PM
They should have used a Crown Eurasian like Ust-Ishim.

Shaikorth
10-08-2015, 08:47 PM
Mota has a notable 'Eurasian' (Proto-Eurasian?) pull in the PCA, but nothing in the formal stats...

http://i58.tinypic.com/2emei6o.png

It's like Ust-Ishim, his place on a global PCA doesn't fully correspond to his ancestry. The total lack of Neanderthal and Denisovan is the thing that supports him being 100% African the most, modern Africans do have traces including those further from Eurasians on a PCA.

Lank
10-08-2015, 08:58 PM
They should have used a Crown Eurasian like Ust-Ishim.
They did use Loschbour, which also gave a significant result. From the paper, compared to the top match to the WE in Yoruba (Sardinian):

D(Yoruba, Mota; Sardinian, Han), D=0.0098 Z=5.008
D(Yoruba, Mota; Loschbour, Han), D=0.0059 Z=3.366

Megalophias
10-08-2015, 09:21 PM
Most of Africa has experienced the Bantu expansion even more recently than this, which has spread genes all over the place. East African (Cushitic?) pastoralists seem to have gone all the way to South Africa. Then there are people who have migrated back and forth across the Sahel. There are Arab migrants in all kinds of places, there is recent European colonial admixture in South Africa (some Khoisan have a lot). The Mbuti speak a Central Sudanic language that they must have picked up from East Africans, the Biaka speak Bantu (and intermarry with them). All kinds of mixing and stirring going on, and that's just some of the stuff that we know about.


I wonder if any of this has anything to do with R1b-V88, despite this particular guy not carrying that signature. I recall the recent analysis on the split between the St Maarten V88 and Sardinians was around 6500 BC. R1b-V88 has also been found in Sardinia, as well as an Iberian farmer carrying a SNP also in the same branch. The timeline fits well with the Neolithic wet Saharan period too.
This guy *didn't* have West Eurasian, remember - he shouldn't have a West Eurasian haplogroup.

The mixture date in Yoruba was put much earlier (like 9000 years ago I think) by ALDER. The authors of this study say it is just off due to the large uncertainty in the estimate, which could be true. But it is also possible that Yoruba got their West Eurasian through a different and earlier route - R1b-V88 carrying people across North Africa, while the East African would be mostly J (and some T etc) carrying people from Asia. That would make sense with the Green Sahara and R1b-V88 split dates you are talking about.

Agamemnon
10-08-2015, 09:59 PM
Glad to see we're finally getting reliable ancient data from Africa, this could have important implications from a linguistic standpoint... More is needed though.

Awale
10-09-2015, 05:32 AM
Was hoping to start a thread on this but you beat me to it... Good work, Passa. :)

King
10-09-2015, 06:23 AM
Don't worry awale you got my thanks regardless :D.

Dubhthach
10-09-2015, 08:25 AM
I see the the guys in UCD (University College Dublin) down the road are involved in this, at least some knowledge in aDNA been built up here in Ireland. It looks like they used same methodology revolving around the Petrous bone in the inner ear. Hopefully now we'll start to see a flood of ancient DNA results show up from all over the place:

Optimal Ancient DNA Yields from the Inner Ear Part of the Human Petrous Bone
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0129102

Dubhthach
10-09-2015, 08:26 AM
Glad to see we're finally getting reliable ancient data from Africa, this could have important implications from a linguistic standpoint... More is needed though.

We have an expression in Irish:
tús maith leath na hOibre
"A good start is half the work"

Now that we have methodology and actual result from Africa I imagine that we'll start seeing more and more results from region. It be great to get some results from Middle East, it might prove some slant on "Basal Eurasian" grouping that is identified as component of EEF.

Awale
10-09-2015, 12:00 PM
Note how admixture levels only decrease in some Ethiopian populations (Wolayta, Afar, Somali, Oromo, Tigray, Amhara, Jews) using Mota as a reference instead of Yoruba. In all other African populations, Eurasian admixture levels rise using Mota.

What's interesting about this is that we don't experience an increase alongside the South Sudanese samples & Anuaks (whom we tend to share the "East African / Nilo-Saharan" component with in admixture runs) and they experience practically as much of a hike as Yorubas do meaning this can't be blamed entirely on their Niger-Congo-related admixture. Does this imply that their non-Niger-Congo-related ancestors experienced Eurasian gene flow after our African ancestors split from them?

ZephyrousMandaru
10-09-2015, 01:01 PM
I don't think we could make an argument that Y-DNA originated in West Eurasia based on this alone. Perhaps if you were to some specific downstream subclade of E1b1b, but E itself? I don't think so, this back migration occurred around 4,500 years ago. Haplogroup E itself is much older. It's similar to the recent discovery of J1 in Samartians, it's possible that it could have originated in the region where the Samartians lived (In fact, I don't think we've even identified the subclades they belong to). But until we have ancient DNA evidence from other parts of the world namely the Middle East, it's still a very speculative hypothesis to claim that J1 arose in the Caucasus or that E arose outside of Africa.

ADW_1981
10-09-2015, 04:06 PM
Yet the match was closer to Stuttgart/Sardinians than it was modern Bedouins + Palestinians, which implies those latter populations have also received genetic input from somewhere else as well.

ZephyrousMandaru
10-09-2015, 04:20 PM
Stuttgart and the Sardinians without WHG admixture and Bedouins without recent Sub-Saharan African admixture would cluster together. This could also be a function of the antiquity of the genome. Similar to how Ust-Ishim doesn't cluster very closely with any modern West Eurasians.

Awale
10-09-2015, 04:22 PM
One other thing I'd note is how the West Eurasian scores in Horn Africans like Somalis, Habeshas, Agaws (Afar & Ethiopian Jews) & Oromos goes up when the Druze are used but then go down when the LBK are used. I honestly dismissed this and mostly still do because the Druze are partially "African" but from what I've seen they're only like ~3% or so "African" (and perhaps some of this ancestry even carries Eurasian ancestry with it?) so would that really shift things by much?

http://oi58.tinypic.com/2u3y6tz.jpg

ZephyrousMandaru
10-09-2015, 04:23 PM
I think we should use Mota as a proxy for Basal Eurasian, or at the very least run him through an ADMIXTURE calculator along with other ancient genomes.

Kale
10-09-2015, 04:55 PM
One other thing I'd note is how the West Eurasian scores in Horn Africans like Somalis, Habeshas, Agaws (Afar & Ethiopian Jews) & Oromos goes up when the Druze are used but then go down when the LBK are used. I honestly dismissed this and mostly still do because the Druze are partially "African" but from what I've seen they're only like ~3% or so "African" (and perhaps some of this ancestry even carries Eurasian ancestry with it?) so would that really shift things by much?

I'm not surprised it goes up in Horners, I mean, Horners likely inherited some ancestry during the Jewish/Christian/Muslim era migration from populations similar to Druze. What I am surprised about though, is how the Eurasian estimates for EVERY OTHER African population goes up by roughly the same amount, ~12%. Maybe it's playing the other way around, maybe the West Eurasian components in Druze are actually harboring some African.

ADW_1981
10-09-2015, 04:59 PM
Stuttgart and the Sardinians without WHG admixture and Bedouins without recent Sub-Saharan African admixture would cluster together. This could also be a function of the antiquity of the genome. Similar to how Ust-Ishim doesn't cluster very closely with any modern West Eurasians.

The paper that's coming out next week stated that Stuttgart could be plotted as 90% Neolithic Anatolian and 10% WHG. In other words, it sounds like only an additional 10% Eurasian Hunter Gatherer was picked up by the farmers who moved into Europe initially. It doesn't sound like the earliest European farmers mixed all that much initially, contrary to what was previously thought. It could be that Bedouin and Palestinians are shifted by a "Teal" component that is ubiquitous in the Middle East today.

ZephyrousMandaru
10-09-2015, 06:13 PM
The paper that's coming out next week stated that Stuttgart could be plotted as 90% Neolithic Anatolian and 10% WHG. In other words, it sounds like only an additional 10% Eurasian Hunter Gatherer was picked up by the farmers who moved into Europe initially. It doesn't sound like the earliest European farmers mixed all that much initially, contrary to what was previously thought. It could be that Bedouin and Palestinians are shifted by a "Teal" component that is ubiquitous in the Middle East today.

The teal component itself is predominantly Near Eastern with some ANE. The Neolithic Anatolian farmers are if I recall correctly mixtures of UHG (Middle Eastern hunters) and Basal Eurasian. I believe Sardinians and Stuttgart might be mixed more with this Middle Eastern variety of WHG, rather than European WHG.

rossa
10-09-2015, 07:06 PM
Isn't there a group in Ethiopia called the Falash that are Jewish (insert Dan Brown Ark of the Covenant stories here), could this compenent and this group have any link?

Awale
10-09-2015, 07:17 PM
I'm not surprised it goes up in Horners, I mean, Horners likely inherited some ancestry during the Jewish/Christian/Muslim era migration from populations similar to Druze.

There's really no evidence of such gene flow into these groups (at least in meaningful amounts that would ever show up in such analyses). The latest signs of West Eurasian gene flow would be in Habeshas from around 2,800 to 3,000 years ago when their ancestors shifted from Central-Cushitic (https://en.wikipedia.org/wiki/Agaw_languages) to South-Semitic and what remains of that ancestry in modern Habeshas whilst seemingly non-negligible is not substantial since most of their ancestry clearly seems to be shared with their Cushitic speaking neighbors (http://anthromadness.blogspot.ae/2015/02/cushitic-admixture-levels-somalis-as.html) whether that's the Xamir ("Afar" samples (http://anthromadness.blogspot.ae/2014/12/afar-samples-from-autosomal-dna-studies.html)) or Somalis. And this admixture in Habeshas that's not seemingly present in groups like Somalis doesn't adequately explain the hike in all the present Horn African groups when the Druze are used.


What I am surprised about though, is how the Eurasian estimates for EVERY OTHER African population goes up by roughly the same amount, ~12%. Maybe it's playing the other way around, maybe the West Eurasian components in Druze are actually harboring some African.

The West Eurasian components in the LBK too? Cos the increase generally sticks around even when they're used. But yes, I find it quite weird how even groups like Mbutis and such are turning up with a "Eurasian" element. This just generally seems... Weird.


Isn't there a group in Ethiopia called the Falash that are Jewish (insert Dan Brown Ark of the Covenant stories here), could this compenent and this group have any link?

That population is present in this study. They're the "Ethiopian Jews". "Falasha" or "Felasha" is just a rather derogatory term coined by the Abyssinian Emperor Yeshaq I (https://en.wikipedia.org/wiki/Yeshaq_I) (random side note; a hymn made on his orders marks the first documented use of the word "Somali" too; the hymn was made to celebrate him kicking this hombre (https://en.wikipedia.org/wiki/Sa%27ad_ad-Din_II)'s ass) during the medieval period to describe Ethiopian Jews / the Beta Israel (https://en.wikipedia.org/wiki/Beta_Israel) who are really just Ethio-Semitized (https://en.wikipedia.org/wiki/Ethiopian_Semitic_languages) Agaws (https://en.wikipedia.org/wiki/Agaw_people). I've gone into them at my blog in the past (http://anthromadness.blogspot.ae/2015/04/the-jewishness-of-ethiopian-jews_7.html).

gihanga.rwanda
10-09-2015, 08:19 PM
The West Eurasian components in the LBK too? Cos the increase generally sticks around even when they're used. But yes, I find it quite weird how even groups like Mbutis and such are turning up with a "Eurasian" element. This just generally seems... Weird.


The results are very weird, I agree. Not the fact that there is admixture necessarily, but the fact that it is pretty homogenous across the board (from 6% in the Mbuti to 9% among the Luhya) among groups (West Africans, Bantus, South Sudanese, Khoe-San) with disparate population histories. For example, the Xun (who again generally lack Bantu or Cushitic admixture) possess just as much purport Eurasian admixture (7.6%) as the Yoruba (7.0-7.4%) according to these stats. Check out the blog post by Maju I linked on Forumbiodiversity, I think he explains my sentiment perfectly. The odd effect it has on Horners is also kind of weird.

Honestly, these authors might have captured evidence of "Basal Eurasians" without realizing it. Treemix consistently captures an admixture edge close to the Dinka and Hadza (but not really) to the root of groups like LBK.

Agamemnon
10-09-2015, 09:09 PM
If anything, we can expect many more "weird" results of the sort in the near future. As usual, this goes on to prove that contemporary populations are unreliable proxies when addressing past demographic events.

gihanga.rwanda
10-09-2015, 10:01 PM
If anything, we can expect many more "weird" results of the sort in the near future. As usual, this goes on to prove that contemporary populations are unreliable proxies when addressing past demographic events.

Idk I don't think the author's captured the full picture. This paper reminds me of that one paper that argued for widespread Siberian/Native American admixture in Europe; months before we were introduced to ANE.

Chad Rohlfsen
10-09-2015, 11:57 PM
Something is odd here. Making inferences about other Africans, because of one African from not that long ago, is a little extreme. I'm pretty sure Africans had Eurasian, or what people call Basal Europeans and West Asians, prior to this. If this person only contained supposed Basal Eurasian, without the 50% WHG/UHG input, then they would appear further from Eurasians. This just reflects a pre-farming, WHG/UHG containing element into Africa, and just one part of Africa. The PCA does have this individual shifted into an odd spot, if it is supposed to be a pre-admixture African. It looks closer to Eurasians, in general. Very odd, indeed.

parasar
10-10-2015, 01:12 AM
Idk I don't think the author's captured the full picture. This paper reminds me of that one paper that argued for widespread Siberian/Native American admixture in Europe; months before we were introduced to ANE.

Which "ghost" turned out to be correct.

gihanga.rwanda
10-10-2015, 02:11 AM
Which "ghost" turned out to be correct.

Well not really, instead of one directional ENA gene flow into Europe they actually found evidence of an ancient, previously undetected population that contributed to both Europeans and Siberians/NAs, among others; on top of that, this population although distinct from WEHG (Western Eurasian Hunter Gatherers) was closer to them than to ENA populations.

Btw people, the purport Neanderthal signal in the Yoruba is apparently inflated; it's supposed to be at .21% but the paper reports it at .7%. So unless their "Eurasian" component is a quarter Neanderthal, the authors have some explaining to do...

Megalophias
10-10-2015, 05:36 AM
Well not really, instead of one directional ENA gene flow into Europe they actually found evidence of an ancient, previously undetected population that contributed to both Europeans and Siberians/NAs, among others; on top of that, this population although distinct from WEHG (Western Eurasian Hunter Gatherers) was closer to them than to ENA populations.

From the original paper: "We do not of course think that there has been substantial gene flow back into Europe from Amazonia. The only plausible explanation we can see for our signal of admixture into the French is that an ancient northern Eurasian population contributed genetic material both to the ancestral population of the Americas, and also to the ancestral population of northern Europe."

Shaikorth
10-10-2015, 06:24 AM
The PCA does have this individual shifted into an odd spot, if it is supposed to be a pre-admixture African. It looks closer to Eurasians, in general. Very odd, indeed.

Not really, Ust-Ishim, WHG and ANE all cluster in different places on a PCA yet they are considered unadmixed and the PCA just tries to place them within modern eurasian variation.

Megalophias
10-10-2015, 06:42 AM
Yes, he is most closely related to Ari, so he ends up projected close to Ari. The PCA is only of East Africans, and PC1 is only partly related to Eurasian ancestry. He may have some kind of older Eurasian ancestry, but it hardly seems likely that he would have *more* of it than other East Africans.

Kurd
10-10-2015, 08:32 AM
The PCAs seem unremarkeable to me folks. Within the Omotic variation on one, and the Ari variation, with slightly less of a Middle Eastern shift than the average Ari, on the other PCA posted.

Is there a link to the page with the stats?

gihanga.rwanda
10-10-2015, 09:42 AM
From the original paper: "We do not of course think that there has been substantial gene flow back into Europe from Amazonia. The only plausible explanation we can see for our signal of admixture into the French is that an ancient northern Eurasian population contributed genetic material both to the ancestral population of the Americas, and also to the ancestral population of northern Europe."

"Ancient northern Eurasian" as in Siberian; I don't think they had any working concept of ANE then, which of course was any fault of their own.

gihanga.rwanda
10-10-2015, 09:46 AM
Not really, Ust-Ishim, WHG and ANE all cluster in different places on a PCA yet they are considered unadmixed and the PCA just tries to place them within modern eurasian variation.

This is genome isn't really ancient, only 4,500 years old; it won't act as erratically as a 45,000 or 24,000 year old genome. It's pretty clear from the PCAs that Mota is situated in close proximity to Ari and Sandawe; but regardless, eurogenes or another paper should clear this up.

gihanga.rwanda
10-10-2015, 09:52 AM
The PCAs seem unremarkeable to me folks. Within the Omotic variation on one, and the Ari variation, with slightly less of a Middle Eastern shift than the average Ari, on the other PCA posted.

Is there a link to the page with the stats?

The Ari are supposed to be ~15-20% Eurasian according to these stats (same is true for the Sandawe), so hopefully another PCA will clear things up. Btw, there are two types of Ari, the blacksmiths appear more "pure" while the cultivators exhibit more Cushitic admixture.

Lank
10-10-2015, 10:14 AM
The PCAs seem unremarkeable to me folks. Within the Omotic variation on one, and the Ari variation, with slightly less of a Middle Eastern shift than the average Ari, on the other PCA posted.
I think it's rather notable that Mota, despite having a lower signal of Eurasian admixture than any modern Africans, clusters with the Ari and most admixed Maasai samples on the most significant dimension (PC1) in the global PCA. This is not necessarily surprising; South Sudanese are a bit closer to Eurasians than West Africans are when overall genetic distance is measured, despite having a slightly lower signal of Eurasian admixture than West Africans. Possibly indicating higher levels of proto-Eurasian ancestry in the South Sudanese (unsurprising considering geography and the necessary origin of OOA migrants in the eastern part of the continent).

The Mota result may indicate even higher levels of proto-Eurasian ancestry, despite the fact that he appears less Eurasian-admixed than all modern Africans. Him carrying Y-DNA E, which is descended from proto-Eurasian CT, and mtDNA L3 would fit nicely with high levels of proto-Eurasian ancestry. Recently, I argued for (http://www.anthrogenica.com/showthread.php?4916-Birthplace-of-Y-DNA-DE-and-E-Africa-or-Eurasia) an African origin of Y-DNA E/DE. A previous study indicated that modern Omotics from Ethiopia have the highest levels of mtDNA L3(xM,N) in the world, and the most isolated/remote Omotic groups show Y-DNA E frequencies approaching 100%.

Awale
10-10-2015, 10:23 AM
The results are very weird, I agree. Not the fact that there is admixture necessarily, but the fact that it is pretty homogenous across the board (from 6% in the Mbuti to 9% among the Luhya) among groups (West Africans, Bantus, South Sudanese, Khoe-San) with disparate population histories.

Yep, I was so weirded out by this that I made another blog post (http://anthromadness.blogspot.ae/2015/10/all-modern-africans-are-part-eurasian.html) expressing skepticism about these results and cited those weiredly homogeneous levels as a reason for my skepticism. I would've been more convinced if the levels varied more (~3% in one group, ~7% in another etc.).


If anything, we can expect many more "weird" results of the sort in the near future. As usual, this goes on to prove that contemporary populations are unreliable proxies when addressing past demographic events.

Quoted for truth...

Kurd
10-10-2015, 11:08 AM
I think it's rather notable that Mota, despite having a lower signal of Eurasian admixture than any modern Africans, clusters with the Ari and most admixed Maasai samples on the most significant dimension (PC1) in the global PCA.

That is sort of an oxymoron. PCAs are a reflection of admixture. Not really notable. Do you have the eigenvalues and eigenvectors for the PCA? FYI, not all admixture component variations can be captured by just 2 principal components.

Anyways, is there a link to the analysis? I would need to see that to be able to form any conclusions.

Passa
10-10-2015, 12:25 PM
The genome is available here: ftp://biodisk.org/Store/Genome/African/

Can anybody here analyse it?

Sangarius
10-10-2015, 12:34 PM
The genome is available here: ftp://biodisk.org/Store/Genome/African/

Can anybody here analyse it?

You should inform Generalissimo.

parasar
10-10-2015, 02:09 PM
I think we should use Mota as a proxy for Basal Eurasian, or at the very least run him through an ADMIXTURE calculator along with other ancient genomes.
Mota has nothing to do with Basal Eurasian. Mota is the same position vis a vis Basal Eurasian that he is with respect to ENA-WHG-ANE. There is perhaps a 2000 generation gap between OoA and the Basal split, so I think Mota would be that many generations drifted away from Basal. One feature Mota would share with Basal is a lack of Denisovan (and perhaps Neandertal) since both SSA and Basal did not make a sojourn to E/SA Asia.

We already knew for a some time based on modern populations that the affinity between Basal and SSA is due to a Sardinian like input to Africa, which the Mota paper confirms with ancient DNA.

The so called basal is showing a little more affinity to African since southern Europe had some continued contact with Africa.
Please see Reich group's San paper referenced earlier in this thread. Post 10.

http://www.pnas.org/content/111/7/2632.full
https://www.newscientist.com/article/dn24988-humanitys-forgotten-return-to-africa-revealed-in-dna#.VG4IwDTF8xE

Back to Africa – but from where?
Reich and his colleagues found that DNA sequences in the Khoisan people most closely resemble some found in people who today live in southern Europe. That, however, does not mean the migration back to Africa started in Italy or Spain. More likely, the migration began in what is now the Middle East.

We know that southern Europeans can trace their ancestry to the Middle East. However, in the thousands of years since they – and the ancestors of the Khoisan – left the region, it has experienced several waves of immigration. These waves have had a significant effect on the genes of people living in the Middle East today, and means southern Europeans are much closer to the original inhabitants of the Levant than modern-day Middle Easterners.

ep2
10-10-2015, 02:09 PM
The lack of Eurasian DNA before 4500 years ago in the region also have been replicated in Sudan uniparental studies (as well as Egyptian autosomal aDNA studies-BMJ, JAMA, old Paabo to some extent).

http://i1274.photobucket.com/albums/y421/amunratheultimate2/Misc2/Ancient%20DNA%20of%20Kushites%20from%20Genetic%20P atterns%20Hassan%20dissertation_zpsrxznegab.jpg
From Hassan(2009) (http://tinyurl.com/pp8s25y)

The Eurasian haplogroups start to appear during the christian era. Kadruka is a neolithic site in Sudan near the Egyptian border.

xKeleix
10-10-2015, 03:04 PM
In all honesty, these ancient genomes keep creating more and more questions. Some questions for those more knowledgeable about the finer details, my comments following:

1. What does this say about haplogroup E?
Clearly it doesn't support the theory that Y DNA Haplogroup E back-migrated if, in fact, Mota is absent of Eurasian admixture. Of course, it doesn't rule out the possibility entirely, but it seems to go against that theory.

2. What does this say about basal Eurasian?
If migration edges are formed using Nilotic groups into Sardinians, shouldn't this then say something about basal Eurasian? Basal Eurasian is still a mystery.

3. Why is there a seemingly Eurasian "pull" in the PCA relative to West Africans if Mota is not admixed?
This seems to have been answered: Mainly, it could be because the 2 dimensions don't show all of the diversity. I suppose it could also have to do with Eastern Omotics and similar populations being more similar to Eurasians in other ways - namely, in that they're more similar to the OOA peoples rather than it being from admixture.

Lank
10-10-2015, 04:43 PM
I wonder how we can reconcile these results with some of the evidence of more ancient Eurasian admixture. Some of the most obvious examples include Y-DNA J(xJ1,J2), which has been found in remote Omotics but never in Cushitic/Semitic Ethiopians, and mtDNA U6a2, a subclade that is centered in Ethiopia and has been dated to 20,000 years ago. And there's the more well-known example of mtDNA M1. It seems most likely they were just living in another part of Africa at the time Mota man was a hunter-gatherer in southwestern Ethiopia.

Modern data should also be studied while we wait for more aDNA. Based on uniparental lineages, there are a number of Omotic tribes in remote southwestern Ethiopia that have probably preserved native ancestry to a greater degree than the Ari.

Kale
10-10-2015, 04:54 PM
In all honesty, these ancient genomes keep creating more and more questions. Some questions for those more knowledgeable about the finer details, my comments following:

1. What does this say about haplogroup E?
Clearly it doesn't support the theory that Y DNA Haplogroup E back-migrated if, in fact, Mota is absent of Eurasian admixture. Of course, it doesn't rule out the possibility entirely, but it seems to go against that theory.

2. What does this say about basal Eurasian?
If migration edges are formed using Nilotic groups into Sardinians, shouldn't this then say something about basal Eurasian? Basal Eurasian is still a mystery.

3. Why is there a seemingly Eurasian "pull" in the PCA relative to West Africans if Mota is not admixed?
This seems to have been answered: Mainly, it could be because the 2 dimensions don't show all of the diversity. I suppose it could also have to do with Eastern Omotics and similar populations being more similar to Eurasians in other ways - namely, in that they're more similar to the OOA peoples rather than it being from admixture.

1) It doesn't suggest that E is of West (or even crown Eurasian) origin. Basal Eurasian is a very likely possibility still.
2) If an African group branched off the human tree 80kya...and Basal Eurasians split off 75kya...Basal Eurasians will be effectively treated as a brother to that 80kya diverged African group because they diverged from everyone else around the same time.
3) In a global PCA I'd expect someone like Mota...with no East or West Eurasian input, but pretty close to the OOA bottleneck (maybe?) to plot equally closer to East and West Eurasians relative to more diverged groups like Mbuti or San (unless of course Mota has basal Eurasian, which would drag it closer to the West Eurasian pole.)

ep2
10-10-2015, 06:34 PM
To gain a better view of the whole region. We also have this full genome study dating the Eurasian admixture in modern Egyptians to 750 years ago.


Using ADMIXTURE and principal-component analysis (PCA) (Figure 1A), we estimated the average proportion of non-African ancestry in the Egyptians to be 80% and dated the midpoint of the admixture event by using ALDER20 to around 750 years ago (Table S2), consistent with the Islamic expansion and dates reported previously. link (http://dx.doi.org/10.1016/j.ajhg.2015.04.019)

Tomenable
10-10-2015, 10:01 PM
affecting even populations such as Yoruba (...) previously thought to be relatively unadmixed, who harbor 6-7% Eurasian ancestry.
Razib Khan noticed ENF ("Sardinian-like") admixture in Yoruba back in 2012!:

http://blogs.discovermagazine.com/gnxp/2012/03/we-are-all-sardinians/#.VhmJzUoyU8E

Lank
10-14-2015, 04:45 PM
From Nature, Ancient DNA from hot climes yields its secrets (http://www.nature.com/news/ancient-dna-from-hot-climes-yields-its-secrets-1.18549?WT.mc_id=TWT_NatureNews):



“Africa is the next step in terms of the population history of humans. It is an obvious place to go and look,” says Eske Willerslev, an evolutionary geneticist at the Natural History Museum of Denmark. He says that his team has also had success in obtaining DNA from African remains, including some that have not been protected by the relatively cool confines of a high-altitude cave. Willerslev expects that researchers will eventually strike lucky in extracting DNA from African remains that are many tens of thousands of years old, rather than a few thousand.

Seems like we can expect to get more ancient DNA from Africa in the near future.

Lank
10-26-2015, 05:03 PM
His Y-DNA was probably E-M329, a 30-40 ky old lineage indigenous to the Horn of Africa.
Mota is now confirmed as E-M329 (E1b1a2) by YFull (http://www.yfull.com/tree/E1b1a2/), i.e. the same lineage that peaks in modern Omotics from southern Ethiopia and shares a distant ancestor with West-Central African E-M2 ~40,000 years ago. Not surprising considering the high levels of drift shared with the Ari.

Lank
11-05-2015, 11:22 PM
Mota's GEDmatch ID: M261275

Eurogenes K13

Population
North_Atlantic -
Baltic -
West_Med -
West_Asian -
East_Med -
Red_Sea -
South_Asian 0.63%
East_Asian -
Siberian 0.20%
Amerindian 1.85%
Oceanian 1.01%
Northeast_African 79.99%
Sub-Saharan 16.33%


Eurogenes K36

Population
Amerindian 0.27%
Arabian -
Armenian -
Basque -
Central_African 13.56%
Central_Euro -
East_African -
East_Asian 0.52%
East_Balkan -
East_Central_Asian -
East_Central_Euro -
East_Med -
Eastern_Euro -
Fennoscandian -
French -
Iberian -
Indo-Chinese -
Italian -
Malayan -
Near_Eastern -
North_African -
North_Atlantic -
North_Caucasian -
North_Sea -
Northeast_African -
Oceanian 0.15%
Omotic 68.15%
Pygmy 13.88%
Siberian -
South_Asian -
South_Central_Asian -
South_Chinese 0.08%
Volga-Ural -
West_African 3.38%
West_Caucasian -
West_Med -


HarappaWorld

Population
S-Indian 1.39%
Baloch 1.44%
Caucasian -
NE-Euro 0.55%
SE-Asian -
Siberian 0.62%
NE-Asian 0.11%
Papuan 0.93%
American 0.99%
Beringian -
Mediterranean -
SW-Asian 1.54%
San 17.75%
E-African 53.47%
Pygmy 11.00%
W-African 10.22%


MDLP K13 Ultimate

Amerindian 0.41%
ANE 4.17%
Arctic -
ASI 1.10%
Caucas-Gedrosia -
EastAsian 0.76%
ENF -
NearEast 7.89%
Oceanian 0.88%
Paleo-African 21.73%
Siberian 1.26%
Subsaharian 61.80%
WHG-UHG -

DMXX
11-06-2015, 07:31 AM
Eurogenes K7 ANE results.



Kit Number: M261275 Elapsed Time: 18.81 seconds


Population
ANE 2.70%
ASE 1.80%
WHG-UHG -
East_Eurasian 2.33%
West_African 20.30%
East_African 65.23%
ENF 7.65%

Arame
11-06-2015, 08:55 AM
If Mota is not an exception then:

1. The Ethopian urheimat of Afro-Asiatic languages is seriously weakened, not to say dead. The origins of AA should be searched in North.
2. Mota is E1b1a while in modern Ethiopia this clade has small frequency. Only 3.4%. It is also low or inexistant in other AA people. On the other hand the E1b1b-M35 show quite well correlation with many Afro-Asiatic people ( exception Semitic J1-P58 and Chadic R1b-V88)
3. The origin of E1b1b-M215>M35 from Ethiopia is also under serious doubt.
4. Mota clustering with Hadza makes more sense than with Omotic people Ari. Hadza are hunter-gatherers like Mota, they speak a language isolate.

And the last thing. I think the chances that the Natufian culture will have E1b1b increases. I would bet on E-M78 the ancestral form of Balkanic E-V13 and Egyptian E-V22, E-V12, and Cushitic E-V12>V32.

bones
11-06-2015, 12:25 PM
Eurogenes K7 ANE results.



Kit Number: M261275 Elapsed Time: 18.81 seconds


Population
ANE 2.70%
ASE 1.80%
WHG-UHG -
East_Eurasian 2.33%
West_African 20.30%
East_African 65.23%
ENF 7.65%


According to Polakos Blogpost on his eurogenes blog, Mota has no west eurasian admixture, but according to the numbers you presented here, there is 7.65% ENF, which must qualify as west eurasian admixture in Mota? Am i missing something here?

DMXX
11-06-2015, 01:06 PM
According to Polakos Blogpost on his eurogenes blog, Mota has no west eurasian admixture, but according to the numbers you presented here, there is 7.65% ENF, which must qualify as west eurasian admixture in Mota? Am i missing something here?

Yes, I noticed that as well. You're not the only one perplexed by the finding.

[Edit]: Kurd's Eurasia K14 for Mota:



Population
N_Amerindian 0.39%
Afansievo_Yamnaya -
Kalash -
Siberian -
S_Amerindian 0.43%
Sub_Saharan 49.11%
SE_Asian 1.16%
E_African 43.96%
SW_Asian -
Neolithic_Balkan_Farmers 0.75%
SHG_WHG 0.42%
Early_European_Farmers -
S_Indian 2.59%
Papuan 1.20%
Oracle


>95% Sub_Saharan and E_African combined, all else looks like noise (bar S_Indian, SE_Asian and Papuan?).

Only workaround I can surmise between these numbers is that Mota actually doesn't have W. Eurasian admixture and the Eurogenes ANE K7 result is just manifesting some Basal Eurasian through the ENF component.

Morci
11-06-2015, 01:14 PM
It seems all of the calculators agree that Mota has some sort of Eastern Eurasian ancestry related to Amerindians and Siberians, too. Yes? Every Horn African gets this small amount of East Asian affinity too, in K7. None of them have ANE, but Mota does. Interesting.

bones
11-06-2015, 01:33 PM
Yes, I noticed that as well. You're not the only one perplexed by the finding.

[Edit]: Kurd's Eurasia K14 for Mota:



>95% Sub_Saharan and E_African combined, all else looks like noise (bar S_Indian, SE_Asian and Papuan?).

Only workaround I can surmise between these numbers is that Mota actually doesn't have W. Eurasian admixture and the Eurogenes ANE K7 result is just manifesting some Basal Eurasian through the ENF component.

Yeah i thought about that as well, that the ENF in Mota showed by the K7 result, is actually Basal Eurasian. It makes sense since we know that basal eurasian is not an african component but at the same time is very divergent from other eurasian components. So if mota has admixture from from basal eurasians then it would not show up in these k36 runs as west eurasian because of how divergent that component is. If Mota has basal eurasian then it is probably from his E y-dna haplogroup.

R.Rocca
11-06-2015, 01:52 PM
If Mota is not an exception then:

1. The Ethopian urheimat of Afro-Asiatic languages is seriously weakened, not to say dead. The origins of AA should be searched in North.
2. Mota is E1b1a while in modern Ethiopia this clade has small frequency. Only 3.4%. It is also low or inexistant in other AA people. On the other hand the E1b1b-M35 show quite well correlation with many Afro-Asiatic people ( exception Semitic J1-P58 and Chadic R1b-V88)
3. The origin of E1b1b-M215>M35 from Ethiopia is also under serious doubt.
4. Mota clustering with Hadza makes more sense than with Omotic people Ari. Hadza are hunter-gatherers like Mota, they speak a language isolate.

And the last thing. I think the chances that the Natufian culture will have E1b1b increases. I would bet on E-M78 the ancestral form of Balkanic E-V13 and Egyptian E-V22, E-V12, and Cushitic E-V12>V32.

Arame, I don't understand the Afro-Asiatic origin theories all that well, but have an interest due to my maternal grandfather's J1. Can you elaborate more on your first point? If we found a similarly aged Arabian desert sample to be 75% ENF but 25% NE African/Ethiopian, wouldn't it leave all options open again relating to Afro-Asiatic? Again, not arguing anything one way or another, just trying to understand, so be gentle :D

Kale
11-06-2015, 04:23 PM
Yeah i thought about that as well, that the ENF in Mota showed by the K7 result, is actually Basal Eurasian. It makes sense since we know that basal eurasian is not an african component but at the same time is very divergent from other eurasian components. So if mota has admixture from from basal eurasians then it would not show up in these k36 runs as west eurasian because of how divergent that component is. If Mota has basal eurasian then it is probably from his E y-dna haplogroup.

It could be since the levels of East Eurasian and West Eurasian in Mota are roughly equal, that it is just a biproduct of him being close to the OOA population. Sharing a bit more drift with both East and West Eurasian than other Africans.

Lank
11-06-2015, 04:39 PM
Eurogenes has posted some interesting TreeMix graphs (http://eurogenes.blogspot.com/2015/11/treemix-graphs-with-ancient-ethiopian.html). Mota has a position closer to Eurasians than other Africans, in accordance with the PCA. In addition, he shows significant amounts of something that looks rather archaic. These are highly preliminary results, but if confirmed they would suggest that this is what really explains most of the "Eurasian" signal reported in African hunter-gatherers and (non-East African) farmers.



3. The origin of E1b1b-M215>M35 from Ethiopia is also under serious doubt.
Trombetta's 2015 paper has already revealed that East African M35 clades are relatively young (dated to 8-12 kya), although Omotic M35 lines have not been properly tested. The deeper branches are most diverse in North Africa, with some spillover into southern Europe, making that the most likely point of origin for now. Considering the age of E-M35 (25 kya), it could not have originated with the Natufians. Also, the presence of M215(xM35) in Ethiopia, peaking in Omotics, supports that it evolved within the continent, but most likely in North Africa.


According to Polakos Blogpost on his eurogenes blog, Mota has no west eurasian admixture, but according to the numbers you presented here, there is 7.65% ENF, which must qualify as west eurasian admixture in Mota? Am i missing something here?
Mota only scores West Eurasian in runs with no East African component. The Eurogenes ANE K7 run sticks out, but there even West Africans are predominantly "East African", so that explains it.

aarnisotka
11-06-2015, 05:16 PM
The table of Eurasian admixture estimates using Yoruba and Mota as the reference. Note how admixture levels only decrease in some Ethiopian populations (Wolayta, Afar, Somali, Oromo, Tigray, Amhara, Jews) using Mota as a reference instead of Yoruba. In all other African populations, Eurasian admixture levels rise using Mota.

http://i60.tinypic.com/2elcj7n.png


I think ancient DNA from other parts of Africa is needed to really make sense of these results, although it's great that we finally have our first sample. Exciting times ahead!

I think what could be going on in there is that since Eurasians are an offshoot of ancient East African diversity what we actually see there is not actual OOA Eurasian stuff, but some old East African. East Africa and West are equally related but as Eurasians descent from East Africans, Eurasians share a closer genetic relationship with eastern SSA diversity.

bored
11-06-2015, 06:03 PM
It could be that there is some Basal Eurasian in Africa and that's what is being called West Eurasian here. Though the homogenous levels at which it supposedly exists throughout Africa is suspect.

Agamemnon
11-06-2015, 06:14 PM
If Mota is not an exception then:

1. The Ethopian urheimat of Afro-Asiatic languages is seriously weakened, not to say dead. The origins of AA should be searched in North.
2. Mota is E1b1a while in modern Ethiopia this clade has small frequency. Only 3.4%. It is also low or inexistant in other AA people. On the other hand the E1b1b-M35 show quite well correlation with many Afro-Asiatic people ( exception Semitic J1-P58 and Chadic R1b-V88)
3. The origin of E1b1b-M215>M35 from Ethiopia is also under serious doubt.
4. Mota clustering with Hadza makes more sense than with Omotic people Ari. Hadza are hunter-gatherers like Mota, they speak a language isolate.

And the last thing. I think the chances that the Natufian culture will have E1b1b increases. I would bet on E-M78 the ancestral form of Balkanic E-V13 and Egyptian E-V22, E-V12, and Cushitic E-V12>V32.

Mota certainly seems archaic, nevertheless I don't think it's enough to inform us about the validity of Ethiopia as the PAA homeland... While I decided to focus on Hebrew and NW Semitic studies this year, I've read a lot about the PAA debate and what I can say for now is that the odds for a Ethiopian urheimat aren't as good as they initially looked. Similarly, I doubt the Natufian horizon provides a valuable alternative at this point, while a Near Eastern urheimat isn't totally out of the question, the Natufian horizon is too young IMO (I'd like to say more about this but a dear friend of mine will be the first to know about this). Though it's doubtful that Mota spoke an early AA dialect, from a genetic standpoint I wouldn't be so quick to dismiss the presence of similar admixture in early AA speakers, Mota's age in itself isn't enough to provide groundbreaking information.

As far as uniparental lineages go, I think it's pretty safe to say that E-M35.1 as a whole was present in the PAA-speaking community, I'll also be surprised if J1 wasn't involved at an early stage (not necessarily PAA stage though). That being said, we should not expect a 1:1 correlation between a given lineage and AA phylogeny, and that's not in the least controversial since no one would expect that kind of fantastic correlation between R1b/R1a and IE phylogeny either.

Arbogan
11-06-2015, 08:51 PM
who lived approximately 4,500 years ago. We use this genome to demonstrate that the Eurasian backflow into Africa came from a population closely related to Early Neolithic farmers, who had colonized Europe 4,000 years earlier

This means that there must have been multiple branches of neolithics in west-asia, which makes the simplistic "barcin EEFs as the only West-Asian Neolithic population" model , obsolete. There probably existed multiple branches that co-existed at the same time, with barcin-like neolithics being their own branch entirely. That makes more sense as an inference, than
some of the ridicules assertations I've seen, that suggests that ancestors of modern west-asians immigrated en masse from central-asia or somewhere more obscure, and in the process, replacing or subduing the venerable "EEFs". Which sounds much like the historical fiction penned by some gentlemen who ascribe to ridicules ideologies.

Padre Organtino
11-06-2015, 09:48 PM
There probably existed multiple branches that co-existed at the same time, with barcin-like neolithics being their own branch entirely. That makes more sense as an inference, than
some of the ridicules assertations I've seen, that suggests that ancestors of modern west-asians immigrated en masse from central-asia or somewhere more obscure, and in the process, replacing or subduing the venerable "EFFs". Which sounds much like the historical fiction penned by some gentlemen who ascribe to ridicules ideologies.

1. EEFs not EFFs

2.Yes, cause its impossible to imagine who genetic groups migrating away from their homelands and getting replaced by totally different people. I mean this is surely not the case with Amerindians,EHGs and porto-celts. Whops, actually it is.

3.0% - this is how much ANE was found is a large number of Neolithic specimens sampled in Europe and Minor Asia. I guess "Gedrosian" like neolithics were hiding somewhere away from geneome scientists:P

4."Somewhere more obscure" is just around the mountains of Turkmenistan where Neolithic cultures have flourished for ages. And that's right next to Iran.

Need I continue?

Arbogan
11-06-2015, 10:31 PM
1. EEFs not EFFs?
A typo. Your point?



2.Yes, cause its impossible to imagine who genetic groups migrating away from their homelands and getting replaced by totally different people. I mean this is surely not the case with Amerindians,EHGs and porto-celts. Whops, actually it is.?
I didn't leave out the possibility. But I highly doubt that the neolithics represented by the barcin samples, were the sole or predominant group of neolithics in west-asia. Rather a subgroup, as this recent finding seems to suggest. And why is this implausible? Compared to the ridicules "large-scale replacement" hypothesis. The natufian and the early Neolithic cultures are old enough to lead into a scenario of differentiation and branching, it's fully possible over a period of 6000 years or more. It would explain the lack of proper diffusion of "EEF" type barcin ancestry amongst most west-asian groups.



3.0% - this is how much ANE was found is a large number of Neolithic specimens sampled in Europe and Minor Asia. I guess "Gedrosian" like neolithics were hiding somewhere away from geneome scientists:P
To date there is no present samples or any research detailing west-Asian Neolithic cultures outside of the the barcin samples, which are located in an area that is very close to Europe. And could look the way they do for many different reasons. The grand picture is still largely unexplored. We're nowhere near the full picture. We don't know when ANE(except that it must have arrived in subsequent waves) arrived in west-asia, we don't know why the barcin samples don't have any ANE or why modern day west-Asians don't have a lot of EEF-type ancestry. We don't even know what neolithics groups outside of a small corner in turkey genetically looked like. Compared to the studies on Mesolithic European hunter gatherers, the Neolithics from west-asia are largely understudied. My point with the above post was to not make hasty inferences without further study, as some people seem to do. Lets wait for some hard evidence. This study, is another piece of the puzzle. People seem to be stuck on hasty notions, because of bronze age Armenians and the barcin samples. Can you say for certain about several areas in, Levant Mesopotamia, Zagros, that the genetic picture was similar? It's still hard to ascertain the answers for many questions surrounding this issue without any evidence.




4."Somewhere more obscure" is just around the mountains of Turkmenistan where Neolithic cultures have flourished for ages. And that's right next to Iran.
Except most Neolithic sites in the Zagros-Mesopotamia-Levant areas are much older than the Jeiton culture, and there are several cultures dated to exist parallel to it in west-asia. The earliest form of systematic plant domestication occurred in the Levantine corridor, in modern day Syria and Lebanon. I'd be careful of accepting "Gedrosia" as some form of real also means nothing. It's a statistical concept, and not hard evidence. The picture is probably much more complex than matrices in admixture. We still don't know what the relationship between central-Asian Neolithic cultures and west-asia is.



Need I continue?

I honestly don't know what the point of this post was. Except as a failed attempt of being rude and correct a simple typo on my end. You're sorely mistaken if you think the point with my posts, is to engage in petty "upmanship", like some other boards. This is critical assesment.

Padre Organtino
11-06-2015, 11:20 PM
I didn't leave out the possibility. But I highly doubt that the neolithics represented by the barcin samples, were the sole or predominant group of neolithics in west-asia. Rather a subgroup, as this recent finding seems to suggest. And why is this implausible? Compared to the ridicules "large-scale replacement" hypothesis. The natufian and the early Neolithic cultures are old enough to lead into a scenario of differentiation and branching, it's fully possible over a period of 6000 years or more. It would explain the lack of proper diffusion of "EEF" type barcin ancestry amongst most west-asian groups.

Everything is "possible" including German Dziebel's "bright" Out of America theory. However, what we observe in Neolithic samples from all over Europe points to the contrary. In fact even ancient Balkan samples show similar EEF/EEF-WHG hybrid types and not modern West Asian shifted pops.
Could there have been ancient differentiation of the Neolithic groups? Yes, indeed it is a most likely scenario. However, those guys who went East have not contributed anything to European gene pool up until the recent historical migrations to SE Europe from Western Asia.



To date there is no present samples or any research detailing west-Asian Neolithic cultures outside of the the barcin samples, which are located in an area that is very close to Europe. And could look the way they do for many different reasons. The grand picture is still largely unexplored. We're nowhere near the full picture. We don't know when ANE(except that it must have arrived in subsequent waves) arrived in west-asia, we don't know why the barcin samples don't have any ANE or why modern day west-Asians don't have a lot of EEF-type ancestry. We don't even know what neolithics groups outside of a small corner in turkey genetically looked like. Compared to the studies on Mesolithic European hunter gatherers, the Neolithics from west-asia are largely understudied. My point with the above post was to not make hasty inferences without further study, as some people seem to do. Lets wait for some hard evidence. This study, is another piece of the puzzle. People seem to be stuck on hasty notions, because of bronze age Armenians and the barcin samples. Can you say for certain about several areas in, Levant Mesopotamia, Zagros, that the genetic picture was similar? It's still hard to ascertain the answers for many questions surrounding this issue without any evidence.

Fine point. Except we can quite certainly say that ANE entry happened somewhere between Neolithic and Bronze Age. Cause Armenian samples clearly show that ANE arrived Anatolia by Bronze Age.




Except most Neolithic sites in the Zagros-Mesopotamia-Levant areas are much older than the Jeiton culture, and there are several cultures dated to exist parallel to it in west-asia. The earliest form of systematic plant domestication occurred in the Levantine corridor, in modern day Syria and Lebanon. I'd be careful of accepting "Gedrosia" as some form of real also means nothing. It's a statistical concept, and not hard evidence. The picture is probably much more complex than matrices in admixture. We still don't know what the relationship between central-Asian Neolithic cultures and west-asia is.


Of course it is older. ANE is a relatively recent newcomer to West Asia despite what some would like to believe. Jeiton is interesting not because it is very old but because it has interacted with some forager types of Central Asia who could have contributed to the gene pool of Neolithic settlers. it is also interesting because it has collapsed due to the growth of aridity and people were forced to migrate from it. And it is also very interesting since we know that people like Kassites (yep, your potential ancestors), Hurrians and etc have suddenly sprung up in the Northern Middle East/Zagros and that has coincided with the spread of Kura-Araxes culture. And (behold the magic!) it happened exactly at the end of Neolithic and the start of the Bronze Age.



I honestly don't know what the point of this post was. Except as a failed attempt of being rude and correct a simple typo on my end. You're sorely mistaken if you think the point with my posts, is to engage in petty "upmanship", like some other boards. This is critical assesment.

K, thx for rather level-headed response. I did overdo it with sarcasm but mostly cause I think you are arguing against rather obvious things (at least when it comes to European Neolithic).

Arame
11-07-2015, 09:33 AM
Rocca


If we found a similarly aged Arabian desert sample to be 75% ENF but 25% NE African/Ethiopian, wouldn't it leave all options open again relating to Afro-Asiatic?

One can always proposes different theories from what autosomal side came a proto-language. I am just choosing the most probable. And it is not a question of autosomes only.
This is the map of E1b1a. The Hg of Mota but also of Ramesses III.

https://upload.wikimedia.org/wikipedia/commons/9/96/Distribution_of_haplogroup_e1b1a_in_Rosa_2007.jpg

And this is the approximative map of E1b1b

http://thegeneticatlas.com/E1b1b.png

That's why I have hard to time imagine that Mota-like people spreaded the PAA. But of course if we find some E1b1b-M35 with same autosomes as Mota then story will change certainly.


Lank


Trombetta's 2015 paper has already revealed that East African M35 clades are relatively young (dated to 8-12 kya), although Omotic M35 lines have not been properly tested. The deeper branches are most diverse in North Africa, with some spillover into southern Europe, making that the most likely point of origin for now.

I know that study of Trombetta. Let's wait do the aDNA confirms it's conclusions or not. After all I am just speculating on one sample :)

For the Natufian I specified the subclade E-M78. http://yfull.com/tree/E-M78/
It's TMRCA fits quite well into Natufian period. And it's daughter subclades distribution also. And I am not linking the Natufian with the PAA start.

Lank
11-07-2015, 10:08 AM
Eurogenes has posted some more (http://eurogenes.blogspot.com/2015/11/treemix-graphs-with-ancient-ethiopian.html) TreeMix graphs, this time including Neanderthals and Denisovans. Now, the divergent migration edge into Mota looks like it's coming from the root of the AMH lineage (just before the split of the Ju'hoan Khoisan). Mota cave is not far from where the Omo remains, the oldest known AMH fossils, have been found.

Arame, the map you posted is for E-M2 (E1b1a1) associated with Niger-Congo speakers. Mota carries E-M329 (E1b1a2), which is found in modern Ethiopians and not elsewhere, reaching 25% in the Maale Omotics from southern Ethiopia. The shared ancestor between Ethiopian E-M329 with West-Central African E-M2 dates to ~40,000 years ago.

Arame
11-07-2015, 01:43 PM
Arame, the map you posted is for E-M2 (E1b1a1) associated with Niger-Congo speakers. Mota carries E-M329 (E1b1a2), which is found in modern Ethiopians and not elsewhere,

Thanks. I didn't knew Mota's SNP under the E1b1a.

Well it doesn't change much for my arguments related to PAA and E-M215>M35. Do it?
It just shows that Mota Hg didn't have much expansion and continued to live in Ethiopa till today.

Inigo Montoya
11-11-2015, 11:01 AM
Eurogenes has posted some more (http://eurogenes.blogspot.com/2015/11/treemix-graphs-with-ancient-ethiopian.html) TreeMix graphs, this time including Neanderthals and Denisovans. Now, the divergent migration edge into Mota looks like it's coming from the root of the AMH lineage (just before the split of the Ju'hoan Khoisan). Mota cave is not far from where the Omo remains, the oldest known AMH fossils, have been found.

Yep. I asked nicely and got this 6616 6617.
Now I wish he had kept Karelia_HG, but beggars can't be choosers, and it's interesting nonetheless.
Notice how, in the 10 edges tree, the two pygmy population no longer receive their archaic admixture from the same source. The edge from basal hominid to Yoruba is almost certainly spurious, and I think the tree tells us why: the Yoruba are heavily admixed with a West pygmy population, and hence received a small unknown hominid component indirectly.
As lank mentions, Mota receives considerable admixture (30%) from basal h. s. s. As he is otherwise the sister group to OOA, I suggest he may represent a back-migration after the Nile valley became passable again during the glacial termination, which assimilated the last population of ancestral humans.
As before (and quite puzzlingly), Ju Hoan receives Eurasian (West Eurasian in this case) admixture that Yoruba doesn't. If you read the whole thread on the blog, you'll see that there quite a few cases where Treemix arrives at the exact opposite conclusion from the authors of the paper.

Vadim Verenich
11-11-2015, 07:21 PM
PLINK v1.90p 64-bit (2 Oct 2015) https://www.cog-genomics.org/plink2
(C) 2005-2015 Shaun Purcell, Christopher Chang GNU General Public License v3
Logging to GB20.log.
Options in effect:
--bfile mota_mergedset
--keep keep.txt
--out GB20
--recode 23

2068000 MB RAM detected; reserving 1034000 MB for main workspace.
Allocated 3278 MB successfully, after larger attempt(s) failed.
3588223 variants loaded from .bim file.
1 person (0 males, 0 females, 1 ambiguous) loaded from .fam.
Ambiguous sex ID written to GB20.nosex .
--keep: 1 person remaining.
Using 1 thread (no multithreaded calculations invoked.
Before main variant filters, 1 founder and 0 nonfounders present.
Calculating allele frequencies... done.
3588223 variants and 1 person pass filters and QC.
Note: No phenotypes present.
--recode 23 to GB20.txt ... done.
-----------------------------------------------------------
-- DIY Dodecad v 2.1 --------------------------------------
[email protected]@@@@@@----
-- Copyright (c) 2011 Dienekes Pontikos [email protected]@[email protected]@---
[email protected]@---
-- More information: [email protected]@@@@@@----
----- Dienekes' Anthropology Blog [email protected]@----------
-------- http://dienekes.blogspot.com [email protected]@----------
----- The Dodecad Ancestry Project [email protected]@@@@@@@@---
-------- http://dodecad.blogspot.com ----------------------
-----------------------------------------------------------

16 ancestral populations
116463 total SNPs
1 flipped SNPs
30586 heterozygous SNPs
0 no-calls
64359 absent SNPs
0.447387 genotype rate
mode genomewide

64359 SNPs missing (no-call or absent)

6690 dQ: 1.001E-07 goal: 1.000E-07

6694 total iterations
9.997E-08 final dQ

----------------------------
FINAL ADMIXTURE PROPORTIONS:
----------------------------

1.56% Amerindian
21.12% Ancestor
0.00% ANE
4.59% ANI
0.00% Arctic
1.86% Australian
0.00% Caucasian
28.51% EastAfrican
1.61% Mesolithic
1.13% NearEast
0.08% Neolithic
0.18% NorthAfrican
0.01% Oceanic
0.14% Siberian
2.31% SouthEastAsian
36.89% Subsaharian

CPU time = 153.15 sec

bones
11-11-2015, 07:55 PM
^So based on that analysis Mota doesnt seem to be entirely without eurasian admixture, or could these results be interpreted differently?

Vadim Verenich
11-11-2015, 08:25 PM
^So based on that analysis Mota doesnt seem to be entirely without eurasian admixture, or could these results be interpreted differently?

I believe it is counter-productive to base our hypothesis off one sample: both variants could apply here (I am inclined to believe the former explanation is slightly better), that's why we need either more ancient samples from Africa or some kind of circumstantial evidence.

bones
11-11-2015, 08:49 PM
I believe it is counter-productive to base our hypothesis off one sample: both variants could apply here (I am inclined to believe the former explanation is slightly better), that's why we need either more ancient samples from Africa or some kind of circumstantial evidence.

What are the two variants you referring to? Could you elaborate a bit, when you say the former explanation is slightly better do you mean that the former explanation is the one where Mota has eurasian admixture or not?

Bernard
01-25-2016, 06:56 PM
Erratum to Gallego Llorente et al. 2015 (https://dl.dropboxusercontent.com/u/26978112/Erratum%20with%20figures.pdf):

The results presented in the Report “Ancient Ethiopian genome reveals extensive Eurasian admixture throughout the African continent“ were affected by a bioinformatics error. A script necessary to convert the input produced by samtools v0.1.19 to be compatible with PLINK was not run when merging the ancient genome, Mota, with the contemporary populations SNP panel, leading to homozygote positions to the human reference genome being dropped as missing data (the analysis of admixture with Neanderthals and Denisovans was not affected). When those positions were included, 255,922 SNP out of 256,540 from the contemporary reference panel could be called in Mota. The conclusion of a large migration into East Africa from Western Eurasia, and more precisely from a source genetically close to the early Neolithic farmers, is not affected. However, the geographic extent of the genetic impact of this migration was overestimated: the Western Eurasian backflow mostly affected East Africa and only a few Sub-Saharan populations; the Yoruba and Mbuti
do not show higher levels of Western Eurasian ancestry compared to Mota.
We thank Pontus Skoglund and David Reich for letting us know about this problem
.

gihanga.rwanda
01-25-2016, 07:24 PM
Erratum to Gallego Llorente et al. 2015 (https://dl.dropboxusercontent.com/u/26978112/Erratum%20with%20figures.pdf):

Just to add on to this, it also appears that East Africans are less admixed than previously assumed. For example, Somalis are at ~30% and Habeshas ~40, a remarkable ~10% drop in purported Western Eurasian (WE) ancestry for both groups. A handful of groups (i.e. Khomani, Karretji,e, and Nama) from South Africa are the only other groups outside of East Africa with significant (no more than ~10%) WE ancestry, however, most of it can be traced to the colonial period and the assimilated of mixed-raced individuals into some of these communities.

https://dl.dropboxusercontent.com/u/...%20figures.pdf

Megalophias
01-25-2016, 07:35 PM
https://dl.dropboxusercontent.com/u/...%20figures.pdf

Dropbox links aren't working. Never mind I found it: https://dl.dropboxusercontent.com/u/26978112/Erratum%20with%20figures.pdf

The admixture actually turns negative in some samples, maybe meaning Mota himself is slightly closer to LBK than they are. I doubt it's significant though.

kingjohn
01-25-2016, 07:45 PM
in ethiopian groups west euroasian 40% is still high
do you understand it is almost half of there genome.
very high in amhara and tigray groups also in ethiopian jews.
regards
adam

Awale
01-25-2016, 08:19 PM
Just to add on to this, it also appears that East Africans are less admixed than previously assumed. For example, Somalis are at ~30% and Habeshas ~40, a remarkable ~10% drop in purported Western Eurasian (WE) ancestry for both groups. A handful of groups (i.e. Khomani, Karretji,e, and Nama) from South Africa are the only other groups outside of East Africa with significant (no more than ~10%) WE ancestry, however, most of it can be traced to the colonial period and the assimilated of mixed-raced individuals into some of these communities.

https://dl.dropboxusercontent.com/u/...%20figures.pdf

It depends... Mota himself might not be 100% African (i.e. Kurd seems to claim that he's close in terms of "Eurasian" ancestry to Aris (~15-20%)..). If he's indeed partially Eurasian then if they've improved the stat once he's used; he could now more easily be taking up some of the West Eurasian ancestry in Horners and producing weird results just like with their former stat before they improved things. Everything from other runs, stats and even Horners' PCA positions fit better with the ~40% for Somalis & ~50% for Tigrinyas estimates as I pointed out recently here (http://anthromadness.blogspot.ae/2016/01/just-how-african-west-eurasian-are-horn.html). But the problem is that now even when Yorubas are used; Horners' West Eurasian scores still clearly go down in mostly the same way which is weird to say the least. I dunno what to make of this now. Honestly a bit annoyed by these imbeciles though, I'm gonna have to do a follow up now at Anthromadness addressing their fuck up.

Shaikorth
01-25-2016, 08:29 PM
It depends... Mota himself might not be 100% African (i.e. Kurd seems to claim that he's close in terms of "Eurasian" ancestry to Aris (~15-20%)..). If he's indeed partially Eurasian then if they've improved the stat once he's used; he could now more easily be taking up some of the West Eurasian ancestry in Horners and producing weird results just like with their former stat before they improved things. Everything from other runs, stats and even Horners' PCA positions fit better with the ~40% for Somalis & ~50% for Tigrinyas estimates as I pointed out recently here (http://anthromadness.blogspot.ae/2016/01/just-how-african-west-eurasian-are-horn.html). But the problem is that now even when Yorubas are used; Horners' West Eurasian scores still clearly go down in mostly the same way which is weird to say the least. I dunno what to make of this now. Honestly a bit annoyed by these imbeciles though, I'm gonna have to do a follow up now at Anthromadness addressing their fuck up.

Mota can be 100% African assuming East African is closer to Eurasian (without preference for West that ENF-like admixture would cause) due to a more recent separation from OOA than Pygmy/West African. Dinka D-stats show they don't have a West Eurasian preference so it's quite possible.

Awale
01-25-2016, 08:42 PM
Mota can be 100% African assuming East African is closer to Eurasian (without preference for West that ENF-like admixture would cause) due to a more recent separation from OOA than Pygmy/West African. Dinka D-stats show they don't have a West Eurasian preference so it's quite possible.

Yeah, this is something myself and Lank (http://www.anthrogenica.com/member.php?37-Lank) have been talking about in respect to "East Africans" for well over a year now... That a closer relationship between EAs and Eurasians seemingly exists. It is indeed possible... But again; their "new results" seem about as funky as their old ones but this time for Eastern Africans.

I'd like to see this run on North Africans & West Eurasians because Somalis cluster very much right in-between Egyptians/Tunisians/Moroccans and Anuaks (http://3.bp.blogspot.com/-QnFwBGkLRd0/Vo5ji7d_atI/AAAAAAAAD_A/oxbK2XcgeYM/s1600/Anuak%2BV%2BNorth%2BAfrica.png) & Tigrinyas sit right between Europeans like Sardinians and Anuaks (http://1.bp.blogspot.com/-NvtY_Ax8ckw/Vo5kzjsnk_I/AAAAAAAAD_Q/FTOT6XiQ35o/s1600/Anuak%2BV%2BSardinian.png) so either Sardinians are ~10% post-OoA "African" admixed (ignoring what Basal Eurasian may or may not be here...) and Egyptians are in fact ~30% "African" instead of their usual ~20% or Anuaks are pretty West Eurasian or generally Eurasian admixed and this is somehow being masked. <- All outcomes I find pretty unlikely.

gihanga.rwanda
01-25-2016, 09:01 PM
I'd like to see this run on North Africans & West Eurasians because Somalis cluster very much right in-between Egyptians/Tunisians/Moroccans and Anuaks (http://3.bp.blogspot.com/-QnFwBGkLRd0/Vo5ji7d_atI/AAAAAAAAD_A/oxbK2XcgeYM/s1600/Anuak%2BV%2BNorth%2BAfrica.png) & Tigrinyas sit right between Europeans like Sardinians and Anuaks (http://1.bp.blogspot.com/-NvtY_Ax8ckw/Vo5kzjsnk_I/AAAAAAAAD_Q/FTOT6XiQ35o/s1600/Anuak%2BV%2BSardinian.png) so either Sardinians are ~10% post-OoA "African" admixed (ignoring what Basal Eurasian may or may not be here...) and Egyptians are in fact ~30% "African" instead of their usual ~20% or Anuaks are pretty West Eurasian or generally Eurasian admixed and this is somehow being masked. <- All outcomes I find pretty unlikely.

Or we shouldn't be using the Anuak or Dinka as references for the majority portion of African ancestry in East Africans; Mota (despite the fact that ~25% of his ancestry derives from an ancient African population at the root of AMH, possibly even more so than the Ju'Hoansi) is more similar to Eurasians than Nilotes, siting in proximity to the Hadza.

gihanga.rwanda
01-25-2016, 09:10 PM
It depends... Mota himself might not be 100% African (i.e. Kurd seems to claim that he's close in terms of "Eurasian" ancestry to Aris (~15-20%)..). If he's indeed partially Eurasian then if they've improved the stat once he's used; he could now more easily be taking up some of the West Eurasian ancestry in Horners and producing weird results just like with their former stat before they improved things. Everything from other runs, stats and even Horners' PCA positions fit better with the ~40% for Somalis & ~50% for Tigrinyas estimates as I pointed out recently here (http://anthromadness.blogspot.ae/2016/01/just-how-african-west-eurasian-are-horn.html). But the problem is that now even when Yorubas are used; Horners' West Eurasian scores still clearly go down in mostly the same way which is weird to say the least. I dunno what to make of this now. Honestly a bit annoyed by these imbeciles though, I'm gonna have to do a follow up now at Anthromadness addressing their fuck up.

I haven't seen any convincing evidence that Mota is anything but 100% African, you're going to have to share the link to Kurd's argument. It appears that Mota and Ari share about the same levels of "San-like" ancestry, in the range of 15-25%, yet the Ari cluster closer to other Horners in relation to Mota, which one can easily argue is due to the ~10% Eurasian in the Ari.

Awale
01-25-2016, 09:11 PM
Or we shouldn't be using the Anuak or Dinka as references for the majority portion of African ancestry in East Africans; Mota (despite the fact that ~25% of his ancestry derives from an ancient African population at the root of AMH, closer to Ju'Hoansi) is closer to Eurasians than Nilotes, siting in proximity to the Hadza.

Yes but he shows strange Eurasian affinities, more than can be explained, in my humble opinion, by some pre-historic as hell "East Africans & Eurasians relationship" (~20% is not a joke...) so he's dubious if we're looking for the positioning of a full-on "African" but even if that is completely ignored; he's not representative of the kind of the African ancestry in Somalis which is closer looking to the non-West-Central-African-related ancestry in the Dinka. But even with your logic there... There's a problem. The exact same drop occurs when Yorubas & LBK are used (Somalis = ~30% & Habehsas = ~40%) which makes no sense given where Somalis cluster between Yorubas and West Eurasians (http://4.bp.blogspot.com/-SNA7KPzLpdY/Vo8EC3UUCYI/AAAAAAAAEA4/bc76ZTduNVw/s1600/18.png) or where Habeshas cluster between Yorubas and West Eurasians (http://4.bp.blogspot.com/-NZEMP4XxxAg/Vo73pse2GII/AAAAAAAAEAg/iK_hJ5MjM9s/s1600/Yoruba%2BV%2BWest%2BEurasian.png). Unless the predominantly or entirely West Eurasian individuals in question have a notable amount of African/SSA ancestry we don't know about (like Sardinians being ~10-20% for example) which I find to be a hard sell. The results for Horn Africans are strange to me just like the results for other Africans were before, something both you and I (http://anthromadness.blogspot.ae/2015/10/all-modern-africans-are-part-eurasian.html) took the time to point out.


I haven't seen any convincing evidence that Mota is anything but 100% African, you're going to have to share the link to Kurd's argument. It appears that Mota and Ari share about the same levels of "San-like" ancestry, in the range of 15-25%, yet the Ari cluster closer to other Horners in relation to Mota, which one can easily argue is due to the ~10% Eurasian in the Ari.

Again, whether or not he's 100% African is kind of irrelevant to your argument... He's not representative of the African element in Somalis but more connected to the Omotic-cluster (http://anthromadness.blogspot.ae/2015/01/the-omotic-cluster-horn-specific-cluster.html) (something Somalis are the very minimum of in the Horn) or some sort of very interesting pre-cursor to it or something along those lines. As for the Kurd stuff, he told me this via PMs. You're welcome to PM him yourself.

For the record, what Kurd told me is that he seems "broadly Eurasian" admixed not "West Eurasian" admixed in particular which lends credence to what Shaikorth suggested.

Chad Rohlfsen
01-25-2016, 09:15 PM
They have Mota at 2.1% West Eurasian.

gihanga.rwanda
01-25-2016, 09:55 PM
They have Mota at 2.1% West Eurasian.

That's only if Mota is modeled as (Yoruba, Druze), which is an inappropriate way to measure his "Eurasianess," also, let's not forget that Mota apparently lacks Neanderthal admixture, which according to the amendment to the study, still stands.

A Basal Eurasian or older African genome would clear a lot of this up.

kingjohn
01-25-2016, 10:13 PM
mota eurogenes k36
there is 2% euroasian but that leavs 98% african
dna
he is the father of africans
before the west euroeasian admixture took place.
p.s on contrary to him ethipopians and somalians do show significant euroasian admixture
adam

Passa
01-25-2016, 10:17 PM
That's only if Mota is modeled as (Yoruba, Druze), which is an inappropriate way to measure his "Eurasianess," also, let's not forget that Mota apparently lacks Neanderthal admixture, which according to the amendment to the study, still stands.

A Basal Eurasian or older African genome would clear a lot of this up.

See here, we may soon get 10 ky old Kenyan HG data: http://www.anthrogenica.com/showthread.php?97-Genetic-Genealogy-and-Ancient-DNA-in-the-News&p=134984&viewfull=1#post134984

Chad Rohlfsen
01-25-2016, 11:15 PM
That's only if Mota is modeled as (Yoruba, Druze), which is an inappropriate way to measure his "Eurasianess," also, let's not forget that Mota apparently lacks Neanderthal admixture, which according to the amendment to the study, still stands.

A Basal Eurasian or older African genome would clear a lot of this up.

Well, it's about 1.8% with LBK, and we have some Africans still in the negatives, so it can't be far off. I'm going to try and get the fixed genome and run these stats myself.

gihanga.rwanda
01-25-2016, 11:22 PM
Yes but he shows strange Eurasian affinities, more than can be explained, in my humble opinion, by some pre-historic as hell "East Africans & Eurasians relationship" (~20% is not a joke...) so he's dubious if we're looking for the positioning of a full-on "African" but even if that is completely ignored; he's not representative of the kind of the African ancestry in Somalis which is closer looking to the non-West-Central-African-related ancestry in the Dinka. But even with your logic there... There's a problem. The exact same drop occurs when Yorubas & LBK are used (Somalis = ~30% & Habehsas = ~40%) which makes no sense given where Somalis cluster between Yorubas and West Eurasians (http://4.bp.blogspot.com/-SNA7KPzLpdY/Vo8EC3UUCYI/AAAAAAAAEA4/bc76ZTduNVw/s1600/18.png) or where Habeshas cluster between Yorubas and West Eurasians (http://4.bp.blogspot.com/-NZEMP4XxxAg/Vo73pse2GII/AAAAAAAAEAg/iK_hJ5MjM9s/s1600/Yoruba%2BV%2BWest%2BEurasian.png). Unless the predominantly or entirely West Eurasian individuals in question have a notable amount of African/SSA ancestry we don't know about (like Sardinians being ~10-20% for example) which I find to be a hard sell. The results for Horn Africans are strange to me just like the results for other Africans were before, something both you and I (http://anthromadness.blogspot.ae/2015/10/all-modern-africans-are-part-eurasian.html) took the time to point out.



Again, whether or not he's 100% African is kind of irrelevant to your argument... He's not representative of the African element in Somalis but more connected to the Omotic-cluster (http://anthromadness.blogspot.ae/2015/01/the-omotic-cluster-horn-specific-cluster.html) (something Somalis are the very minimum of in the Horn) or some sort of very interesting pre-cursor to it or something along those lines. As for the Kurd stuff, he told me this via PMs. You're welcome to PM him yourself.

For the record, what Kurd told me is that he seems "broadly Eurasian" admixed not "West Eurasian" admixed in particular which lends credence to what Shaikorth suggested.


Well, it's about 1.8% with LBK, and we have some Africans still in the negatives, so it can't be far off. I'm going to try and get the fixed genome and run these stats myself.


See here, we may soon get 10 ky old Kenyan HG data: http://www.anthrogenica.com/showthread.php?97-Genetic-Genealogy-and-Ancient-DNA-in-the-News&p=134984&viewfull=1#post134984

Very interesting, I will go hunch and bet that these humans will share ancestry with the divergent San-like ancestry found in the Hadze, Sandawe, Omotics, and Mota. These were the likely indigenous inhabitants of East Africa south of the Ethiopian highlands.

Shaikorth
01-25-2016, 11:27 PM
Well, it's about 1.8% with LBK, and we have some Africans still in the negatives, so it can't be far off. I'm going to try and get the fixed genome and run these stats myself.

Some Nilotics who look more "Eurasian" on PCA's than Yorubas and Pygmies have no Eurasian in the model regardless of whether Mota or Yoruba was used to estimate it. Maybe it's just that Mota is closer to OOA than even those? Checking for East Eurasian ancestry also would likely do to test these ideas, closer to OOA should bring Africans closer to all Eurasians.

Chad Rohlfsen
01-26-2016, 12:04 AM
The one we have is no good. There's no use in trying to run any stats with it. Yoruba do have some West Eurasian. I'm sure it will be re-run by others, and this will be cleared up.

puntDNALKing
01-26-2016, 03:19 AM
Yeah, this is something myself and Lank (http://www.anthrogenica.com/member.php?37-Lank) have been talking about in respect to "East Africans" for well over a year now... That a closer relationship between EAs and Eurasians seemingly exists. It is indeed possible... But again; their "new results" seem about as funky as their old ones but this time for Eastern Africans.

I'd like to see this run on North Africans & West Eurasians because Somalis cluster very much right in-between Egyptians/Tunisians/Moroccans and Anuaks (http://3.bp.blogspot.com/-QnFwBGkLRd0/Vo5ji7d_atI/AAAAAAAAD_A/oxbK2XcgeYM/s1600/Anuak%2BV%2BNorth%2BAfrica.png) & Tigrinyas sit right between Europeans like Sardinians and Anuaks (http://1.bp.blogspot.com/-NvtY_Ax8ckw/Vo5kzjsnk_I/AAAAAAAAD_Q/FTOT6XiQ35o/s1600/Anuak%2BV%2BSardinian.png) so either Sardinians are ~10% post-OoA "African" admixed (ignoring what Basal Eurasian may or may not be here...) and Egyptians are in fact ~30% "African" instead of their usual ~20% or Anuaks are pretty West Eurasian or generally Eurasian admixed and this is somehow being masked. <- All outcomes I find pretty unlikely.

Here are Mota's results using the puntDNAL k15, puntDNAL k8 African only and Ethiohelix K10 African only.

puntDNAL k15

Population
S_Indian -
Mediterranean -
Siberian -
Wht_Nile_River 25.40
Amerindian 0.93
S_African 13.74
E_Asian -
Caucasian -
NE_European -
Omo_River 52.68
W_African 4.35
Horn_Of_Africa 2.79
Oceanian 0.11
Beringian -
SW_Asian -

puntDNAL k8

Population
Nilo_Saharan 29.06
Ubangian_Congo 7.36
W_Benue_Congo 0.38
Eastern_HG 7.53
E_Benue_Congo 12.69
Omotic 24.69
Southern_HG 14.06
Western_Semitic 4.25

Ethiohelix K10 African only

Population
Nilo-Saharan 20.46
East-Africa2 5.26
Mbuti-Pygmy 6.57
East_Africa1 7.57
Khoi-San 13.02
West_Africa 4.05
Hadza 7.19
Biaka-Pygmy 3.56
North-Africa -
Omotic 32.31

Mota is most likely 3% Eurasian and 97 Percent African, 73% East African, 8% Central African, 13% South African and the rest 5% West African.

pgbk87
01-26-2016, 03:21 AM
Very interesting, I will go hunch and bet that these humans will share ancestry with the divergent San-like ancestry found in the Hadze, Sandawe, Omotics, and Mota. These were the likely indigenous inhabitants of East Africa south of the Ethiopian highlands.

I am going to bet they will be the best proxy for the ALEA ancestry in Cushites and Ethio-Semites (Ancestral Lowland East African). It will be much more interesting if I am right :P

Kurd
01-26-2016, 04:22 AM
For the record, what Kurd told me is that he seems "broadly Eurasian" admixed not "West Eurasian" admixed in particular which lends credence to what Shaikorth suggested.

There are a couple of problems with Mota's genome.

In the Erratum to Gallego Llorente et al. 2015 : "leading to homozygote positions to the human reference genome being dropped as missing data (the analysis of admixture with Neanderthals and Denisovans was not affected). When those positions were included, 255,922 SNP out of 256,540 from the contemporary reference panel could be called in Mota"

What this means is that positions which are homozygotes for the minor allele (MAF=100%) , and positions which are homozygotes for the major allele (MAF=0%) were not included in the sequence!! The positions homozygotes for the major allele, may not affect the analysis much, but the positons homozygotes for the minor allele, most certainly will affect the analysis. In fact, whoever wrote the erratum is being a little generous by saying that Neanderthal and Denisovan analysis was not affected. I believe it actually was, but because we are dealing with values less than 1% it may not affect things as much. So this is 1st problem.

The 2nd problem with Mota is that of uracil residues, a common miscoding lesion in ancient DNA that results from the deamination of cytosine. I did 2 sets of Dstat runs, one with transition and transversion SNPs, the other with transverions only (to mitigate some of the effects of sequencing errors). This I did with the uncorrected Mota genome. With transitions included, Mota appeared less W Eurasian shifted that the other African genomes in the comparison, except when it came to Onge, where he was more Onge shifted. Once I removed transition SNPs ( transitions are more susceptible to sequencing errors in ancient genomes) from the run, he started to appear more W Eurasian.

Once we get the corrected genome, I will re-do the analysis. It is likely that he will be substantially more W Eurasian shifted than in these runs, especially when I remove transition SNPs.



L1
L2
R1
OUT
D
Z


MbutiPygmy
.MOTA
LBK
Karitiana
-0.0005
-0.159


Yoruba
.MOTA
LBK
Karitiana
0.0002
0.077


Khomani
.MOTA
LBK
Karitiana
0.007
2.158


Masai_Kinyawa
.MOTA
LBK
Karitiana
0.0211
6.829


Somali
.MOTA
LBK
Karitiana
0.0385
12.298


Ethiopian_Jew
.MOTA
LBK
Karitiana
0.0478
14.49










TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z


Yoruba
.MOTA
LBK
Karitiana
-0.0061
-1.03


MbutiPygmy
.MOTA
LBK
Karitiana
-0.0032
-0.494


Khomani
.MOTA
LBK
Karitiana
0.0029
0.443


Masai_Kinyaw
.MOTA
LBK
Karitiana
0.0125
2.086


Somali
.MOTA
LBK
Karitiana
0.03
5.016


Ethiopian_Je
.MOTA
LBK
Karitiana
0.0425
6.915

Kurd
01-26-2016, 04:23 AM
L1
L2
R1
OUT
D
Z


MbutiPygmy
.MOTA
Druze
Karitiana
0.001
0.357


Yoruba
.MOTA
Druze
Karitiana
0.0036
1.419


Khomani
.MOTA
Druze
Karitiana
0.0061
2.208


Masai_Kinyawa
.MOTA
Druze
Karitiana
0.0196
7.464


Somali
.MOTA
Druze
Karitiana
0.0327
12.378


Ethiopian_Jew
.MOTA
Druze
Karitiana
0.0418
14.677










TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z


Yoruba
.MOTA
Druze
Karitiana
-0.0018
-0.362


MbutiPygmy
.MOTA
Druze
Karitiana
-0.0003
-0.047


Khomani
.MOTA
Druze
Karitiana
0.0037
0.689


Masai_Kinyaw
.MOTA
Druze
Karitiana
0.0128
2.523


Somali
.MOTA
Druze
Karitiana
0.0278
5.514


Ethiopian_Je
.MOTA
Druze
Karitiana
0.0369
7.096

Kurd
01-26-2016, 04:24 AM
L1
L2
R1
OUT
D
Z


MbutiPygmy
.MOTA
BedouinB
Karitiana
0.0016
0.548


Yoruba
.MOTA
BedouinB
Karitiana
0.0045
1.63


Khomani
.MOTA
BedouinB
Karitiana
0.0051
1.705


Masai_Kinyawa
.MOTA
BedouinB
Karitiana
0.0201
7.182


Somali
.MOTA
BedouinB
Karitiana
0.034
12.141


Ethiopian_Jew
.MOTA
BedouinB
Karitiana
0.0416
13.87










TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z


Yoruba
.MOTA
BedouinB
Karitiana
-0.0012
-0.22


MbutiPygmy
.MOTA
BedouinB
Karitiana
-0.001
-0.167


Khomani
.MOTA
BedouinB
Karitiana
0.0037
0.651


Masai_Kinyaw
.MOTA
BedouinB
Karitiana
0.0133
2.522


Somali
.MOTA
BedouinB
Karitiana
0.0293
5.563


Ethiopian_Je
.MOTA
BedouinB
Karitiana
0.0371
6.733

Kurd
01-26-2016, 04:25 AM
L1
L2
R1
OUT
D
Z


MbutiPygmy
.MOTA
Kotias
Karitiana
-0.0038
-0.885


Yoruba
.MOTA
Kotias
Karitiana
-0.0036
-0.908


Khomani
.MOTA
Kotias
Karitiana
-0.0013
-0.316


Masai_Kinyawa
.MOTA
Kotias
Karitiana
0.0055
1.364


Somali
.MOTA
Kotias
Karitiana
0.0144
3.6


Ethiopian_Jew
.MOTA
Kotias
Karitiana
0.0196
4.704










TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z


Yoruba
.MOTA
Kotias
Karitiana
-0.0077
-0.978


Khomani
.MOTA
Kotias
Karitiana
-0.0031
-0.379


MbutiPygmy
.MOTA
Kotias
Karitiana
-0.0005
-0.056


Masai_Kinyaw
.MOTA
Kotias
Karitiana
0.0034
0.438


Somali
.MOTA
Kotias
Karitiana
0.017
2.219


Ethiopian_Je
.MOTA
Kotias
Karitiana
0.0261
3.283

Kurd
01-26-2016, 04:27 AM
L1
L2
R1
OUT
D
Z


Somali
.MOTA
Onge
Karitiana
-0.0074
-1.847


Ethiopian_Jew
.MOTA
Onge
Karitiana
-0.0073
-1.687


Masai_Kinyawa
.MOTA
Onge
Karitiana
-0.002
-0.476


Khomani
.MOTA
Onge
Karitiana
-0.0013
-0.291


Yoruba
.MOTA
Onge
Karitiana
0.0027
0.64


MbutiPygmy
.MOTA
Onge
Karitiana
0.0027
0.582










TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z


Ethiopian_Je
.MOTA
Onge
Karitiana
-0.0168
-1.737


Somali
.MOTA
Onge
Karitiana
-0.0163
-1.784


Masai_Kinyaw
.MOTA
Onge
Karitiana
-0.006
-0.006


MbutiPygmy
.MOTA
Onge
Karitiana
-0.0003
-0.035


Yoruba
.MOTA
Onge
Karitiana
-0.0002
-0.017


Khomani
.MOTA
Onge
Karitiana
0.001
0.107

Awale
01-26-2016, 04:52 AM
Well, it's about 1.8% with LBK, and we have some Africans still in the negatives, so it can't be far off. I'm going to try and get the fixed genome and run these stats myself.

Quoting Chad saying that these stats seem normal to him... You've won the argument, my Tutsi broski, I give up. The Sheikh's jokes aside though... I want you to look at something so you'll understand why these stats seem messed up in regards to Horn Africans like Somalis and Habeshas (I'd say Chad is right in that things mostly make sense for various other Africans now):

http://oi66.tinypic.com/2h37zol.jpg

^ That's not Mota V LBK there... That's Yoruba V LBK and we have essentially the exact same scores as when Mota is used by a difference of about ~1%. Right off the bat, this tells you we're getting rather asinine results in regards to Horn Africans like Somalis and Tigrinyas' results. We've been compared in this way to the likes of Yorubas or Dinkas or Anuaks + a completely West Eurasian (i.e. Sardinians) or mostly West Eurasian population and we know clearly that these aren't results one should expect either via formal stats or ADMIXTURE:

http://2.bp.blogspot.com/-Rf_DORbQeyA/VJ88hXbsV4I/AAAAAAAAB7U/VL6qNboFj4U/s1600/jjj.png

.
.

http://3.bp.blogspot.com/-7OA6NeHHdxo/VJTdh-XNzKI/AAAAAAAAB4Y/VgvWohRDC18/s1600/105r589.jpg

And of course, Tigrinyas cluster closer to full West Eurasians like LBKs than to Yorubas (http://4.bp.blogspot.com/-NZEMP4XxxAg/Vo73pse2GII/AAAAAAAAEAg/iK_hJ5MjM9s/s1600/Yoruba%2BV%2BWest%2BEurasian.png) = the idea they would be ~40% (the usual Somali range which is more consistent with how Somalis place between Yorubas and some Southern Levantines (http://4.bp.blogspot.com/-SNA7KPzLpdY/Vo8EC3UUCYI/AAAAAAAAEA4/bc76ZTduNVw/s1600/18.png)) in this model makes no sense. I can't say there might be something wrong with how they actually ran the analysis itself this time around (would be interested to see what Chad turns up with) but just that something, whatever it might be, is producing what look to be unreliable results for Cushitic and Ethio-Semitic speaking Horn Africans.


Mota is most likely 3% Eurasian and 97 Percent African, 73% East African, 8% Central African, 13% South African and the rest 5% West African.

Thanks for the input, broski. Interestingly, he doesn't actually pull that close to Eurasians in this global PCA from David (https://drive.google.com/file/d/0B9o3EYTdM8lQS0VoWEtaS1ZGV2M/view?usp=sharing). In fact, Somalis and the Beta Israel samples pretty much sit between him and Northern Africans and him and Sardinians respectively in this case too but noticeably more south than with the other PCAs which makes sense as he's somewhat north of where Anuaks would sit.

Chad Rohlfsen
01-26-2016, 05:36 AM
I think that if the general thought of 5% West Eurasian in the Yoruba and Mbuti is correct, then Mota is probably around 7%. I may already have the right one. I'll let everyone know as soon as I know.

vettor
01-26-2016, 06:25 AM
a computer glitch ?!

http://dispatchesfromturtleisland.blogspot.com.au/2016/01/oops-part-of-major-ethiopia-ancient-dna.html

puntDNALKing
01-26-2016, 06:41 AM
@Kurd

The Dstat for Mota being Onge shifted has a low Z score, so do you think we should put significant stock in it? I agree with you on all your other assumptions which illustrate the makeup of Mota. They have an affinity to East Africans and South Africans and a minor affinity to central Africans. Furthermore if you put Omotic populations in the mix the Z score would be higher than Ethiopians. I did an admixture which included Mota and East Africans, Mota scored 85% Omotic and Ariblacksmith score 93% Omotic.

Kurd
01-26-2016, 11:40 AM
@Kurd

The Dstat for Mota being Onge shifted has a low Z score, so do you think we should put significant stock in it? I agree with you on all your other assumptions which illustrate the makeup of Mota. They have an affinity to East Africans and South Africans and a minor affinity to central Africans. Furthermore if you put Omotic populations in the mix the Z score would be higher than Ethiopians. I did an admixture which included Mota and East Africans, Mota scored 85% Omotic and Ariblacksmith score 93% Omotic.

Punt, I forgot to mention that the Z score is low, especially with my runs involving transversions only, because I was working with a lower number of SNPs. The Z score whether positive or negative would likely have been quiet higher had I had more transversions.

With regards to your ADMIXTURE results, I am quiet confident they will change, if you use a corrected Mota sample. My guess is a significant number of positions for which Mota is homozygous for the minor allele (MAF=100%) were not called, as well as ones for which Mota is homozygous for the major allele (MAF=0%). The former positions once called should have a significant impact on formal as well as ADMIXTURE results. I am curious to see how his increased affinity to ENA changes with the updated data

Edit: A slight discrepancy between ADMIXTURE results and formal stats is expected, because the ADMIXTURE clusters, including the SSA one, are based on populations that have drifted since the time of Mota.

Shaikorth
01-26-2016, 12:08 PM
The one we have is no good. There's no use in trying to run any stats with it. Yoruba do have some West Eurasian. I'm sure it will be re-run by others, and this will be cleared up.

Yoruba showed no preference towards West or East Eurasia when run using high-coverage sequences which gets rid of ascertainment issues.

Granted it is still possible that they have West Eurasian, because Mbuti and Dinka had an East Eurasian preference over French. Assuming East Eurasians have old African and Yoruba have West Eurasian it'd explain those stats.

Work with complete sequences so far has been minimal, if someone has 10 terabytes of space, some time and willingness to obtain a certificate...https://www.simonsfoundation.org/life-sciences/simons-genome-diversity-project-dataset/

Awale
01-26-2016, 02:56 PM
Quoting Chad saying that these stats seem normal to him... You've won the argument, my Tutsi broski, I give up. The Sheikh's jokes aside though...

It seems I misunderstood here btw... You weren't replying to me there but you were using your phone and it somehow messed up to quote both me and Chad (as you said via PMs). Apologies for taking it the wrong way, khayo.





L1
L2
R1
OUT
D
Z


MbutiPygmy
.MOTA
LBK
Karitiana
-0.0005
-0.159


Yoruba
.MOTA
LBK
Karitiana
0.0002
0.077


Khomani
.MOTA
LBK
Karitiana
0.007
2.158


Masai_Kinyawa
.MOTA
LBK
Karitiana
0.0211
6.829


Somali
.MOTA
LBK
Karitiana
0.0385
12.298


Ethiopian_Jew
.MOTA
LBK
Karitiana
0.0478
14.49










TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z


Yoruba
.MOTA
LBK
Karitiana
-0.0061
-1.03


MbutiPygmy
.MOTA
LBK
Karitiana
-0.0032
-0.494


Khomani
.MOTA
LBK
Karitiana
0.0029
0.443


Masai_Kinyaw
.MOTA
LBK
Karitiana
0.0125
2.086


Somali
.MOTA
LBK
Karitiana
0.03
5.016


Ethiopian_Je
.MOTA
LBK
Karitiana
0.0425
6.915



Could you run these again but with Yorubas in Mota's place?

gihanga.rwanda
01-26-2016, 06:31 PM
^ That's not Mota V LBK there... That's Yoruba V LBK and we have essentially the exact same scores as when Mota is used by a difference of about ~1%. Right off the bat, this tells you we're getting rather asinine results in regards to Horn Africans like Somalis and Tigrinyas' results. We've been compared in this way to the likes of Yorubas or Dinkas or Anuaks + a completely West Eurasian (i.e. Sardinians) or mostly West Eurasian population and we know clearly that these aren't results one should expect either via formal stats or ADMIXTURE:


I agree, after looking at the East African results, they do look pretty inconsistent, do you also notice that the purported WE ancestry in the Luhya and Bantu Kenya increases significantly. This is something that needs to be addressed, but I am surprised Pontus Skoglund and David Reich didn't catch this in their review. Like I said in our pm's, they're really mishandling our chap Mota.

Kurd
01-27-2016, 03:37 AM
Could you run these again but with Yorubas in Mota's place?

These are with about 30K transversions only. The Z scores will likely be higher if more SNPs are used. I will try to do one with transitions and transversions and more markers if I have some time tomorrow.



TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z
SNPs


MbutiPygmy
Yoruba
LBK
Karitiana
0.0028
0.841
30088


Khomani
Yoruba
LBK
Karitiana
0.0092
2.88
30088


Masai_Kinyawa
Yoruba
LBK
Karitiana
0.0191
7.894
30087


Somali
Yoruba
LBK
Karitiana
0.0358
13.4
30088


Ethiopian_Jew
Yoruba
LBK
Karitiana
0.049
14.81
30089





TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z
SNPs


MbutiPygmy
Yoruba
Druze
Karitiana
0.0008
0.28
30240


Khomani
Yoruba
Druze
Karitiana
0.0055
2.078
30240


Masai_Kinyawa
Yoruba
Druze
Karitiana
0.0146
7.087
30239


Somali
Yoruba
Druze
Karitiana
0.0294
12.293
30240


Ethiopian_Jew
Yoruba
Druze
Karitiana
0.039
13.85
30241





TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z
SNPs


MbutiPygmy
Yoruba
BedouinB
Karitiana
0
0.002
30240


Khomani
Yoruba
BedouinB
Karitiana
0.0047
1.69
30240


Masai_Kinyawa
Yoruba
BedouinB
Karitiana
0.0142
6.229
30241


Somali
Yoruba
BedouinB
Karitiana
0.0303
12.069
30240


Ethiopian_Jew
Yoruba
BedouinB
Karitiana
0.038
12.77
30241

Kurd
01-27-2016, 03:39 AM
TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z
SNPs


Khomani
Yoruba
Kotias
Karitiana
0.0041
1.012
27415


MbutiPygmy
Yoruba
Kotias
Karitiana
0.0058
1.372
27415


Masai_Kinyawa
Yoruba
Kotias
Karitiana
0.0115
3.565
27414


Somali
Yoruba
Kotias
Karitiana
0.0251
6.836
27415


Ethiopian_Jew
Yoruba
Kotias
Karitiana
0.033
7.73
27416





TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z
SNPs


MbutiPygmy
Yoruba
Anatolia Neolithic
Karitiana
0.003
1.004
30159


Khomani
Yoruba
Anatolia Neolithic
Karitiana
0.0106
3.525
30156


Masai_Kinyawa
Yoruba
Anatolia Neolithic
Karitiana
0.0208
9.224
30155


Somali
Yoruba
Anatolia Neolithic
Karitiana
0.039
15.53
30158


Ethiopian_Jew
Yoruba
Anatolia Neolithic
Karitiana
0.0516
17.05
30157

Kurd
01-27-2016, 03:40 AM
L1
L2
R1
OUT
D
Z
SNPs


MbutiPygmy
.MOTA
Anatolia_Neolithic
Karitiana
-0.0004
-0.125
135489


Yoruba
.MOTA
Anatolia_Neolithic
Karitiana
0.0008
0.296
135487


Khomani
.MOTA
Anatolia_Neolithic
Karitiana
0.0079
2.648
135485


Masai_Kinyawa
.MOTA
Anatolia_Neolithic
Karitiana
0.0213
7.49
135484


Somali
.MOTA
Anatolia_Neolithic
Karitiana
0.0389
13.813
135488


Ethiopian_Jew
.MOTA
Anatolia_Neolithic
Karitiana
0.0487
15.884
135486











TRANSVERSION SNPs ONLY


L1
L2
R1
OUT
D
Z
SNPs


Yoruba
.MOTA
Anatolia_Neolithic
Karitiana
-0.004
-0.73
24565


MbutiPygmy
.MOTA
Anatolia_Neolithic
Karitiana
-0.0004
-0.062
24567


Khomani
.MOTA
Anatolia_Neolithic
Karitiana
0.0063
1.047
24563


Masai_Kinyaw
.MOTA
Anatolia_Neolithic
Karitiana
0.0163
2.943
24562


Somali
.MOTA
Anatolia_Neolithic
Karitiana
0.0347
6.328
24566


Ethiopian_Je
.MOTA
Anatolia_Neolithic
Karitiana
0.047
8.347
24564

puntDNALKing
01-27-2016, 05:25 AM
@ Kurd

Clearly, the transverse SNPs are more accurate, despite the fact of the low number of SNPs. Your Dstat correctly illustrates Mota's affinity to Eurasian populations when compared with Yoruba, despite their recent Eurasian drift.

@ Awale

I agree with you in regards to the admixture results being wacky. If they used Yoruba as a reference source for the african component and Anatolian for the Eurasian component, then Somalis and Oromos would be 50% Eurasian and Ethiopians around 58-60% Eurasian. However, if you use east africans like Mota or even Omotic and Nilotic populations, then the Eurasian percentage decreases. Horners would decrease by 8-10% in their Eurasian results.

Kurd
01-27-2016, 12:32 PM
@ Kurd

Clearly, the transverse SNPs are more accurate, despite the fact of the low number of SNPs. Your Dstat correctly illustrates Mota's affinity to Eurasian populations when compared with Yoruba, despite their recent Eurasian drift.
.

I agree, but at 24K SNPs the margin of error is slightly higher, and Mbuti"s and Yoruba's rankings with regards to Anatolians are certainly within the margin of error. I expect significant change with an updated Mota genome. BTW, anyone seen the updated genome for download yet?

Awale
01-27-2016, 06:21 PM
I agree, but at 24K SNPs the margin of error is slightly higher, and Mbuti"s and Yoruba's rankings with regards to Anatolians are certainly within the margin of error. I expect significant change with an updated Mota genome. BTW, anyone seen the updated genome for download yet?

Btw... Were you using the Lazaridis et al. Kenyan Somalis? They're, on average, just a percent or two less West Eurasian than Pagani et al.'s Somalian and Ethiopian Somali samples who're much more representative of peeps like the Sheikh and Punt (https://drive.google.com/file/d/0B2ARnUeK-Y8WSDh3VGR5WGtIYXM/view?usp=sharing). Thanks for the analyses btw, ya Kurdi. :)

puntDNALKing
01-27-2016, 07:36 PM
I agree, but at 24K SNPs the margin of error is slightly higher, and Mbuti"s and Yoruba's rankings with regards to Anatolians are certainly within the margin of error. I expect significant change with an updated Mota genome. BTW, anyone seen the updated genome for download yet?

No, I haven't seen the updated genome. Maybe this update will form an informative cluster in admixture, in unsupervised mode of course.

Kurd
01-27-2016, 07:59 PM
Btw... Were you using the Lazaridis et al. Kenyan Somalis? They're, on average, just a percent or two less West Eurasian than Pagani et al.'s Somalian and Ethiopian Somali samples who're much more representative of peeps like the Sheikh and Punt (https://drive.google.com/file/d/0B2ARnUeK-Y8WSDh3VGR5WGtIYXM/view?usp=sharing). Thanks for the analyses btw, ya Kurdi. :)

Welcome, ya Sheikh. My Somalis have the prefix Ayudo

Awale
01-27-2016, 08:07 PM
Welcome, ya Sheikh. My Somalis have the prefix Ayudo

Yep, those would be them Kenyans. Nearly half of them are somewhat possibly Borana Oromo admixed (http://anthromadness.blogspot.ae/2015/12/kenyan-somalis-not-all-borana-admixed.html) (that's seemingly why they pull away from other Somalis and toward non-Somali Ethiopian populations like Habeshas) and they seem to be just a smidgen, on average, less West Eurasian than other Somalis. Try switching them out in your dataset for Pagani et al.'s samples if you can but it's okay if that's too much trouble.

Lank
01-27-2016, 08:17 PM
It seems unlikely that Mota has notable Eurasian admixture, especially if South Sudanese do not. In the updated formal stats, Mota shows just 2% Eurasian relative to Yoruba. Admixture estimates for Africans (using formal stats) are nearly unchanged when using Mota instead of Yoruba.

ADMIXTURE should not be used as evidence for Eurasian admixture, as opposed to African structure. Before we had South Sudanese samples, they used to show 10% Eurasian in ADMIXTURE (and then it was mainly W Eurasian, although D stats tell a different story). Of course, Mota is even more Eurasian-shifted than South Sudanese, let alone Yoruba. Non-negligible Eurasian admixture in Mota is not likely unless it is also found in South Sudanese, and if it is then it's not probable that it's from the Neolithic wave.

Awale
01-27-2016, 09:13 PM
It seems unlikely that Mota has notable Eurasian admixture, especially if South Sudanese do not. In the updated formal stats, Mota shows just 2% Eurasian relative to Yoruba. Admixture estimates for Africans (using formal stats) are nearly unchanged when using Mota instead of Yoruba.

ADMIXTURE should not be used as evidence for Eurasian admixture, as opposed to African structure. Before we had South Sudanese samples, they used to show 10% Eurasian in ADMIXTURE (and then it was mainly W Eurasian, although D stats tell a different story). Of course, Mota is even more Eurasian-shifted than South Sudanese, let alone Yoruba. Non-negligible Eurasian admixture in Mota is not likely unless it is also found in South Sudanese, and if it is then it's not probable that it's from the Neolithic wave.

Plus there's there's the dubious nature of the "Eurasian" admixture he shows. From what I gather, it's all over the place... Like part South Asian, part West Eurasian, part East Eurasian etc. etc. <- It's unlikely as hell that he's some sort of cluster-fuck of Eurasian admixture. This seems more like an affinity than anything else.

kingjohn
01-27-2016, 09:43 PM
2-5% west euroasian admixture is still isn't high{not even 2 digits}
regards
adam

The_Lyonnist
01-27-2016, 11:05 PM
the extent of this so-called reflux demonstrates rather that the man does not come from Africa. It comes from Sumer is the first civilization. Why would the man he wait to get out of Africa to found a civilization? In short, all this support from non-African origin of man.

gihanga.rwanda
01-28-2016, 01:59 PM
It seems unlikely that Mota has notable Eurasian admixture, especially if South Sudanese do not. In the updated formal stats, Mota shows just 2% Eurasian relative to Yoruba. Admixture estimates for Africans (using formal stats) are nearly unchanged when using Mota instead of Yoruba.

ADMIXTURE should not be used as evidence for Eurasian admixture, as opposed to African structure. Before we had South Sudanese samples, they used to show 10% Eurasian in ADMIXTURE (and then it was mainly W Eurasian, although D stats tell a different story). Of course, Mota is even more Eurasian-shifted than South Sudanese, let alone Yoruba. Non-negligible Eurasian admixture in Mota is not likely unless it is also found in South Sudanese, and if it is then it's not probable that it's from the Neolithic wave.

Hey, do you have any idea what's causing the drop in LBK-like ancestry in East Africans? The Mota run isn't surprising IMHO, it's the Yoruba run that doesn't make much sense?

Lank
01-28-2016, 05:52 PM
Hey, do you have any idea what's causing the drop in LBK-like ancestry in East Africans? The Mota run isn't surprising IMHO, it's the Yoruba run that doesn't make much sense?
I'm no expert on the nature of the statistics, but Yoruba do have significant ancient East African ancestry. So they're not really an outgroup to ancestral East Africans. Furthermore, East Africans themselves had significant "divergent African" (not Y-DNA E or mtDNA L3) admixture, prior to Eurasian gene flow. Some of divergent mtDNA lineages shared with West Africans are dated to the Holocene.

I think the large difference between using Druze or LBK estimates is surprising. Druze might have 3% SSA, or something like that, which should not result in the ~10% difference we're seeing.

Awale
01-28-2016, 05:59 PM
I think the large difference between using Druze or LBK estimates is surprising. Druze might have 3% SSA, or something like that, which should not result in the ~10% difference we're seeing.

And again, we've seen stats where the likes of Anuaks or Yorubas have been used before with a population other than the Druze who don't really show any SSA element autosomally (like Sardinians) and these results are not what to expect. Also makes no sense given where Tigrinyas or Somalis cluster in a global PCA when compared to Yorubas (Tigrinyas would seriously cluster where the off-spring of a somewhat West Eurasian admixed Yoruba and a full-blown Sardinian would plot... Roughly anyway...), also makes no sense when compared to dozens of ADMIXTURE runs. Funky stuff is all I can say.

Kurd
01-29-2016, 02:07 PM
Hey, do you have any idea what's causing the drop in LBK-like ancestry in East Africans? The Mota run isn't surprising IMHO, it's the Yoruba run that doesn't make much sense?

The crux of the argument for a significant back migration event(s) into Africa post Mota was that Eurasian levels are much higher now in the Mota area than during Mota's time, thus indicating Mota types were replaced with groups (Somalis Ethiopians, etc) which are much more Eurasian admixed. This is shown in formal stats. The other claim made was that these back migrations also had a significant event on sub-saharans also, because even stats show them also more Eurasian shifted than Mota.

All of this of course is based on a Mota genome that was flawed to begin with. Apparently, ~250K homozygous positions were not called in Mota's genome, which IMO would significantly affect the outcome of any analysis, since I expect a substantial portion of those positions to be homozygous for the minor allele (MAF=100%). The updated genome will most likely show Eurasian admixture in Mota, which of course will negated the argument for widespread sub-saharan influence of the back migrations, and also reduce the magnitude of the Eurasian geneflow due to back migrations post Mota on E Africans.

As for the drop in LBK in E Africans, I believe that the back-migrants shared more drift with Anatolians (ENF) than with LBK, hence the decreased affinity of E Africans to LBK vs Anatolians. These stats illustrate this. You can clearly see that Somalis and Ethiopian Jews share far more drift with Anatolians than either LBK or Druze. If I have time later I will do a run comparing Druze to LBK, and also one using transversions only.



L1
L2
R1
OUT
D
Z
SNPs


Anatolia_Neolithic
Druze
Ethiopian_Je
Karitiana
0.0071
5.15
168435


Anatolia_Neolithic
Druze
Somali
Karitiana
0.006
4.406
168437


Anatolia_Neolithic
LBK_EN
Ethiopian_Jew
Karitiana
0.0048
2.839
168429


Anatolia_Neolithic
LBK_EN
Somali
Karitiana
0.0039
2.303
168431


Anatolia_Neolithic
LBK_EN
Khomani
Karitiana
0.0036
2.001
168428


Anatolia_Neolithic
LBK_EN
Yoruba
Karitiana
0.0034
1.95
168430


Anatolia_Neolithic
LBK_EN
Masai_Kinyawa
Karitiana
0.0032
1.892
168433


Anatolia_Neolithic
LBK_EN
MbutiPygmy
Karitiana
0.0027
1.385
168432


Anatolia_Neolithic
Druze
Masai_Kinyaw
Karitiana
0.0015
1.043
168439


Anatolia_Neolithic
Druze
Khomani
Karitiana
0.0011
0.783
168434

kingjohn
01-30-2016, 03:41 PM
manica admit
about the mistake her team made the west euroasian genefloiw was limited to
east africa not west central africa http://www.scientificamerican.com/article/error-found-in-study-of-first-ancient-african-genome/
regards
adam