PDA

View Full Version : Teal discovered ??!!



Pages : 1 [2] 3

vettor
11-20-2015, 08:56 PM
posted on another site
Bichon

Switzerland Grotte du Bichon M 13,560–13,770 cal. BP I2a U5b1h
he is positive for I2a1a2a L1286 which downstream from missing y-dna SNPs I2a1 P37.2/PF4004 and I2a1a CTS595

Krefter
11-20-2015, 09:30 PM
posted on another site
Bichon

Switzerland Grotte du Bichon M 13,560–13,770 cal. BP I2a U5b1h
he is positive for I2a1a2a L1286 which downstream from missing y-dna SNPs I2a1 P37.2/PF4004 and I2a1a CTS595

I2a1a2a has been found in Mesolithic Hungary and Sweden, Neolithic Hungary and Sweden. It could have been popular. There was a poster here who had I2a1a2a.

Tomenable
11-21-2015, 03:36 AM
With future ancient South Asian HG samples, ANE will be obsolete for the most of the Eurasia.

When it comes to modern populations, the highest % of ANE admixture is among Western Siberians:

Modern Western Siberians are 57% ANE:

From: "Reconstructing Genetic History of Siberian and Northeastern European Populations" (2015):

http://biorxiv.org/content/early/2015/10/18/029421

Abstract:

"Siberia and Western Russia are home to over 40 culturally and linguistically diverse indigenous ethnic groups. Yet, genetic variation of peoples from this region is largely uncharacterized. We present whole-genome sequencing data from 28 individuals belonging to 14 distinct indigenous populations from that region. We combine these datasets with additional 32 modern-day and 15 ancient human genomes to build and compare autosomal, Y-DNA and mtDNA trees. Our results provide new links between modern and ancient inhabitants of Eurasia. Siberians share 38% of ancestry with descendants of the 45,000-year-old Ust-Ishim people, who were previously believed to have no modern-day descendants. Western Siberians trace 57% of their ancestry to the Ancient North Eurasians, represented by the 24,000-year-old Siberian Mal'ta boy. In addition, Siberians admixtures are present in lineages represented by Eastern European hunter-gatherers from Samara, Karelia, Hungary and Sweden (from 8,000-6,600 years ago), as well as Yamnaya culture people (5,300-4,700 years ago) and modern-day northeastern Europeans. These results provide new evidence of ancient gene flow from Siberia into Europe."

Data Supplements:

http://biorxiv.org/content/early/2015/10/18/029421.figures-only

Data Figures:

http://biorxiv.org/content/biorxiv/suppl/2015/10/18/029421.DC1/029421-2.pdf

Krefter
11-21-2015, 08:25 AM
When it comes to modern populations, the highest % of ANE admixture is among Western Siberians:

Modern Western Siberians are 57% ANE:

From: "Reconstructing Genetic History of Siberian and Northeastern European Populations" (2015):
.......

Then why do they look just like Chinese or Japanese or Cambodian etc? They are not 57% from MA1 people. Native Americans on the other hand don't look totally East Asian so it is more believable. We have to take in account phenotype. Crazy numbers like that come from academics who are not familiar with the basics ancient DNA has revealed.

Nothing has refuted the idea of MA1-type ancestry yet. All the evidence used that CHG is a pure type of Eurasian with no MA1 or WHG admixture has been proven wrong with formal stats the paper didn't test. Some posters are saying ANE ancestry is fake because it's hard to believe a random person who lived in Siberia 24,000 years ago could be so connected to so many people today. There's no solid evidence against ANE ancestry as described by Laz and by Davidski.

Krefter
11-21-2015, 09:00 AM
Interesting D-stats provided by Davidski and others.

https://docs.google.com/spreadsheets/d/1TPByWTZL2eOhq21QvpeV-PFe7Ev04FAcxeD-72fyMnA/edit#gid=0

----------------------------------------------------------------------------------

CHG's placement on human tree?

Putting humans into a simple tree is very difficult. With ancient genomes which moderns are a mixture of it is much easier. Laz 2014 created a very reasonable tree with ancient genomes of human genetics. According to Laz's tree CHG has WHG-like and Basal Eurasian ancestry. Along with this CHG probably has MA1-type ancestry(will have to be confirmed with more formal stats). In no way does the data suggest CHG was pure Basal Eurasian like Jones 2015 modeled CHG.

We'll need many more formal stats, ADMIXTURE, PCA, etc., etc., to understand where in the human tree CHG is.

----------------------------------------------------------------------------

CHG ancestry in West Asia?

Formal stats confirm there's a lot ancestry closely related to CHG in South Asia and Yamnaya. But what about West Asia? ADMIXTURE and F3-drift stats from Jones 2015 tell us CHG's closest modern relatives live in the Caucasus. F3-drift stats didn't show an especially close relation with other West Asians. However CHG component of ADMIXTURE is also very high in the rest of West Asia.

We're going to need formal stats, not just ADMIXTURE, to back up the idea of CHG-like ancestry in West Asia. F3(not drift) shows that Assyrian/Lezgin/Turkish fit as a mixture of CHG+EEF(best proxy of Neolithic Near East?).

Furthermore D-stats showed that EEF and WHG have a closer relation to modern Caucasus than to CHG. EEF especially has a much closer relation to modern Caucasus. Lezgin appears to have MA1 or European-like ancestry on top of CHG. Loschbour and MA1 is closer to Lezgin as opposed to CHG than Stuttgart. Weirdly Yamnaya(had lots of ANE and WHG) wasn't as significantly closer to Lezgin as opposed to CHG.

bored
11-21-2015, 09:11 AM
Is it possible that ANE has been overestimated in Lezgins and maybe some other groups?

Tomenable
11-21-2015, 10:28 AM
Then why do they look just like Chinese or Japanese or Cambodian etc?

Eastern Siberians look like East Asians. Western Siberians, the ones with 57% ANE, have a distinct appearance on their own, IIRC.

Modern Siberians form 2 autosomal clusters - Eastern Siberians are the result of relatively recent East Asian westward migrations.

Shaikorth
11-21-2015, 10:59 AM
West Siberian Mansis were modeled as 57% ANE in the study because they grouped with Mal'ta and Afontova Gora in TreeMix and had a 43% East Siberian migration edge. Native Americans grouped with East Asians and had a 41% ANE edge in the same tree which is in line with what we knew already. The results seem plausible.

qpAdm models also support the idea, for example this one from David:

Mansi
Han 0.367
MA1 0.561
Georgian 0.073

chisq 0.278, tail prob 0.870155

Tomenable
11-21-2015, 12:35 PM
No other Amerindians score more ANE than Karitiana, who can be modelled as 42,5% ANE + 57% East Asian - see this spreadsheet:

https://docs.google.com/spreadsheets/d/1JVGdg2UsN3jYWgaoxAZu-QsAmuCaq3kT7FvqSXwUsAA/pubhtml

Arbogan
11-21-2015, 01:26 PM
Interesting D-stats provided by Davidski and others.

https://docs.google.com/spreadsheets/d/1TPByWTZL2eOhq21QvpeV-PFe7Ev04FAcxeD-72fyMnA/edit#gid=0

----------------------------------------------------------------------------------

CHG's placement on human tree?

Putting humans into a simple tree is very difficult. With ancient genomes which moderns are a mixture of it is much easier. Laz 2014 created a very reasonable tree with ancient genomes of human genetics. According to Laz's tree CHG has WHG-like and Basal Eurasian ancestry. Along with this CHG probably has MA1-type ancestry(will have to be confirmed with more formal stats). In no way does the data suggest CHG was pure Basal Eurasian like Jones 2015 modeled CHG.

We'll need many more formal stats, ADMIXTURE, PCA, etc., etc., to understand where in the human tree CHG is.

----------------------------------------------------------------------------

CHG ancestry in West Asia?

Formal stats confirm there's a lot ancestry closely related to CHG in South Asia and Yamnaya. But what about West Asia? ADMIXTURE and F3-drift stats from Jones 2015 tell us CHG's closest modern relatives live in the Caucasus. F3-drift stats didn't show an especially close relation with other West Asians. However CHG component of ADMIXTURE is also very high in the rest of West Asia.

We're going to need formal stats, not just ADMIXTURE, to back up the idea of CHG-like ancestry in West Asia. F3(not drift) shows that Assyrian/Lezgin/Turkish fit as a mixture of CHG+EEF(best proxy of Neolithic Near East?).

Furthermore D-stats showed that EEF and WHG have a closer relation to modern Caucasus than to CHG. EEF especially has a much closer relation to modern Caucasus. Lezgin appears to have MA1 or European-like ancestry on top of CHG. Loschbour and MA1 is closer to Lezgin as opposed to CHG than Stuttgart. Weirdly Yamnaya(had lots of ANE and WHG) wasn't as significantly closer to Lezgin as opposed to CHG. I wonder can anyone technically savy do D3/D4 statistics on kurds or iranians? I'm very interested in what our genetic relationship to CHG is. And how much CHG type ancestry we harbour.

Krefter
11-21-2015, 02:52 PM
Eastern Siberians look like East Asians. Western Siberians, the ones with 57% ANE, have a distinct appearance on their own, IIRC.

Modern Siberians form 2 autosomal clusters - Eastern Siberians are the result of relatively recent East Asian westward migrations.

Mansi people. (https://www.google.com/search?q=Mansi+people&espv=2&biw=1280&bih=699&source=lnms&tbm=isch&sa=X&ved=0ahUKEwisk8LF4aHJAhWCaz4KHTELBncQ_AUIBygB#imgr c=e8Meys6VoZ8LAM%3A) They do look East Asian. Some in Google images look European-admixed. What I suspect is European admixture that's heavy in MA1-affinity(via WHG and EHG) is causing them to have unrealistically high ANE scores. Unless unexplianble selection for traits is going on it is impossible for Mansi or any-other Siberians to be mostly descended from MA1. The overwhelmingly East Asian affinity of Siberians is obvious in their features. I've looked at Siberian mtDNA and it is clear they're not the brother of Native Americans who stayed in Siberia. They have East Asian ancestry that came into Siberi after the ancestors of Native Americans left.
http://www.nba.fi/hanti/kuvat1/SUK_36_2_lbox.jpg

Shaikorth
11-21-2015, 03:03 PM
Physical anthropology is irrelevant here. Ainus for one are more East Asian than Native Americans or any Siberians and have a diverging appearence, selection is what matters. Karitianas look quite a bit like East Asians. MA-1's and Afontova Gora's looks when alive are uncertain but Russians classified the skeletal type as some kind of East Eurasian.

parasar
11-21-2015, 04:39 PM
West Siberian Mansis were modeled as 57% ANE in the study because they grouped with Mal'ta and Afontova Gora in TreeMix and had a 43% East Siberian migration edge. Native Americans grouped with East Asians and had a 41% ANE edge in the same tree which is in line with what we knew already. The results seem plausible.

qpAdm models also support the idea, for example this one from David:

Mansi
Han 0.367
MA1 0.561
Georgian 0.073

chisq 0.278, tail prob 0.870155

Another Yeniseian population - the Kets.

http://biorxiv.org/content/early/2015/08/13/024554
"speakers of this language family occupied vast territories of Western and Central Siberia, from northern Mongolia in the south to the middle Yenisei River in the north, and from the Irtysh River in the west to the Angara River in the east3,4. Most Yeniseian-speaking tribes used to live south of the current Ket settlements. Ancestors of the Yeniseian people were tentatively associated5 with the Karasuk culture (3200-2700 YBP) of the upper Yenisei6 ...

ANE ancestry in Kets can be estimated using various f4-ratios from 27% to 62% (depending on the dataset and reference populations), vs. 2% in Nganasans, 30 ‒ 39% in Karitiana, and 23 ‒ 28% in Mayans (Suppl. file S7, see details in Suppl. Information Section 8). Integrating data by different methods, we conservatively estimate that Kets have the highest degree of ANE ancestry among all investigated modern Eurasian populations west of Chukotka and Kamchatka. We speculate that ANE ancestry in Kets was acquired in the Altai region, where the Bronze Age Okunevo culture was located, with a surprisingly close genetic proximity to Mal'ta ... Compared to Kets, Mal'ta was probably closer only to Motala12, although with a non-significant Z-score of -1.1 ... with the individual Ket884 removed: Z-score for f4(Mal'ta, Yoruba; Ket891, Motala12) became even less significant, -0.4 (Fig. 3A). Mal'ta ancestry in Kets was further supported by the TreeMix42 analysis (Fig. 2C).

In the full-genome dataset without transitions (main source of ancient DNA biases41), the Ket891 genome was the fourth best hit for Mal'ta, after Motala12, Karitiana, and Mixe ... Motala12 is the best hit for Mal'ta in our f3 statistic set-up ... the Kets were consistently placed at the top of the Eurasian spectrum of f3(Yoruba; Mal'ta, X) values (Suppl. Fig. 7.36, 7.37) or within the American spectrum (Suppl. Figs. 7.38-39) by statistics f3(Yoruba; Mal'ta, Ket891) and f3(Yoruba; Mal'ta, Ket884+891) computed for two datasets combining the Ket genomes and SNP array data (Suppl. Table 1) ...

U4, predominant in Kets .. haplogroup U, especially its U4 and U5 branches48, may be considered as a marker of West European hunter-gatherer (WHG) and ANE ancestry. In this light, high prevalence of haplogroup U4 in Kets and Selkups (Suppl. file S8) correlates well with large degrees of ANE ancestry in these populations ... more than 90% of Kets had haplogroup Q1a (of subclade Q1a2a1 as shown in our study) (Suppl. file S12), while haplogroups I1a2 and I2a1b3a occurred in just two Ket individuals ... the modern haplogroup Q forms another sister-branch of haplogroup R20. It is tempting to hypothesize that haplogroup Q1a correlates with ANE ancestry on a global scale"

Perhaps the EDAR in Motala is connected to later Siberians.

https://upload.wikimedia.org/wikipedia/commons/5/50/Ket_women_and_children_1914.png

Shaikorth
11-21-2015, 04:52 PM
ANE ancestry in Kets can be estimated using various f4-ratios from 27% to 62% (depending on the dataset and reference populations), vs. 2% in Nganasans, 30 ‒ 39% in Karitiana, and 23 ‒ 28% in Mayans (Suppl. file S7, see details in Suppl. Information Section 8). Integrating data by different methods, we conservatively estimate that Kets have the highest degree of ANE ancestry among all investigated modern Eurasian populations west of Chukotka and Kamchatka. We speculate that ANE ancestry in Kets was acquired in the Altai region, where the Bronze Age Okunevo culture was located, with a surprisingly close genetic proximity to Mal'ta ... Compared to Kets, Mal'ta was probably closer only to Motala12, although with a non-significant Z-score of -1.1 ... with the individual Ket884 removed: Z-score for f4(Mal'ta, Yoruba; Ket891, Motala12) became even less significant, -0.4 (Fig. 3A). Mal'ta ancestry in Kets was further supported by the TreeMix42 analysis (Fig. 2C).



I'll add that this study inferred the ANE percentages using a f4 ratio and assuming a two-way mix of ANE and East Eurasian. The changing East Asian references are reason for varying ANE estimates for each population. The lowest numbers are for models where the East Asian reference population is East Siberian, who have ANE themselves. Nganasans for instance had 2% ANE when modelled as MA-1+Yakut, but 30% as MA-1 + Dai.

Krefter
11-21-2015, 05:02 PM
Physical anthropology is irrelevant here. Ainus for one are more East Asian than Native Americans or any Siberians and have a diverging appearence, selection is what matters. Karitianas look quite a bit like East Asians. MA-1's and Afontova Gora's looks when alive are uncertain but Russians classified the skeletal type as some kind of East Eurasian.

Physical anthropology isn't irrelevant. From pictures and videos I've seen of Siberians they look exactly like East Asians. How could this be the case if they're mostly from a people totally unrelated to East Asians, besides being Eurasian? We can't just use selection as an excuse. DNA is more reliable however physical anthropology can be used as a sanity check. I don't know about anthropology and I wouldn't be surprised if under Pale skin and slanted eyes there's lots of features differnt from East Asians who tend to be uniform. But without being an anthropologist I'm very confident no Siberians are mostly from MA1's people.

Shaikorth
11-21-2015, 05:07 PM
Physical anthropology isn't irrelevant. From pictures and videos I've seen of Siberians they look exactly like East Asians. How could this be the case if they're mostly from a people totally unrelated to East Asians, besides being Eurasian? We can't just use selection as an excuse. The same excuse is used for mostly North European ancestry in S/C Asia. DNA is more reliable however physical anthropology can be used as a sanity check. I don't know about anthropology and I wouldn't be surprised if under Pale skin and slanted eyes there's lots of features differnt from East Asians who tend to be uniform. But without being an anthropologist I'm very confident no Siberians are mostly from MA1's people.

Even old school anthropologists have distinguished between East Asian, East Siberian, West Siberian and Native American people, so they certainly do not look exactly the same. Ainus' distinct appearence has been noted for a long time (and genetic studies show selection differences) but their overall ancestry is more East Asian than anything else. The point is that the high ANE in West Siberians is picked up by exactly the same genetic analyses that sees it in Native Americans, so it is not a fluke. It might be EHG instead of MA-1-like ANE, but that is irrelevant to differences between them and Native Americans, because EHG is also better fit than MA-1 for Native Americans.

Krefter
11-21-2015, 05:17 PM
Even old school anthropologists have distinguished between East Asian, East Siberian, West Siberian and Native American people, so they certainly do not look exactly the same. Ainus' distinct appearence has been noted for a long time (and genetic studies show selection differences) but their overall ancestry is more East Asian than anything else. The point is that the high ANE in West Siberians is picked up by exactly the same genetic analyses that sees it in Native Americans, so it is not a fluke. It might be EHG instead of MA-1-like ANE, but that is irrelevant to differences between them and Native Americans, because EHG is also better fit than MA-1 for Native Americans.

I agree with that. There's also the possibility European admixture is raising MA1 score. Besides Native Americans, MA1's closest relatives are Europeans.

parasar
11-21-2015, 05:38 PM
I agree with that. There's also the possibility European admixture is raising MA1 score. Besides Native Americans, MA1's closest relatives are Europeans.

Europeans have a wide variation of affinity to MA-1 (Maltese to Karelians). So I would not say that as a whole MA1's closest relatives in Eurasia are Europeans.

puntDNALKing
11-21-2015, 06:46 PM
David (Eurogenes) predicted this teal component and I also found it in my puntDNAL K11 and named it Caucas-Gedrosia. The influence of this component is higher among the north europeans than south europeans and among West Asia and South Asia. The Yamnaya also score 40 percent and considering I still don't have the CHG samples, I think it is likely those samples will form a more pure form of this component. Kurd has acquired the CHG samples and I wonder how they would score in the puntDNAL K11. I am sure they would score 90 percent on the Caucas-Gedrosia component.

Kurd
11-21-2015, 06:47 PM
I was able to convert and merge all datasets including CHG and Bichon, and Anthrogenica project members I will start posting stats shortly against CHG. If there are any special requests regarding Dstats or f3, you can post here. I was able to salvage 400K markers on most genomes.

evon
11-21-2015, 06:54 PM
David (Eurogenes) predicted this teal component and I also found it in my puntDNAL K11 and named it Caucas-Gedrosia. The influence of this component is higher among the north europeans than south europeans and among West Asia and South Asia. The Yamnaya also score 40 percent and considering I still don't have the CHG samples, I think it is likely those samples will form a more pure form of this component. Kurd has acquired the CHG samples and I wonder how they would score in the puntDNAL K11. I am sure they would score 90 percent on the Caucas-Gedrosia component.

Do you have a spreadsheet for K11? Would like to compare other European results with my own family..

Me:
3 Caucas-Gedrosia 16.72

Paternal aunt:
3 Caucas-Gedrosia 16.44

Maternal grandmother:
3 Caucas-Gedrosia 18.67

Maternal grandfathers brother:
3 Caucas-Gedrosia 20.87

Seems western Norwegians score around 15-20% Teal/CHG..

MfA
11-21-2015, 07:51 PM
I've edited Davidski's last PCA below. Interesting Yamnaya/Sintashta is lowest in Iran even less so than Armenians. And South-Central Asians got lion share of the CHG/-like admixture. Added the "-like" here because I'm suspecting most of this admixture are probably from very similar, yet local variant what I'd expect from Central Asian hunters.

Armenians and Iranians are on CHG cline
Armenians and North Caucasians are on Southwest Asian cline
Iranians and North Caucasians are on Yamnaya/Sintashta cline
North Caucasians and South-Central Asians are on Early/Middle Farmers cline

http://abload.de/img/ancient_west_eurasia_2aqul.png

DMXX
11-21-2015, 08:01 PM
I've edited Davidski's last PCA below. Interesting Yamnaya/Sintashta is lowest in Iran even less so than Armenians. And South-Central Asians got lion share of the CHG/-like admixture. Added the "-like" here because I'm suspecting most of this admixture are probably from very similar, yet local variant what I'd expect from Central Asian hunters.


Excellent work. Somewhat in line with the IBD data we saw a couple of months back (North Caucasians as well as Pamiri Tajiks display the most direct affinity to the BA steppe populations, followed by South-Central Asians, then West Asians). I don't think this proposition should be a surprise to anyone who has been observing the data in real-time since Haak et al.

I'd also guess that, if it were not for the ASE-related input in modern East Iranic speaking populations, they would plot closer to North Caucasians and modern West Iranic speaking groups. Probably on a cline between the North Caucasus and Iran, albeit shifted more northwards than Kurds and Iranians (overlapping with some North Caucasians possibly?).

Interesting that "West Iranics", per this modified plot, have more CHG ancestry than Armenians. Was this confirmed as well in Jones et al.? We'll have to wait for Kurd's formal stats to observe any corroboration along these lines.

Kurd
11-21-2015, 08:04 PM
This set is done with 2 outgroups to get an idea of absolute position vs Kotias.



OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Mbuti
SATP
Kotias
Gorilla
-0.451
-79
181537


Mbuti
Iceman
Kotias
Gorilla
-0.412
-80
251578


Mbuti
Abkhasian
Kotias
Gorilla
-0.405
-100
252973


Mbuti
Georgian
Kotias
Gorilla
-0.405
-100
252973


Mbuti
Adygei
Kotias
Gorilla
-0.400
-100
252973


Mbuti
Yamanya
Kotias
Gorilla
-0.400
-100
252171


Mbuti
RISE_baArm
Kotias
Gorilla
-0.399
-83
170899


Mbuti
RISE_baSin
Kotias
Gorilla
-0.399
-89
234998


Mbuti
RISE_baYam
Kotias
Gorilla
-0.397
-97
247719


Mbuti
RISE_baAndrov
Kotias
Gorilla
-0.396
-94
252482


Mbuti
Armenian
Kotias
Gorilla
-0.394
-100
252973


Mbuti
Lithuanian
Kotias
Gorilla
-0.391
-100
252973


Mbuti
Kalash
Kotias
Gorilla
-0.388
-100
252973


Mbuti
Samara_HG
Kotias
Gorilla
-0.388
-64
149957


Mbuti
Samara_HG
Kotias
Gorilla
-0.388
-64
149957


Mbuti
Tajik_Pomiri
Kotias
Gorilla
-0.387
-100
252973


Mbuti
Iranian
Kotias
Gorilla
-0.386
-100
252973


Mbuti
Stuttgart
Kotias
Gorilla
-0.386
-85
249150


Mbuti
LBK_EN
Kotias
Gorilla
-0.385
-100
252648


Mbuti
Balochi
Kotias
Gorilla
-0.384
-100
252973


Mbuti
Pathan
Kotias
Gorilla
-0.383
-100
252973


Mbuti
Brahui
Kotias
Gorilla
-0.382
-100
252973


Mbuti
Bichon
Kotias
Gorilla
-0.382
-74
170381


Mbuti
Motala_HG
Kotias
Gorilla
-0.381
-90
247982


Mbuti
Karelia_HG
Kotias
Gorilla
-0.379
-68
244839


Mbuti
Makrani
Kotias
Gorilla
-0.378
-100
252973


Mbuti
Burusho
Kotias
Gorilla
-0.378
-100
252973


Mbuti
LaBrana1
Kotias
Gorilla
-0.377
-71
240288


Mbuti
Loschbour
Kotias
Gorilla
-0.377
-72
250804


Mbuti
RISE_baOku
Kotias
Gorilla
-0.368
-59
124414


Mbuti
MA1
Kotias
Gorilla
-0.365
-60
181609


Mbuti
Kostenki14
Kotias
Gorilla
-0.357
-63
238571


Mbuti
Ust_Ishim
Kotias
Gorilla
-0.312
-57
252477

Tomenable
11-21-2015, 08:05 PM
Yamna people can be modelled genetically as a mix of EHG and either "Teal" or CHG. This also applies already to Khvalynsk people, even though in slightly different proportions. In anthropological terms I'm not sure how things looked like in Yamna culture (maybe the population was already so intermixed that it comprised a single anthropological type), but a morphological duality of Khvalynsk population can be observed.

Let's quote Mathieson's study:

"The unusually large cemetery at Khvalynsk contained southern Europeoid and northern Europeoid cranio-facial types, consistent with the possibility that people from the northern and southern steppes mingled and were buried here."

I guess that originally (before they merged into one population) EHG = northern types and "Teal" or CHG = southern types.

I wonder which cranio-facial type or types can be attributed to males SVP35 with R1b (grave 12) and SVP46 with R1a (grave 1) ???

Were both of them of northern Europeoid cranio-facial type, or was one or both of them of southern Europeoid type?

Kurd
11-21-2015, 08:05 PM
This one compares against Georgian



POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Georgian
SATP
Kotias
Gorilla
-0.0565
-8.344
181537


Georgian
Abkhasian
Kotias
Gorilla
0.0008
0.37
252973


Georgian
Iceman
Kotias
Gorilla
0.0026
0.49
251578


Georgian
Yamnaya
Kotias
Gorilla
0.0066
2.15
252171


Georgian
Adygei
Kotias
Gorilla
0.0076
4.103
252973


Georgian
RISE_baYam
Kotias
Gorilla
0.0104
3.065
247719


Georgian
RISE_baArm
Kotias
Gorilla
0.0119
2.491
170899


Georgian
RISE_baSin
Kotias
Gorilla
0.0132
3.21
234998


Georgian
RISE_baAndrov
Kotias
Gorilla
0.0138
3.754
252482


Georgian
Armenian
Kotias
Gorilla
0.0172
8.369
252973


Georgian
Armenian
Kotias
Gorilla
0.0172
8.369
252973


Georgian
RISE_baSca
Kotias
Gorilla
0.0198
4.787
247881


Georgian
Lithuanian
Kotias
Gorilla
0.0202
8.32
252973


Georgian
Kalash
Kotias
Gorilla
0.0246
10.111
252973


Georgian
Samara_HG
Kotias
Gorilla
0.0253
3.91
149957


Georgian
Samara_HG
Kotias
Gorilla
0.0253
3.91
149957


Georgian
Tajik_Pomiri
Kotias
Gorilla
0.0266
11.57
252973


Georgian
LBK_EN
Kotias
Gorilla
0.0274
9.974
252648


Georgian
Stuttgart
Kotias
Gorilla
0.0276
5.948
249150


Georgian
Iranian
Kotias
Gorilla
0.0282
13.029
252973


Georgian
Balochi
Kotias
Gorilla
0.0318
15.527
252973


Georgian
Pathan
Kotias
Gorilla
0.0322
16.4
252973


Georgian
Motala_HG
Kotias
Gorilla
0.0333
9.117
247982


Georgian
Karelia_HG
Kotias
Gorilla
0.0335
5.554
244839


Georgian
Brahui
Kotias
Gorilla
0.034
16.717
252973


Georgian
LaBrana1
Kotias
Gorilla
0.0382
6.585
240288


Georgian
Loschbour
Kotias
Gorilla
0.0387
7.274
250804


Georgian
Bichon
Kotias
Gorilla
0.0395
7.241
170381


Georgian
Makrani
Kotias
Gorilla
0.0407
20.405
252973


Georgian
MA1
Kotias
Gorilla
0.0529
7.842
181609


Georgian
RISE_baOku
Kotias
Gorilla
0.0547
8.081
124414


Georgian
Kostenki14
Kotias
Gorilla
0.0682
11.466
238571


Georgian
Ust_Ishim
Kotias
Gorilla
0.1222
21.534
252477

puntDNALKing
11-21-2015, 08:09 PM
Do you have a spreadsheet for K11? Would like to compare other European results with my own family..

Me:
3 Caucas-Gedrosia 16.72

Paternal aunt:
3 Caucas-Gedrosia 16.44

Maternal grandmother:
3 Caucas-Gedrosia 18.67

Maternal grandfathers brother:
3 Caucas-Gedrosia 20.87

Seems western Norwegians score around 15-20% Teal/CHG..

I am at work and I will post the spreadsheet once I get home. Keep in mind the caucas-gedrosia component is not pure teal or CHG because I did not use the recent CHG sample in this run, however, it is a mix of ANE and Teal-like component.

Kurd
11-21-2015, 08:27 PM
POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Mbuti
.Zephyrous
KOTIAS
Gorilla
-0.3876
-75.363
102536


Mbuti
.Hanna
KOTIAS
Gorilla
-0.3867
-75.613
102469


Mbuti
.Kurd_C2
KOTIAS
Gorilla
-0.3818
-69.862
73819


Mbuti
.Arbogan
KOTIAS
Gorilla
-0.3802
-71.995
86250


Mbuti
.DMXX
KOTIAS
Gorilla
-0.3791
-70.762
86310


Mbuti
.Pak_Gujjar
KOTIAS
Gorilla
-0.3772
-75.561
101278


Mbuti
.Rukha
KOTIAS
Gorilla
-0.3764
-66.932
73784


Mbuti
.Sein
KOTIAS
Gorilla
-0.3763
-72.744
101345


Mbuti
.Kurd_C1
KOTIAS
Gorilla
-0.3754
-65.818
73767


Mbuti
.Passa
KOTIAS
Gorilla
-0.3754
-65.676
73553


Mbuti
.Bol_Nat
KOTIAS
Gorilla
-0.374
-71.55
100512


Mbuti
.Dluffy
KOTIAS
Gorilla
-0.3711
-67.207
73702


Mbuti
.Farid
KOTIAS
Gorilla
-0.371
-64.55
73693


Mbuti
.Bored
KOTIAS
Gorilla
-0.3707
-63.905
73823


Mbuti
.Khana
KOTIAS
Gorilla
-0.3705
-66.489
73729


Mbuti
.Kandhari
KOTIAS
Gorilla
-0.3692
-65.956
73751


Mbuti
.Zara
KOTIAS
Gorilla
-0.3658
-62.537
73847


Mbuti
.Kenji
KOTIAS
Gorilla
-0.3635
-64.533
73681


Mbuti
.Jesus
KOTIAS
Gorilla
-0.3625
-66.051
73653


Mbuti
.Varun
KOTIAS
Gorilla
-0.3611
-67.965
102384


Mbuti
KURD
KOTIAS
Gorilla
-0.3601
-64.961
73702


Mbuti
.Awale
KOTIAS
Gorilla
-0.228
-38.383
73758

Kurd
11-21-2015, 08:28 PM
POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Georgian
.Hanna
KOTIAS
Gorilla
0.0122
2.276
102469


Georgian
.Zephyrous
KOTIAS
Gorilla
0.0129
2.433
102536


Georgian
.Kurd_C2
KOTIAS
Gorilla
0.0159
2.845
73819


Georgian
.Arbogan
KOTIAS
Gorilla
0.0207
3.997
86250


Georgian
.DMXX
KOTIAS
Gorilla
0.0217
4.241
86310


Georgian
.Rukha
KOTIAS
Gorilla
0.0217
3.9
73784


Georgian
.Passa
KOTIAS
Gorilla
0.0228
4.228
73553


Georgian
.Kurd_C1
KOTIAS
Gorilla
0.0241
4.275
73767


Georgian
.Pak_Gujjar
KOTIAS
Gorilla
0.0265
4.99
101278


Georgian
.Sein
KOTIAS
Gorilla
0.0281
5.485
101345


Georgian
.Bored
KOTIAS
Gorilla
0.0295
5.251
73823


Georgian
.Bol_Nat
KOTIAS
Gorilla
0.0296
5.728
100512


Georgian
.Dluffy
KOTIAS
Gorilla
0.0305
5.638
73702


Georgian
.Kandhari
KOTIAS
Gorilla
0.0318
5.759
73751


Georgian
.Khana
KOTIAS
Gorilla
0.0319
5.56
73729


Georgian
.Farid
KOTIAS
Gorilla
0.0324
5.404
73693


Georgian
.Zara
KOTIAS
Gorilla
0.0376
6.142
73847


Georgian
KURD
KOTIAS
Gorilla
0.0377
6.922
73702


Georgian
.Kenji
KOTIAS
Gorilla
0.04
7.137
73681


Georgian
.Jesus
KOTIAS
Gorilla
0.042
7.244
73653


Georgian
.Varun
KOTIAS
Gorilla
0.0469
8.878
102384


Georgian
.Awale
KOTIAS
Gorilla
0.2051
36.082
73758

Tomenable
11-21-2015, 08:40 PM
Yamna people can be modelled genetically as a mix of EHG and either "Teal" or CHG. This also applies already to Khvalynsk people, even though in slightly different proportions. In anthropological terms I'm not sure how things looked like in Yamna culture (maybe the population was already so intermixed that it comprised a single anthropological type), but a morphological duality of Khvalynsk population can be observed.

Let's quote Mathieson's study:

"The unusually large cemetery at Khvalynsk contained southern Europeoid and northern Europeoid cranio-facial types, consistent with the possibility that people from the northern and southern steppes mingled and were buried here."

I guess that originally (before they merged into one population) EHG = northern types and "Teal" or CHG = southern types.

I wonder which cranio-facial type or types can be attributed to males SVP35 with R1b (grave 12) and SVP46 with R1a (grave 1) ???

Were both of them of northern Europeoid cranio-facial type, or was one or both of them of southern Europeoid type?

Reconstructions of Khvalynsk people can be found here:

http://www.forumbiodiversity.com/showthread.php/38378-First-Indo-Europeans?p=1025629&viewfull=1#post1025629

MfA
11-21-2015, 08:53 PM
[TABLE]
How about Kurd_N ?

Kurd
11-21-2015, 09:26 PM
How about Kurd_N ?

Forgot them, Ill post in a couple of hours

Edit: the result will reflect the average of 6 of them

MfA
11-21-2015, 09:34 PM
Forgot them, Ill post in a couple of hours

Edit: the result will reflect the average of 6 of them

I was wondering the lone 23andMe V3 sample though, I thought you had a sample.

Tomenable
11-21-2015, 09:46 PM
I'm not sure why some people consider Yamna as the "original Indoeuropeans". That culture was not the first stage of PIE, but the last one:

According to linguist Robert Stephen Paul Beekes: "There seems to be no doubt that the Yamnaya culture represents the LAST phase of an Indo-European linguistic unity, although there were probably already significant dialectal differences within it."

Marija Gimbutas who was the original author of the Kurgan Hypothesis also didn't consider Yamna as the earliest PIE, but a later stage.

Gimbutas saw early stages of PIE in Chalcolithic steppe cultures which preceded Yamna - Samara and Khvalynsk cultures.

According to Mayu's blog, Corded Ware was descended from PIE groups which emigrated from the steppe during Early Yamna phase:

http://forwhattheywereweare.blogspot.com/p/blog-page.html

AFAIK, all Yamna samples collected so far are from later phases of Yamna culture, so they are people who stayed in that part of the steppe after several other groups had already emigrated in various directions before. Which may be the reason why all that we can see there is ht35.

All Yamna samples tested so far, are from period called by Mayu "Indoeuropean stage 3", not from "stage 2":

Stage 2:

http://4.bp.blogspot.com/-CeK4gX-YKHI/U1dzIlx0RjI/AAAAAAAAChs/M1ZJy-pavXY/s1600/IE2.png

Stage 3:

http://4.bp.blogspot.com/-4lp4uK-eJ0M/U1dzIjfPofI/AAAAAAAACho/6Ljb1_71WlY/s1600/IE3.png

By the time of Stage 3 some haplogroups and subclades - such as R1b-L51 - could already be outside of the steppe zone.

Maybe R1b-L51 - which is absent from Yamna samples known to date - was in Coţofeni culture or in Ezero culture ???

Tomenable
11-21-2015, 10:15 PM
^^^
PIEs were a patriarchal, clan based, polygynous society. Inheriting and family ties were male-based (when they married, a woman was entering the family of her husband, ceasing to be part of her former family). Each clan was descended in terms of Y-DNA from a common male ancestor (its founder), and members of each clan carried a different haplogroup or at least a different subclade. Any migrations were also clan-based. According to Mayu during Stage 2 (Early Yamna), there were migrations in 3 directions - to the south-west into the Balkans, to the north-west into Central Europe and to the north-east (founders of Afanasevo culture). We can suppose that each of those Early Yamna migrating clans or tribes, carried a different haplogroup or at least a different subclade as their main marker (if more than one clan moved, then of course they could carry more than one specific subclade or haplogroup). So - for example - maybe those migrating into the Balkans carried predominantly R1b-L51, those migrating into Central Europe carried predominantly R1a-Z283 and those migrating into areas where they later created the Afanasevo culture carried predominantly R1a-Z93*. We are left for example with R1b-Z2103 which, it seems, stayed in the region until the Late Yamna period.

*And also R1a-Tarim, it seems.

ZephyrousMandaru
11-21-2015, 10:17 PM
Is CHG something like a combination of Basal Eurasian + UHG?

Arbogan
11-21-2015, 10:40 PM
POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Mbuti
.Zephyrous
KOTIAS
Gorilla
-0.3876
-75.363
102536


Mbuti
.Hanna
KOTIAS
Gorilla
-0.3867
-75.613
102469


Mbuti
.Kurd_C2
KOTIAS
Gorilla
-0.3818
-69.862
73819


Mbuti
.Arbogan
KOTIAS
Gorilla
-0.3802
-71.995
86250


Mbuti
.DMXX
KOTIAS
Gorilla
-0.3791
-70.762
86310


Mbuti
.Pak_Gujjar
KOTIAS
Gorilla
-0.3772
-75.561
101278


Mbuti
.Rukha
KOTIAS
Gorilla
-0.3764
-66.932
73784


Mbuti
.Sein
KOTIAS
Gorilla
-0.3763
-72.744
101345


Mbuti
.Kurd_C1
KOTIAS
Gorilla
-0.3754
-65.818
73767


Mbuti
.Passa
KOTIAS
Gorilla
-0.3754
-65.676
73553


Mbuti
.Bol_Nat
KOTIAS
Gorilla
-0.374
-71.55
100512


Mbuti
.Dluffy
KOTIAS
Gorilla
-0.3711
-67.207
73702


Mbuti
.Farid
KOTIAS
Gorilla
-0.371
-64.55
73693


Mbuti
.Bored
KOTIAS
Gorilla
-0.3707
-63.905
73823


Mbuti
.Khana
KOTIAS
Gorilla
-0.3705
-66.489
73729


Mbuti
.Kandhari
KOTIAS
Gorilla
-0.3692
-65.956
73751


Mbuti
.Zara
KOTIAS
Gorilla
-0.3658
-62.537
73847


Mbuti
.Kenji
KOTIAS
Gorilla
-0.3635
-64.533
73681


Mbuti
.Jesus
KOTIAS
Gorilla
-0.3625
-66.051
73653


Mbuti
.Varun
KOTIAS
Gorilla
-0.3611
-67.965
102384


Mbuti
KURD
KOTIAS
Gorilla
-0.3601
-64.961
73702


Mbuti
.Awale
KOTIAS
Gorilla
-0.228
-38.383
73758



I think this correlates well with dodecad K12b scores. Assyrians and Armenians always had higher allele sharing with Caucasians. Despite in overall west-iranics being more similar to Caucasians than the former. So it looks like you can see an even, rising distribution of this form of ancestry from south-east to north-west asia. Beginning from south-central asia/north-india, going to the Iranian plateau, to eastern Anatolia and south-eastern Caucasus, and maxing out in north-west caucasus. This is totally in line with Caucasus like allele frequencies on admixture tests. And probably gives us a good idea of their immigration routes south-east. From the Caucasus, via Iran to south-central asia/north india. South and east of these areas, the Caucasus-like admixture seems to rapidly decline.

I wonder if you could include Iraqi-jews and the mandean samples, and Lebanese and druze, in your next run.

I'd also do a run with as the donor target being satsurblia, which seemed more similar to south-central Asians than kotias. That might give a more representative picture of this form of ancestry in south-Asians/south central Asians and maybe west-iranics. Thanks for doing these runs. It's highly appreciated.

Shaikorth
11-21-2015, 10:52 PM
I agree with that. There's also the possibility European admixture is raising MA1 score. Besides Native Americans, MA1's closest relatives are Europeans.

The recent study about Kets actually had a Ket individual closer to MA-1 than any Native Americans in a f4 test, using a full-genome dataset which is the highest quality that can be hoped for. Using the Human Origins (Reich lab's SNP set) data all Native Americans and some Europeans were closer than that Ket, their idea was that the larger sample sizes in the Human Origins dataset skewed the results but I think there may be issues with SNP selection or perhaps homozygosity of the set's samples too.

Then there are these, I think the modern samples again consist of a few high-coverage sequenced individuals. Modern Europeans very likely aren't responsible for ANE affinity in West Siberia.
http://s16.postimg.org/sy3n09xb9/dstat.jpg

Kurd
11-21-2015, 11:16 PM
To even the playing field, I applied threshold filters to remove any SNPs which are missing in >7% of the database (~175 individuals). Here are the results sorted with most similar profile to Kotias on top.

Most of you may be left wondering as to why ADMIXTURE shows Balochis Brahuis and Makranis scoring the highest CHG component, whereas Dstats show Caucuses groups most similar overall to Kotias.

Well, here is my million dollar answer. Although Balochis et al score highest in CHG, they have some S Asian admixture that works against them when their whole genome is compared. I sit in the same boat, although I score high CHG, the other parts of my admixture, E Asian, African, etc, distance me from Kotias when a genome wide comparison is made. It is kind of like oracle distance in admixture. It is hypothetically possible to score 70% CHG in ADMIXTURE, yet be distant from Kotias in oracles, if the other 30% is say African or E Asian.



POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Georgian
SATSURBILA
KOTIAS
Gorilla
-0.0647
-7.652
58028


Georgian
Abkhasian
KOTIAS
Gorilla
-0.0015
-0.624
81229


Georgian
Iceman
KOTIAS
Gorilla
0.0023
0.342
80901


Georgian
Yamnaya
KOTIAS
Gorilla
0.0035
0.945
81173


Georgian
RISE_baYam
KOTIAS
Gorilla
0.0045
0.95
79724


Georgian
Adygei
KOTIAS
Gorilla
0.0049
2.328
81229


Georgian
RISE_baSin
KOTIAS
Gorilla
0.0073
1.397
75521


Georgian
RISE_baAndrov
KOTIAS
Gorilla
0.0098
2.19
81113


Georgian
Kurd_N5
KOTIAS
Gorilla
0.0104
1.967
81032


Georgian
Kurd_N4
KOTIAS
Gorilla
0.0106
1.943
81035


Georgian
.Hanna
KOTIAS
Gorilla
0.0115
2.036
77612


Georgian
.Kurd_C2
KOTIAS
Gorilla
0.0122
2.115
68286


Georgian
.Zephyrous
KOTIAS
Gorilla
0.013
2.286
77627


Georgian
RISE_baSca
KOTIAS
Gorilla
0.0142
2.687
79975


Georgian
Armenian
KOTIAS
Gorilla
0.0143
6.152
81229


Georgian
Armenian
KOTIAS
Gorilla
0.0143
6.152
81229


Georgian
Samara_HG
KOTIAS
Gorilla
0.0143
1.735
49910


Georgian
Samara_HG
KOTIAS
Gorilla
0.0143
1.735
49910


Georgian
Lithuanian
KOTIAS
Gorilla
0.0174
6.343
81229


Georgian
Kurd_N3
KOTIAS
Gorilla
0.0174
3.353
81037


Georgian
RISE_baArm
KOTIAS
Gorilla
0.0181
2.571
54524


Georgian
RISE_baArm
KOTIAS
Gorilla
0.0181
2.571
54524


Georgian
.Arbogan
KOTIAS
Gorilla
0.0186
3.457
80867


Georgian
.Pak_Gujjar
KOTIAS
Gorilla
0.0194
3.406
76750


Georgian
Kurd_N1
KOTIAS
Gorilla
0.0202
3.618
81040


Georgian
.Kurd_SE
KOTIAS
Gorilla
0.0205
3.687
68276


Georgian
.Passa
KOTIAS
Gorilla
0.0209
3.632
68056


Georgian
.Rukha
KOTIAS
Gorilla
0.0213
3.739
68256


Georgian
.DMXX
KOTIAS
Gorilla
0.0223
4.168
80938


Georgian
Kurd_N6
KOTIAS
Gorilla
0.0224
4.203
81038


Georgian
LBK_EN
KOTIAS
Gorilla
0.0228
6.878
81222


Georgian
Kalash
KOTIAS
Gorilla
0.0235
8.227
81229


Georgian
Kurd_N2
KOTIAS
Gorilla
0.0236
4.395
80358


Georgian
.Kurd_C1
KOTIAS
Gorilla
0.0239
4.233
68240


Georgian
Iranian
KOTIAS
Gorilla
0.0255
9.833
81229


Georgian
Tajik_Pomiri
KOTIAS
Gorilla
0.0256
9.49
81229


Georgian
.Sein
KOTIAS
Gorilla
0.0263
4.917
76800


Georgian
Pathan
KOTIAS
Gorilla
0.028
12.656
81229


Georgian
.Bol_Nat
KOTIAS
Gorilla
0.0281
5.058
76137


Georgian
.Dluffy
KOTIAS
Gorilla
0.0281
4.999
68184


Georgian
Karelia_HG
KOTIAS
Gorilla
0.0288
4.111
79927


Georgian
Balochi
KOTIAS
Gorilla
0.0302
12.606
81229


Georgian
.Khana
KOTIAS
Gorilla
0.0303
5.035
68216


Georgian
.Bored
KOTIAS
Gorilla
0.0304
5.335
68295


Georgian
Stuttgart
KOTIAS
Gorilla
0.0305
5.817
79944


Georgian
.Kandhari
KOTIAS
Gorilla
0.031
5.38
68224


Georgian
Motala_HG
KOTIAS
Gorilla
0.0312
7.318
80653


Georgian
Brahui
KOTIAS
Gorilla
0.0321
13.629
81229


Georgian
Starcevo_EN
KOTIAS
Gorilla
0.0321
2.78
24829


Georgian
.Farid
KOTIAS
Gorilla
0.0321
5.383
68180


Georgian
KURD
KOTIAS
Gorilla
0.0347
6.124
68192


Georgian
LaBrana1
KOTIAS
Gorilla
0.0349
4.95
77381


Georgian
Tajik_Afghan
KOTIAS
Gorilla
0.0351
12.03
81041


Georgian
.Zara
KOTIAS
Gorilla
0.0353
5.751
68312


Georgian
Makrani
KOTIAS
Gorilla
0.0368
16.317
81229


Georgian
.Kenji
KOTIAS
Gorilla
0.0369
6.415
68163


Georgian
Bichon
KOTIAS
Gorilla
0.0384
5.886
54082


Georgian
Loschbour
KOTIAS
Gorilla
0.0403
6.551
80509


Georgian
.Jesus
KOTIAS
Gorilla
0.0409
6.687
68149


Georgian
.Varun
KOTIAS
Gorilla
0.0422
7.342
77516


Georgian
MA1
KOTIAS
Gorilla
0.0529
6.539
58356


Georgian
Kostenki14
KOTIAS
Gorilla
0.0596
8.195
76753


Georgian
Ust_Ishim
KOTIAS
Gorilla
0.1165
18.509
81076


Georgian
.Awale
KOTIAS
Gorilla
0.203
34.745
68231

Kurd
11-21-2015, 11:21 PM
I was wondering the lone 23andMe V3 sample though, I thought you had a sample.

I just posted the Kurd_N results in the above table. I was able to separate them. They were part of a dataset I got from Reich Lab. Can't really tell from their code numbers who genotyped them. They may be diaspora from Turkey, perhaps migrants to Kazakhstan.

Krefter
11-21-2015, 11:55 PM
The recent study about Kets actually had a Ket individual closer to MA-1 than any Native Americans in a f4 test, using a full-genome dataset which is the highest quality that can be hoped for. Using the Human Origins (Reich lab's SNP set) data all Native Americans and some Europeans were closer than that Ket, their idea was that the larger sample sizes in the Human Origins dataset skewed the results but I think there may be issues with SNP selection or perhaps homozygosity of the set's samples too.

Then there are these, I think the modern samples again consist of a few high-coverage sequenced individuals. Modern Europeans very likely aren't responsible for ANE affinity in West Siberia.
http://s16.postimg.org/sy3n09xb9/dstat.jpg

ANE was diluted in Russia because of backmigration from the West starting in the Bronze age. Yamnaya and EHG have more MA1-affinity than modern Europeans.

parasar
11-22-2015, 12:24 AM
ANE was diluted in Russia because of backmigration from the West starting in the Bronze age. Yamnaya and EHG have more MA1-affinity than modern Europeans.

And perhaps ANE was diluted in the Kets & Mansi too because of a subsequent Eastern Eurasian (usually called East Asian) input. Siberians may very well have been all ANE prior to the East Asian influx. No actually they were, as AG2 and MA1 were indeed early Siberians!

Kurd
11-22-2015, 01:08 AM
POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Georgian
KOTIAS
SATSURBILA
Gorilla
-0.0689
-9.26
58028


Georgian
RISE_baYam
SATSURBILA
Gorilla
-0.0024
-0.429
66016


Georgian
Abkhasian
SATSURBILA
Gorilla
-0.0018
-0.651
67537


Georgian
Iceman
SATSURBILA
Gorilla
-0.0003
-0.031
67209


Georgian
Yamnaya
SATSURBILA
Gorilla
0.0015
0.324
67498


Georgian
Lezgin
SATSURBILA
Gorilla
0.0023
0.79
67537


Georgian
Chechen
SATSURBILA
Gorilla
0.0061
2.031
67537


Georgian
Adygei
SATSURBILA
Gorilla
0.0065
2.57
67537


Georgian
RISE_baSin
SATSURBILA
Gorilla
0.0103
1.616
62060


Georgian
Kurd_N4
SATSURBILA
Gorilla
0.0116
1.86
67390


Georgian
Armenian
SATSURBILA
Gorilla
0.012
4.299
67537


Georgian
RISE_baSca
SATSURBILA
Gorilla
0.0121
1.86
66308


Georgian
RISE_baArm
SATSURBILA
Gorilla
0.0127
1.52
44427


Georgian
RISE_baArm
SATSURBILA
Gorilla
0.0127
1.52
44427


Georgian
RISE_baAndrov
SATSURBILA
Gorilla
0.0128
2.41
67395


Georgian
Lithuanian
SATSURBILA
Gorilla
0.0132
4.043
67537


Georgian
Kurd_N2
SATSURBILA
Gorilla
0.0146
2.304
66818


Georgian
Kalash
SATSURBILA
Gorilla
0.0187
5.429
67537


Georgian
Kurd_N3
SATSURBILA
Gorilla
0.0196
3.062
67396


Georgian
Iranian_Jew
SATSURBILA
Gorilla
0.0197
6.617
67537


Georgian
Kurd_N5
SATSURBILA
Gorilla
0.0202
3.187
67390


Georgian
Samara_HG
SATSURBILA
Gorilla
0.0208
1.982
41079


Georgian
Hungarian
SATSURBILA
Gorilla
0.021
8.025
67537


Georgian
Kurd_N1
SATSURBILA
Gorilla
0.0217
3.378
67394


Georgian
Iraqi_Jew
SATSURBILA
Gorilla
0.0226
7.028
67537


Georgian
Bulgarian
SATSURBILA
Gorilla
0.0231
7.959
67537


Georgian
Tajik_Pomiri
SATSURBILA
Gorilla
0.0236
7.31
67537


Georgian
LBK_EN
SATSURBILA
Gorilla
0.0237
5.899
67531


Georgian
Iranian
SATSURBILA
Gorilla
0.0245
8.141
67537


Georgian
Stuttgart
SATSURBILA
Gorilla
0.025
3.934
66405


Georgian
Pathan
SATSURBILA
Gorilla
0.027
9.668
67537


Georgian
Loschbour
SATSURBILA
Gorilla
0.0274
3.923
66923


Georgian
Balochi
SATSURBILA
Gorilla
0.0278
9.97
67537


Georgian
Bichon
SATSURBILA
Gorilla
0.0289
3.41
44003


Georgian
Sindhi
SATSURBILA
Gorilla
0.0294
10.074
67537


Georgian
Motala_HG
SATSURBILA
Gorilla
0.0305
6.07
67019


Georgian
Tajik_Afghan
SATSURBILA
Gorilla
0.0307
8.33
67391


Georgian
Brahui
SATSURBILA
Gorilla
0.0311
11.377
67537


Georgian
Karelia_HG
SATSURBILA
Gorilla
0.0313
3.698
66393


Georgian
Burusho
SATSURBILA
Gorilla
0.0322
11.255
67537


Georgian
Makrani
SATSURBILA
Gorilla
0.0346
12.477
67537


Georgian
RISE_baOku
SATSURBILA
Gorilla
0.0408
3.669
31937


Georgian
LaBrana1
SATSURBILA
Gorilla
0.043
5.045
63897


Georgian
Starcevo_EN
SATSURBILA
Gorilla
0.0489
3.583
20201


Georgian
MA1
SATSURBILA
Gorilla
0.0522
5.535
48361


Georgian
Kostenki14
SATSURBILA
Gorilla
0.0589
6.857
63568


Georgian
Nganasan
SATSURBILA
Gorilla
0.0769
14.48
67537


Georgian
Ulchi
SATSURBILA
Gorilla
0.0794
16.746
67537


Georgian
Ust_Ishim
SATSURBILA
Gorilla
0.1133
16.102
67392



Kalash's position is noteworthy in comparison to neighboring SC Asians. Transition SNPs have been excluded

Arbogan
11-22-2015, 01:41 AM
POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


Georgian
KOTIAS
SATSURBILA
Gorilla
-0.0689
-9.26
58028


Georgian
RISE_baYam
SATSURBILA
Gorilla
-0.0024
-0.429
66016


Georgian
Abkhasian
SATSURBILA
Gorilla
-0.0018
-0.651
67537


Georgian
Iceman
SATSURBILA
Gorilla
-0.0003
-0.031
67209


Georgian
Yamnaya
SATSURBILA
Gorilla
0.0015
0.324
67498


Georgian
Lezgin
SATSURBILA
Gorilla
0.0023
0.79
67537


Georgian
Chechen
SATSURBILA
Gorilla
0.0061
2.031
67537


Georgian
Adygei
SATSURBILA
Gorilla
0.0065
2.57
67537


Georgian
RISE_baSin
SATSURBILA
Gorilla
0.0103
1.616
62060


Georgian
Kurd_N4
SATSURBILA
Gorilla
0.0116
1.86
67390


Georgian
Armenian
SATSURBILA
Gorilla
0.012
4.299
67537


Georgian
RISE_baSca
SATSURBILA
Gorilla
0.0121
1.86
66308


Georgian
RISE_baArm
SATSURBILA
Gorilla
0.0127
1.52
44427


Georgian
RISE_baArm
SATSURBILA
Gorilla
0.0127
1.52
44427


Georgian
RISE_baAndrov
SATSURBILA
Gorilla
0.0128
2.41
67395


Georgian
Lithuanian
SATSURBILA
Gorilla
0.0132
4.043
67537


Georgian
Kurd_N2
SATSURBILA
Gorilla
0.0146
2.304
66818


Georgian
Kalash
SATSURBILA
Gorilla
0.0187
5.429
67537


Georgian
Kurd_N3
SATSURBILA
Gorilla
0.0196
3.062
67396


Georgian
Iranian_Jew
SATSURBILA
Gorilla
0.0197
6.617
67537


Georgian
Kurd_N5
SATSURBILA
Gorilla
0.0202
3.187
67390


Georgian
Samara_HG
SATSURBILA
Gorilla
0.0208
1.982
41079


Georgian
Hungarian
SATSURBILA
Gorilla
0.021
8.025
67537


Georgian
Kurd_N1
SATSURBILA
Gorilla
0.0217
3.378
67394


Georgian
Iraqi_Jew
SATSURBILA
Gorilla
0.0226
7.028
67537


Georgian
Bulgarian
SATSURBILA
Gorilla
0.0231
7.959
67537


Georgian
Tajik_Pomiri
SATSURBILA
Gorilla
0.0236
7.31
67537


Georgian
LBK_EN
SATSURBILA
Gorilla
0.0237
5.899
67531


Georgian
Iranian
SATSURBILA
Gorilla
0.0245
8.141
67537


Georgian
Stuttgart
SATSURBILA
Gorilla
0.025
3.934
66405


Georgian
Pathan
SATSURBILA
Gorilla
0.027
9.668
67537


Georgian
Loschbour
SATSURBILA
Gorilla
0.0274
3.923
66923


Georgian
Balochi
SATSURBILA
Gorilla
0.0278
9.97
67537


Georgian
Bichon
SATSURBILA
Gorilla
0.0289
3.41
44003


Georgian
Sindhi
SATSURBILA
Gorilla
0.0294
10.074
67537


Georgian
Motala_HG
SATSURBILA
Gorilla
0.0305
6.07
67019


Georgian
Tajik_Afghan
SATSURBILA
Gorilla
0.0307
8.33
67391


Georgian
Brahui
SATSURBILA
Gorilla
0.0311
11.377
67537


Georgian
Karelia_HG
SATSURBILA
Gorilla
0.0313
3.698
66393


Georgian
Burusho
SATSURBILA
Gorilla
0.0322
11.255
67537


Georgian
Makrani
SATSURBILA
Gorilla
0.0346
12.477
67537


Georgian
RISE_baOku
SATSURBILA
Gorilla
0.0408
3.669
31937


Georgian
LaBrana1
SATSURBILA
Gorilla
0.043
5.045
63897


Georgian
Starcevo_EN
SATSURBILA
Gorilla
0.0489
3.583
20201


Georgian
MA1
SATSURBILA
Gorilla
0.0522
5.535
48361


Georgian
Kostenki14
SATSURBILA
Gorilla
0.0589
6.857
63568


Georgian
Nganasan
SATSURBILA
Gorilla
0.0769
14.48
67537


Georgian
Ulchi
SATSURBILA
Gorilla
0.0794
16.746
67537


Georgian
Ust_Ishim
SATSURBILA
Gorilla
0.1133
16.102
67392



Kalash's position is noteworthy in comparison to neighboring SC Asians. Transition SNPs have been excluded

It definitely looks like satsurblia is a somewhat better ancestral proxy than kotias for south-central Asians and south-asians. CHG might have been more gedrosia-like at one point. Perhaps saturblia and kotias underwent a similar process to the bronze age and iron age Armenians, the remaining CHGs that went north-wards, gradually became more kotias like, which seems closer to modern day Caucasians than satsurblia.

If my theory is correct, it would mean that a satsurblia-like immigration left the Caucasus earlier and reached Iran, south-central asia and south-asia, much earlier than the later CHGs that mixed into yamnaya. Which imo makes sense. Since the Caucasus allele frequencies are closer to other west-Eurasian signals than Gedrosia. I definitely think that satsurblia and perhaps something even earlier, represents eastern CHG.

Thanks again my friend, for your efforts with D3 statistics.

Kurd
11-22-2015, 01:52 AM
Just for fun....



POP2
POP1
TARGET
OUTGROUP
D
Z
SNP


LBK_EN
Kostenki14
Ust_Ishim
Gorilla
-0.0198
-2.531
87728


LBK_EN
Ulchi
Ust_Ishim
Gorilla
-0.0198
-4.114
93095


LBK_EN
MA1
Ust_Ishim
Gorilla
-0.0182
-2.155
66793


LBK_EN
Nganasan
Ust_Ishim
Gorilla
-0.0175
-3.557
93095


LBK_EN
Samara_HG
Ust_Ishim
Gorilla
-0.017
-2.067
56572


LBK_EN
Samara_HG
Ust_Ishim
Gorilla
-0.017
-2.067
56572


LBK_EN
LaBrana1
Ust_Ishim
Gorilla
-0.016
-2.19
88276


LBK_EN
BICHON
Ust_Ishim
Gorilla
-0.0136
-1.911
61543


LBK_EN
RISE_baOku
Ust_Ishim
Gorilla
-0.0133
-1.473
44322


LBK_EN
Loschbour
Ust_Ishim
Gorilla
-0.0132
-2.04
92251


LBK_EN
Karelia_HG
Ust_Ishim
Gorilla
-0.0131
-1.802
91574


LBK_EN
.Pak_Gujjar
Ust_Ishim
Gorilla
-0.0094
-1.512
88058


LBK_EN
.Bored
Ust_Ishim
Gorilla
-0.0089
-1.445
78406


LBK_EN
Motala_HG
Ust_Ishim
Gorilla
-0.0084
-1.808
92431


LBK_EN
.Bol_Nat
Ust_Ishim
Gorilla
-0.0081
-1.419
87361


LBK_EN
RISE_baAndrov
Ust_Ishim
Gorilla
-0.0053
-1.03
92913


LBK_EN
Lithuanian
Ust_Ishim
Gorilla
-0.0044
-1.293
93095


LBK_EN
Burusho
Ust_Ishim
Gorilla
-0.0039
-1.144
93095


LBK_EN
.Sein
Ust_Ishim
Gorilla
-0.0031
-0.532
88104


LBK_EN
Yamnaya
Ust_Ishim
Gorilla
-0.0025
-0.578
93036


LBK_EN
.Kurd_SE
Ust_Ishim
Gorilla
-0.0025
-0.416
78377


LBK_EN
.Dluffy
Ust_Ishim
Gorilla
-0.0022
-0.337
78272


LBK_EN
Kalash
Ust_Ishim
Gorilla
-0.0017
-0.462
93095


LBK_EN
.Kurd_C2
Ust_Ishim
Gorilla
-0.0015
-0.263
78392


LBK_EN
Pathan
Ust_Ishim
Gorilla
-0.0014
-0.429
93095


LBK_EN
.Kenji
Ust_Ishim
Gorilla
-0.0009
-0.15
78247


LBK_EN
RISE_baSca
Ust_Ishim
Gorilla
-0.0006
-0.106
91524


LBK_EN
Sindhi
Ust_Ishim
Gorilla
-0.0006
-0.19
93095


LBK_EN
.Khana
Ust_Ishim
Gorilla
-0.0004
-0.06
78310


LBK_EN
Tajik_Afghan
Ust_Ishim
Gorilla
0.0007
0.16
92892


LBK_EN
.Hanna
Ust_Ishim
Gorilla
0.001
0.186
89020


LBK_EN
Iceman
Ust_Ishim
Gorilla
0.0013
0.19
92655


LBK_EN
Hungarian
Ust_Ishim
Gorilla
0.0013
0.413
93095


LBK_EN
RISE_baSin
Ust_Ishim
Gorilla
0.0017
0.3
85888


LBK_EN
RISE_baYam
Ust_Ishim
Gorilla
0.0017
0.336
91171


LBK_EN
.Rukha
Ust_Ishim
Gorilla
0.0018
0.297
78344


LBK_EN
.Varun
Ust_Ishim
Gorilla
0.002
0.332
88904


LBK_EN
.KURD
Ust_Ishim
Gorilla
0.0021
0.617
78290


LBK_EN
.Kandhari
Ust_Ishim
Gorilla
0.0023
0.362
78313


LBK_EN
Kurd_N5
Ust_Ishim
Gorilla
0.0032
0.565
92881


LBK_EN
Chechen
Ust_Ishim
Gorilla
0.0036
1.081
93095


LBK_EN
.Farid
Ust_Ishim
Gorilla
0.0037
0.604
78262


LBK_EN
RISE_irArm
Ust_Ishim
Gorilla
0.0039
0.211
10239


LBK_EN
Adygei
Ust_Ishim
Gorilla
0.0044
1.436
93095


LBK_EN
Kurd_N4
Ust_Ishim
Gorilla
0.005
0.886
92882


LBK_EN
Tajik_Pomir
Ust_Ishim
Gorilla
0.0053
1.51
93095


LBK_EN
Balochi
Ust_Ishim
Gorilla
0.0058
1.771
93095


LBK_EN
Bulgarian
Ust_Ishim
Gorilla
0.0064
2.065
93095


LBK_EN
Lezgin
Ust_Ishim
Gorilla
0.0064
1.885
93095


LBK_EN
Abkhasian
Ust_Ishim
Gorilla
0.007
2.193
93095


LBK_EN
Brahui
Ust_Ishim
Gorilla
0.0077
2.296
93095


LBK_EN
Georgian
Ust_Ishim
Gorilla
0.0084
2.57
93095


LBK_EN
Stuttgart
Ust_Ishim
Gorilla
0.0084
1.474
91531


LBK_EN
Iranian
Ust_Ishim
Gorilla
0.0092
2.756
93095


LBK_EN
Armenian
Ust_Ishim
Gorilla
0.0099
3.128
93095


LBK_EN
.Arbogan
Ust_Ishim
Gorilla
0.0105
1.843
92686


LBK_EN
Kurd_N1
Ust_Ishim
Gorilla
0.0112
1.86
92886


LBK_EN
.Kurd_C1
Ust_Ishim
Gorilla
0.0113
1.878
78332


LBK_EN
Iranian_Jew
Ust_Ishim
Gorilla
0.0121
3.671
93095


LBK_EN
Kurd_N3
Ust_Ishim
Gorilla
0.0122
2.144
92886


LBK_EN
Iraqi_Jew
Ust_Ishim
Gorilla
0.0122
3.397
93095


LBK_EN
.DMXX
Ust_Ishim
Gorilla
0.0125
2.269
92761


LBK_EN
SATP
Ust_Ishim
Gorilla
0.0126
1.554
67386


LBK_EN
.Passa
Ust_Ishim
Gorilla
0.0127
2.094
78129


LBK_EN
Makrani
Ust_Ishim
Gorilla
0.0136
4.18
93095


LBK_EN
Starcevo_EN
Ust_Ishim
Gorilla
0.0138
1.148
28100


LBK_EN
Kurd_N2
Ust_Ishim
Gorilla
0.0144
2.569
92114


LBK_EN
RISE_baArm
Ust_Ishim
Gorilla
0.0146
2.094
61616


LBK_EN
RISE_baArm
Ust_Ishim
Gorilla
0.0146
2.094
61616


LBK_EN
KK1
Ust_Ishim
Gorilla
0.0157
2.595
81069


LBK_EN
.Zephyrous
Ust_Ishim
Gorilla
0.0179
2.956
89031


LBK_EN
.Jesus
Ust_Ishim
Gorilla
0.0196
3.12
78228


LBK_EN
.Zara
Ust_Ishim
Gorilla
0.0197
3.116
78418


LBK_EN
.Awale
Ust_Ishim
Gorilla
0.1436
22.657
78334

puntDNALKing
11-22-2015, 02:19 AM
I am at work and I will post the spreadsheet once I get home. Keep in mind the caucas-gedrosia component is not pure teal or CHG because I did not use the recent CHG sample in this run, however, it is a mix of ANE and Teal-like component.

Here is a link to the spreadsheet: here (https://www.dropbox.com/s/5z7qgxgwuacntmf/spreadsheetelven%20%281%29.xlsx?dl=0)

Krefter
11-22-2015, 02:21 AM
This tree makes the most sense to me. Click to get a bigger image.
http://www.anthrogenica.com/attachment.php?attachmentid=6679&stc=1

Mystery Eurasian contributing to proto-MA1 and proto-Dai is supported by.
(Kostienki WHG/MA1; Dai/Onge Chimp)<0
(Kostienki WHG/MA1; Ust-Ishim Chimp)=0

Kostienki and proto-MA1 in same Western branch is supported by
(Ust-Ishim/East Asian MA1; Kostenki Chimp)<0
(Ust-Ishim/East Asian Kostenki; MA1 Chimp)<0

proto-MA1 and proto-Kostienki admixture in WHG is supported by.
(Kostienki WHG; MA1 Chimp)<0
(MA1 WHG; Kostienki Chimp)<0

Basal Eurasian admixture in CHG and EEF is supported by.
(CHG/EEF MA1; Ust-Ishim Chimp)<0
(CHG/EEF Kostenki; Ust-Ishim Chimp)<0

WHG admixture in CHG and EEF is supported by.
(MA1 WHG; CHG/EEF Chimp)<0
(Kostenki WHG; CHG/EEF Chimp)<0

MA1-related admixture in CHG is supported by.
((EEF WHG; Ust-Ishim Chimp)-(CHG WHG; Ust-Ishim Chimp))+(EEF CHG; MA1 Chimp)<0
(WHG EHG; CHG Chimp)<0
(MA1 WHG; CHG Chimp)<(MA1 WHG; Armenian Chimp)<(MA1 WHG; EEF Chimp)
(EEF CHG; MA1 Chimp)<0

WHG admixture more related to European WHGs(Loschbour, Bichon) in EEF than in CHG is supported by.
F3(Kotais EEF; WHG)=.1+(Very significantly more drift)
(((EEF WHG; Ust-Ishim Chimp)-(CHG WHG; Ust-Ishim Chimp))+(Mbuti CHG; EuroWHG Chimp))-(Mbuti EEF; EuroWHG Chimp)=0.025+(pretty significant).

Kale
11-22-2015, 02:40 AM
Having the stats to support an idea is nothing without the stats to refute the alternatives.

For example...
(Kostienki WHG/MA1; Dai/Onge Chimp)<0
Or for my own sanity flip it transitively...
(Chimp Dai/Onge: MA1/WHG Kostenki) <0
Only shows common geneflow between Dai/Onge + MA1/WHG
The possibilities of that are...
1) Kostenki has something basal to the Dai/Onge/Ma1/WHG split
2) After the East West split, Dai/Onge contribute to Ma1/WHG
3) After the East West split, Ma1/WHG contribute to Dai/Onge
4) Another branch concurrent to or precedeing the East West split contributes to both Dai/Onge and Ma1/WHG, but not Kostenki

Now the process of elimination...
1 can be reasonably eliminated via...
Chimp Ust-Ishim ; K14 WHG/Dai = 0

Unless I can dig up some more stats (I'm trying now) I don't think 2, 3, or 4 can be eliminated.

nuadha
11-22-2015, 03:23 AM
ANE was diluted in Russia because of backmigration from the West starting in the Bronze age. Yamnaya and EHG have more MA1-affinity than modern Europeans.

why is france so high on that list?

Kurd
11-22-2015, 03:33 AM
An outgroup was used as a source to get an idea which Source 2 would be most similar to Kotias



SOURCE 1
SOURCE 2
TARGET
F3
STD ERROR
Z
SNP


Mbuti
SATSURBILA
KOTIAS
-0.0060
0.0052
-1.17
26687


Mbuti
Yamnaya
KOTIAS
0.0211
0.0047
4.53
58005


Mbuti
RISE_baYam
KOTIAS
0.0227
0.0047
4.844
50201


Mbuti
RISE_baArm
KOTIAS
0.0261
0.0052
5.002
29138


Mbuti
RISE_baSin
KOTIAS
0.0262
0.0050
5.273
43776


Mbuti
RISE_baAndrov
KOTIAS
0.0262
0.0049
5.362
51122


Mbuti
Iceman
KOTIAS
0.0267
0.0050
5.355
39133


Mbuti
Starcevo_EN
KOTIAS
0.0277
0.0069
3.994
12308


Mbuti
RISE_baSca
KOTIAS
0.0289
0.0050
5.83
47894


Mbuti
Samara_HG
KOTIAS
0.0293
0.0058
5.058
24760


Mbuti
Stuttgart
KOTIAS
0.0318
0.0052
6.168
47120


Mbuti
LBK_EN
KOTIAS
0.0319
0.0048
6.606
60810


Mbuti
Motala_HG
KOTIAS
0.0320
0.0050
6.402
56591


Mbuti
Karelia_HG
KOTIAS
0.0321
0.0052
6.151
39876


Mbuti
Loschbour
KOTIAS
0.0355
0.0052
6.798
46124


Mbuti
LaBrana1
KOTIAS
0.0390
0.0055
7.07
38755


Mbuti
Bichon
KOTIAS
0.0410
0.0057
7.185
31103


Mbuti
MA1
KOTIAS
0.0413
0.0057
7.229
29488


Mbuti
Kostenki14
KOTIAS
0.0503
0.0056
9.024
38764

Kurd
11-22-2015, 03:34 AM
SOURCE 1
SOURCE 2
TARGET
F3
STD ERROR
Z
SNP


LBK_EN
SATSURBILA
KOTIAS
-0.0079
0.0049
-1.588
26329


LBK_EN
RISE_baYam
KOTIAS
0.0292
0.0047
6.168
48933


LBK_EN
MA1
KOTIAS
0.0316
0.0056
5.664
28755


LBK_EN
Yamnaya
KOTIAS
0.0321
0.0047
6.776
55992


LBK_EN
RISE_baArm
KOTIAS
0.0361
0.0052
6.897
28668


LBK_EN
Samara_HG
KOTIAS
0.0379
0.0057
6.594
24503


LBK_EN
Karelia_HG
KOTIAS
0.0393
0.0052
7.551
39393


LBK_EN
Kostenki14
KOTIAS
0.0399
0.0053
7.465
37979


LBK_EN
RISE_baAndrov
KOTIAS
0.0400
0.0050
7.935
50013


LBK_EN
RISE_baSin
KOTIAS
0.0405
0.0051
7.968
43022


LBK_EN
RISE_baSca
KOTIAS
0.0459
0.0050
9.262
47208


LBK_EN
Motala_HG
KOTIAS
0.0497
0.0051
9.684
55137


LBK_EN
Bichon
KOTIAS
0.0513
0.0056
9.147
30635


LBK_EN
Loschbour
KOTIAS
0.0525
0.0053
9.93
45533


LBK_EN
LaBrana1
KOTIAS
0.0526
0.0055
9.652
38368


LBK_EN
Iceman
KOTIAS
0.0602
0.0052
11.476
39231


LBK_EN
Stuttgart
KOTIAS
0.0692
0.0055
12.629
47119


LBK_EN
Starcevo_EN
KOTIAS
0.0696
0.0068
10.204
12346

Kurd
11-22-2015, 03:58 AM
Stronger signal of admixture has been highlighted in yellow



SOURCE 1
SOURCE 2
TARGET
F3
STD ERROR
Z
SNP


Dai
KOTIAS
Tajik_Afghan
-0.0183
0.0012
-15.93
73681


Dai
KOTIAS
.Kurd_SE
-0.0147
0.0031
-4.683
41911


Dai
KOTIAS
.Sein
-0.0143
0.0032
-4.473
46193


Dai
KOTIAS
.Bol_Nat
-0.0125
0.0032
-3.946
45731


Dai
KOTIAS
.Dluffy
-0.0119
0.0033
-3.666
41502


Dai
KOTIAS
.Rukha
-0.0113
0.0033
-3.393
41702


Dai
KOTIAS
Burusho
-0.0107
0.0009
-11.433
83129


Dai
KOTIAS
Pathan
-0.0101
0.0010
-10.195
82811


Dai
KOTIAS
Sindhi
-0.0099
0.0010
-10.072
82626


Dai
KOTIAS
.Varun
-0.0098
0.0033
-2.973
46712


Dai
KOTIAS
.Kenji
-0.0093
0.0034
-2.708
41587


Dai
KOTIAS
.Kandhari
-0.0075
0.0036
-2.098
41707


Dai
KOTIAS
.Bored
-0.0069
0.0033
-2.076
41627


Dai
KOTIAS
Tajik_Pomiri
-0.0060
0.0011
-5.207
78010


Dai
KOTIAS
.DMXX
-0.0054
0.0032
-1.686
48597


Dai
KOTIAS
Balochi
-0.0050
0.0010
-4.818
82938


Dai
KOTIAS
.Jesus
-0.0046
0.0033
-1.392
41797


Dai
KOTIAS
.Khana
-0.0042
0.0042
-1.001
41395


Dai
KOTIAS
Adygei
-0.0033
0.0011
-3.057
81928


Dai
KOTIAS
Iranian
-0.0033
0.0011
-2.885
78224


Dai
KOTIAS
Brahui
-0.0029
0.0010
-2.846
82926


Dai
KOTIAS
Makrani
-0.0027
0.0010
-2.588
83464


Dai
KOTIAS
Lezgin
-0.0022
0.0012
-1.909
78449


Dai
KOTIAS
Chechen
-0.0008
0.0012
-0.705
78264


Dai
KOTIAS
.Hanna
-0.0003
0.0034
-0.091
46552


Dai
KOTIAS
Georgian
0.0007
0.0012
0.621
78774


Dai
KOTIAS
.Kurd_C1
0.0015
0.0034
0.456
41689


Dai
KOTIAS
.Kurd_C2
0.0027
0.0035
0.76
41621


Dai
KOTIAS
Armenian
0.0029
0.0012
2.412
79197


Dai
KOTIAS
.Passa
0.0029
0.0034
0.84
41592


Dai
KOTIAS
Kurd_N1
0.0035
0.0037
0.958
48238


Dai
KOTIAS
Kurd_N4
0.0044
0.0034
1.28
48446


Dai
KOTIAS
Bulgarian
0.0062
0.0012
5.068
79554


Dai
KOTIAS
.Pak_Gujjar
0.0075
0.0046
1.611
45898


Dai
KOTIAS
.Arbogan
0.0076
0.0038
2.008
48352


Dai
KOTIAS
Hungarian
0.0087
0.0012
7.534
82833


Dai
KOTIAS
Iranian_Jew
0.0097
0.0013
7.693
77814


Dai
KOTIAS
Iraqi_Jew
0.0103
0.0014
7.608
74614


Dai
KOTIAS
Kurd_N2
0.0136
0.0043
3.171
48250


Dai
KOTIAS
Lithuanian
0.0166
0.0013
12.517
78550


Dai
KOTIAS
Kurd_N3
0.0186
0.0049
3.796
48024


Dai
KOTIAS
.Zephyrous
0.0188
0.0044
4.285
46051


Dai
KOTIAS
Kalash
0.0194
0.0013
15.025
79072


Dai
KOTIAS
Ulchi
0.0227
0.0010
23.327
76213


Dai
KOTIAS
.Farid
0.0227
0.0054
4.182
41204


Dai
KOTIAS
Nganasan
0.0629
0.0016
39.44
69846


Dai
KOTIAS
.Zara
0.0803
0.0093
8.643
40686

Hanna
11-22-2015, 05:54 AM
This set is done with 2 outgroups to get an idea of absolute position vs Kotias.


Who are Kotias?

kenji.aryan
11-22-2015, 09:36 AM
Going off topic.
Did you know "Kotia" is also a village located in Kanina tehsil, district Mahendragarh, Haryana.

paulgill
11-22-2015, 10:36 AM
Going off topic.
Did you know "Kotia" is also a village located in Kanina tehsil, district Mahendragarh, Haryana.

Hahahaha.............................

bored
11-22-2015, 10:50 AM
Going off topic.
Did you know "Kotia" is also a village located in Kanina tehsil, district Mahendragarh, Haryana.

Moreover it's quite interesting that the people of Haryana have called themselves Kotias since time immemorial.


Jk

Hanna
11-22-2015, 01:00 PM
Going off topic.
Did you know "Kotia" is also a village located in Kanina tehsil, district Mahendragarh, Haryana. ''Also'' but who are/is the former?
Excuse my ignorance.

rozenfeld
11-22-2015, 01:08 PM
''Also'' but who are/is the former?
Excuse my ignorance.

Kotias is the name given to mesolithic hunter-gatherer from Georgia, whose remains were analyzed. The name is given after the cave, Kotias Kade, where remains were found.

Padre Organtino
11-22-2015, 01:37 PM
Kotias is the name given to mesolithic hunter-gatherer from Georgia, whose remains were analyzed. The name is given after the cave, Kotias Kade, where remains were found.

It's Kotias Klde. It is translated as Kotia's Cliff.

alan
11-22-2015, 01:56 PM
Posted this on another thread but belongs here equally:

For those thinking that CHG is some wide signal stretching from the Caucasus all the way along the Stans of inner Asia, this does not seem likely archaeologically - certainly not in the Palaeolithic. There is a huge distinction between the origins of the Palaeolithic groups. For example the Georgian and Armenian areas clearly have a Gravettian flavour and indeed the Caucasus does have a very early Gravettian presence c 3800BC that pre-dates any European Gravettian by a very long time. As far as I know the Zarzians of the Zagros are seen as being of ultimately local Aurignacian origin.

East of the Zagros and certainly east of Iran there is no Gravettian or Aurignacian. From at least the more eastern Stans, through Altai to Baikal, Mongolia and NW China the cultures are very different from those in Europe and anywhere is SW or south-central Asia from at least 43000BC to 10000BC. The last commonality between SW Asia and that area I just described to the east probably is as far back as the Emiran c. 45000BC. It then had a completely different trajectory from the rest of Eurasia for the next 35000 years because there was no Aurignacian or Gravettian waves into those areas. Archaeology would tend to suggest the Caucasus should be not too far from Aurignacian basal but marginally closer to Europe through the shared Gravettian Ahmarian links and somewhat more distant from Siberia. However the separation of the Siberian early upper palaeolithic, the Aurignacian and the very early Caucasus proto-Gravettian all happened about 45-38000BC so they may have not been hugely distinct way back then until separation took its toll.

Also most of the area if you draw a line as suggested above from the top and bottom of the Caspian and project eastwards towards Altai etc was a cold desert during the LGM and indeed there is very little evidence for settlement for a long period after. So its hard to see any commonality stretching from the Caucasus through the Stans in the Palaeolithic and anyway.

I think you only see contact (apparently moving in an east to west direction) between those zones again in the post-Younger Dryas phase (basically the terminal palaeolithic/start of Mesolithic) when you see pressure flaked microblades spreading west.

As the older CHG pre-dates this, I cannot see how there can be a widespread CHG group spreading way to the east of the Caucasus.

alan
11-22-2015, 02:07 PM
This tree makes the most sense to me. Click to get a bigger image.
http://www.anthrogenica.com/attachment.php?attachmentid=6679&stc=1

Mystery Eurasian contributing to proto-MA1 and proto-Dai is supported by.
(Kostienki WHG/MA1; Dai/Onge Chimp)<0
(Kostienki WHG/MA1; Ust-Ishim Chimp)=0

Kostienki and proto-MA1 in same Western branch is supported by
(Ust-Ishim/East Asian MA1; Kostenki Chimp)<0
(Ust-Ishim/East Asian Kostenki; MA1 Chimp)<0

proto-MA1 and proto-Kostienki admixture in WHG is supported by.
(Kostienki WHG; MA1 Chimp)<0
(MA1 WHG; Kostienki Chimp)<0

Basal Eurasian admixture in CHG and EEF is supported by.
(CHG/EEF MA1; Ust-Ishim Chimp)<0
(CHG/EEF Kostenki; Ust-Ishim Chimp)<0

WHG admixture in CHG and EEF is supported by.
(MA1 WHG; CHG/EEF Chimp)<0
(Kostenki WHG; CHG/EEF Chimp)<0

MA1-related admixture in CHG is supported by.
((EEF WHG; Ust-Ishim Chimp)-(CHG WHG; Ust-Ishim Chimp))+(EEF CHG; MA1 Chimp)<0
(WHG EHG; CHG Chimp)<0
(MA1 WHG; CHG Chimp)<(MA1 WHG; Armenian Chimp)<(MA1 WHG; EEF Chimp)
(EEF CHG; MA1 Chimp)<0

WHG admixture more related to European WHGs(Loschbour, Bichon) in EEF than in CHG is supported by.
F3(Kotais EEF; WHG)=.1+(Very significantly more drift)
(((EEF WHG; Ust-Ishim Chimp)-(CHG WHG; Ust-Ishim Chimp))+(Mbuti CHG; EuroWHG Chimp))-(Mbuti EEF; EuroWHG Chimp)=0.025+(pretty significant).

looking at the archaeology, the Caucasus has a unique very early proto-Gravettian which broke off from the Ahmerian in the Levant c. 38000BC a couple of thousand years after the Aurignacian but around ten thousand years before the normal Gravettian of Europe. Note though that they all are independent waves from the same Levantine/Anatolian Ahmerian culture - the main difference being that the three waves are spread out over the period c. 40000-30000BC so the genetics of the source Ahmarian population in the Levant area will not have remained static.
My guess is the source Ahmarian (and indeed its Emeran predecessor) population in the Levant 45000-40000BC were basal and this source population started on a journey from basal to WHG-like in the period 40000-30000BC. The Caucasus Gravettian founders fell about 38000BC so in theory it would seem to me they should be essentially basal but marginally shifted in the direction towards WHG (which in its full form did not yet exist at the time of the migration to the Caucasus).

Kurd
11-22-2015, 02:31 PM
Posted this on another thread but belongs here equally:

For those thinking that CHG is some wide signal stretching from the Caucasus all the way along the Stans of inner Asia, this does not seem likely archaeologically - certainly not in the Palaeolithic. There is a huge distinction between the origins of the Palaeolithic groups. For example the Georgian and Armenian areas clearly have a Gravettian flavour and indeed the Caucasus does have a very early Gravettian presence c 3800BC that pre-dates any European Gravettian by a very long time. As far as I know the Zarzians of the Zagros are seen as being of ultimately local Aurignacian origin.

East of the Zagros and certainly east of Iran there is no Gravettian or Aurignacian. From at least the more eastern Stans, through Altai to Baikal, Mongolia and NW China the cultures are very different from those in Europe and anywhere is SW or south-central Asia from at least 43000BC to 10000BC. The last commonality between SW Asia and that area I just described to the east probably is as far back as the Emiran c. 45000BC. It then had a completely different trajectory from the rest of Eurasia for the next 35000 years because there was no Aurignacian or Gravettian waves into those areas. Archaeology would tend to suggest the Caucasus should be not too far from Aurignacian basal but marginally closer to Europe through the shared Gravettian Ahmarian links and somewhat more distant from Siberia. However the separation of the Siberian early upper palaeolithic, the Aurignacian and the very early Caucasus proto-Gravettian all happened about 45-38000BC so they may have not been hugely distinct way back then until separation took its toll.

Also most of the area if you draw a line as suggested above from the top and bottom of the Caspian and project eastwards towards Altai etc was a cold desert during the LGM and indeed there is very little evidence for settlement for a long period after. So its hard to see any commonality stretching from the Caucasus through the Stans in the Palaeolithic and anyway.

I think you only see contact (apparently moving in an east to west direction) between those zones again in the post-Younger Dryas phase (basically the terminal palaeolithic/start of Mesolithic) when you see pressure flaked microblades spreading west.

As the older CHG pre-dates this, I cannot see how there can be a widespread CHG group spreading way to the east of the Caucasus.

Agreed, the only contribution of Satsurbila to the autosomal structure of SC Asians such as the Baloch, appears to be that he is ancestral to caucasian groups, such as Baloch that moved to SC Asia during the Neolithic and Bronze ages

parasar
11-22-2015, 04:52 PM
This tree makes the most sense to me. Click to get a bigger image.
http://www.anthrogenica.com/attachment.php?attachmentid=6679&stc=1
...

Such a tree cannot be constructed because the input from MA1 to Satsurblia (CHG) is not significantly different from the input from MA1 to Stuttgart (EEF).
D(Yoruba,MA1;Satsurblia,Stuttgart)= D:-0.0096, Z:-1.42

parasar
11-22-2015, 04:59 PM
Posted this on another thread but belongs here equally:

For those thinking that CHG is some wide signal stretching from the Caucasus all the way along the Stans of inner Asia, this does not seem likely archaeologically - certainly not in the Palaeolithic. There is a huge distinction between the origins of the Palaeolithic groups. For example the Georgian and Armenian areas clearly have a Gravettian flavour and indeed the Caucasus does have a very early Gravettian presence c 3800BC that pre-dates any European Gravettian by a very long time. As far as I know the Zarzians of the Zagros are seen as being of ultimately local Aurignacian origin.

East of the Zagros and certainly east of Iran there is no Gravettian or Aurignacian. From at least the more eastern Stans, through Altai to Baikal, Mongolia and NW China the cultures are very different from those in Europe and anywhere is SW or south-central Asia from at least 43000BC to 10000BC. The last commonality between SW Asia and that area I just described to the east probably is as far back as the Emiran c. 45000BC. It then had a completely different trajectory from the rest of Eurasia for the next 35000 years because there was no Aurignacian or Gravettian waves into those areas. Archaeology would tend to suggest the Caucasus should be not too far from Aurignacian basal but marginally closer to Europe through the shared Gravettian Ahmarian links and somewhat more distant from Siberia. However the separation of the Siberian early upper palaeolithic, the Aurignacian and the very early Caucasus proto-Gravettian all happened about 45-38000BC so they may have not been hugely distinct way back then until separation took its toll.

Also most of the area if you draw a line as suggested above from the top and bottom of the Caspian and project eastwards towards Altai etc was a cold desert during the LGM and indeed there is very little evidence for settlement for a long period after. So its hard to see any commonality stretching from the Caucasus through the Stans in the Palaeolithic and anyway.

I think you only see contact (apparently moving in an east to west direction) between those zones again in the post-Younger Dryas phase (basically the terminal palaeolithic/start of Mesolithic) when you see pressure flaked microblades spreading west.

As the older CHG pre-dates this, I cannot see how there can be a widespread CHG group spreading way to the east of the Caucasus.

But how is this early 38000BC presence relevant to Satsurblia CHG whose ancestors arrived in the Caucasus "after the LGM at 17.9–16.2 ka cal. BP?"

Tag Heuer
11-22-2015, 05:35 PM
Agreed, the only contribution of Satsurbila to the autosomal structure of SC Asians such as the Baloch, appears to be that he is ancestral to caucasian groups, such as Baloch that moved to SC Asia during the Neolithic and Bronze ages

so that means Baloch were Caucasians in the past?

alan
11-22-2015, 05:38 PM
But how is this early 38000BC presence relevant to Satsurblia CHG whose ancestors arrived in the Caucasus "after the LGM at 17.9–16.2 ka cal. BP?"

because the late Palaeolithic in Georgia (and Armenia) from which the DNA was found had a material culture of Gravettian origin. The Caucasus had Gravettian material also in pre-LGM times and since 38000BC. So it seems that even if the Caucasus had abandonment in between they likely sheltered nearby.

Even if the post-LGM Georgian hunters with this Gravettian material were a fresh influx from eastern Europe who came through the Caucasus or along the Black Sea shores during or after the LGM, this doesnt change the fact that Gravettian remains are unknown at any time in the area east of Caucasus in Asia. So, the idea that there was some kind of similar CHG type signal stretching from the Caucasus deep into the Stans of central Asia completely contradicts the archaeology.

I would maybe stretch to a similar signal in the the Zagros (where there was an Aurignacian whose signal was likely similar to the early proto-Gravettian in the Caucasus)/south Caspian shore but east of that you are in a different world in terms of Palaeolithic archaeology where neither Aurignacian or Gravettian ever went and where the a different early wave a few millenia earlier which included Ust Ishm went instead with no follow up waves.

That said, all of these groups who settled SW Asia, north-central Asia and Siberia dispersed from ultimate origins in SW Asia (Levant area really) and they all shared a common root there c. 45000BC. So I suspect they all started off as default basal and only diverged after dispersal into separate autosomal DNA groups over the next 10000 years.

Kurd
11-22-2015, 05:55 PM
so that means Baloch were Caucasians in the past?

They were and are still are significantly caucasian. In ADMIXTURE, they seem to have best preserved the autosomal structure modal to CHG. This corroborates Baloch historian claims that many of their tribes are decended from Kurds.

In fact many Pakistani Baloch have the last name Kurd. An internet search "kurd baloch" or "kurd balochistan" can give you more information

Anabasis
11-22-2015, 06:15 PM
They were and are still are significantly caucasian. In ADMIXTURE, they seem to have best preserved the autosomal structure modal to CHG. This corroborates Baloch historian claims that many of their tribes are decended from Kurds.

In fact many Pakistani Baloch have the last name Kurd. An internet search "kurd baloch" or "kurd balochistan" can give you more information

Well i couldnt get the relationship between the baloch-caucasian interaction and baloch-kurd verbal definations. İts irrelevant with topic as far as kurds are not best proxy for chg in west asia.

Kurd
11-22-2015, 06:22 PM
Well i couldnt get the relationship between the baloch-caucasian interaction and baloch-kurd verbal definations. İts irrelevant with topic as far as kurds are not best proxy for chg in west asia.


What are you so confused about? Its really not that complicated

Gravetto-Danubian
11-22-2015, 07:44 PM
because the late Palaeolithic in Georgia (and Armenia) from which the DNA was found had a material culture of Gravettian origin. The Caucasus had Gravettian material also in pre-LGM times and since 38000BC. So it seems that even if the Caucasus had abandonment in between they likely sheltered nearby.

Even if the post-LGM Georgian hunters with this Gravettian material were a fresh influx from eastern Europe who came through the Caucasus or along the Black Sea shores during or after the LGM, this doesnt change the fact that Gravettian remains are unknown at any time in the area east of Caucasus in Asia. So, the idea that there was some kind of similar CHG type signal stretching from the Caucasus deep into the Stans of central Asia completely contradicts the archaeology.

I would maybe stretch to a similar signal in the the Zagros (where there was an Aurignacian whose signal was likely similar to the early proto-Gravettian in the Caucasus)/south Caspian shore but east of that you are in a different world in terms of Palaeolithic archaeology where neither Aurignacian or Gravettian ever went and where the a different early wave a few millenia earlier which included Ust Ishm went instead with no follow up waves.

That said, all of these groups who settled SW Asia, north-central Asia and Siberia dispersed from ultimate origins in SW Asia (Levant area really) and they all shared a common root there c. 45000BC. So I suspect they all started off as default basal and only diverged after dispersal into separate autosomal DNA groups over the next 10000 years.

Then the "Epigravettian" traits in these Caucasian samples must be "cultural diffusions" or "retentions"; because the stats strongly suggest that CHG refuged in Western Asia somewhere, and had no further contact with Europe.

Arbogan
11-22-2015, 07:51 PM
so that means Baloch were Caucasians in the past?
Rather. Caucasians and south central asians share common ancestry from a satsurblia like population. Which would explain why chechens and Lezgins always have high gedrosia. Iranians have high gedrosia-caucasus for the very same reason. Some satsurblia like group must have split and moved east earlier than expected. As Kotias seems to have much less south central asian affinity than satsurblia. Gedrosia in south central asia basically just signals archaic eastern CHG. Its why caucasus-gedrosia components are almost inseparable on admixture matrices until higher Ks.

I think once we have more skeletons from neolithics cultures, and groups ancestral to South west asians and south indians. We'll get the full picture for west asia and the caucasus. This probably the first important discovery for west asians from archaeo-genetics

MfA
11-22-2015, 08:16 PM
Satsurblia 13.2 ky
Docecad K12b


32.98% Gedrosia
- Siberian
- Northwest_African
- Southeast_Asian
- Atlantic_Med
7.08% North_European
- South_Asian
0.01% East_African
- Southwest_Asian
- East_Asian
59.02% Caucasus
0.91% Sub_Saharan

http://i.imgur.com/VuYBw4j.png


If I just flip Caucasus and Gedrosia results, the new location will be in error range to the Belt caves that Pinhasi team soon to release a paper about.

http://abload.de/img/hutofvo1h.png

Kurd
11-22-2015, 08:23 PM
Well i couldnt get the relationship between the baloch-caucasian interaction and baloch-kurd verbal definations. İts irrelevant with topic as far as kurds are not best proxy for chg in west asia.

Incorrect, Kurds are just a good proxy, if not better for CHG, than other W Asian groups. Here are the latest results from my run at K14 in ADMIXTURE. Kotias, Satsurbila, and Bichon were also used to source TRANSVERSION only allele frequencies.



IND
Caucuses_Hunter_Gatherer


KOTIAS
66.90%


Brahui
66.76%


Makrani
65.89%


Balochi
65.70%


Brahui
65.46%


Brahui
64.85%


Makrani
64.69%


Balochi
64.33%


Brahui
64.14%


Brahui
63.98%


Brahui
63.65%


Balochi
63.33%


Brahui
63.32%


Makrani
63.01%


Brahui
62.70%


Brahui
61.61%


Brahui
60.95%


Balochi
60.91%


Balochi
60.88%


Makrani
60.76%


Balochi
60.53%


Brahui
60.44%


Makrani
60.31%


Makrani
60.18%


Brahui
59.83%


Brahui
59.30%


Makrani
59.01%


Makrani
58.40%


Makrani
58.22%


Brahui
58.22%


Makrani
58.19%


Balochi
57.89%


Makrani
57.41%


Makrani
57.26%


Balochi
57.21%


Brahui
57.05%


Balochi
57.01%


Balochi
56.94%


Balochi
56.72%


Makrani
55.71%


Balochi
55.64%


Brahui
55.48%


Makrani
54.93%


Balochi
54.57%


Brahui
54.38%


Balochi
54.11%


Makrani
54.10%


Brahui
53.94%


Brahui
52.84%


Balochi
52.00%


Makrani
51.99%


Abkhasian
51.31%


Makrani
51.21%


Brahui
50.85%


Makrani
50.51%


Makrani
50.23%


Balochi
50.17%


Balochi
49.52%


Balochi
49.51%

Kurd
11-22-2015, 08:26 PM
Kurds are highlighted green:



IND
Caucuses_Hunter_Gatherer


KOTIAS
66.90%


Kurd_C1
48.78%


Balochi
48.73%


Abkhasian
48.52%


Georgian
48.33%


Georgian
48.08%


Abkhasian
47.93%


Georgian
47.64%


Balochi
47.21%


Georgian
47.19%


Abkhasian
47.15%


Abkhasian
47.08%


Georgian
47.03%


kurd_N
46.91%


Iranian
46.89%


Arbogan
46.87%


Lezgin
46.72%


Tajik_Pomiri
46.69%


Lezgin
46.66%


Georgian
46.63%


Iranian
46.63%


Lezgin
46.31%


Kumyk
46.07%


Pathan
45.91%


Abkhasian
45.76%


Abkhasian
45.65%


Adygei
45.54%


Georgian
45.49%


Pathan
45.29%


kurd_N
45.24%


Iranian_Jew
45.14%


Adygei
44.88%


kurd_N
44.79%


Turkish
44.65%


Iranian
44.37%


Uzbek_Afghan
44.30%


Iraqi_Jew
44.29%


Lezgin
44.24%


Adygei
44.24%


Adygei
44.18%


Makrani
44.02%


Pathan
43.66%


Armenian
43.62%


Georgian
43.58%


Adygei
43.55%


Pathan
43.55%


Pathan
43.45%


Tajik_Pomiri
43.35%


Chechen
43.35%


kurd_N
43.09%


Pathan
42.93%


Georgian_Jew
42.89%


Georgian_Jew
42.89%


Adygei
42.88%


Pathan
42.78%


Abkhasian
42.77%


Adygei
42.75%


Burusho
42.70%


Adygei
42.68%


kurd_N
42.65%


Chechen
42.62%


Armenian
42.61%


Georgian
42.47%


Pathan
42.38%


Iranian_Jew
42.36%


Pashtun_Afghan
42.35%


Pathan
42.32%


Pashtun_Afghan
42.30%


.Rukha
42.28%


.Kurd_C2
42.19%


Iranian
42.13%


Iranian
42.04%


kurd_N
41.98%


Georgian
41.87%


Chechen
41.83%


Kumyk
41.80%


Lezgin
41.76%


Georgian_Jew
41.69%


Iranian_Jew
41.61%


Lezgin
41.48%

Krefter
11-22-2015, 08:29 PM
Such a tree cannot be constructed because the input from MA1 to Satsurblia (CHG) is not significantly different from the input from MA1 to Stuttgart (EEF).
D(Yoruba,MA1;Satsurblia,Stuttgart)= D:-0.0096, Z:-1.42

This is because Stuttgart has more WHG and less Basal Eurasian. Lazardis 2014 explains how there can be MA1 ancestry in West Asia even though MA1 is more related to Europeans.

Kurd
11-22-2015, 08:32 PM
This run is showing good potential for being turned into a Gedmatch calculator:




S_Indian
CHG
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo


CHG
0.08














EHG_Yamnaya
0.14
0.13













W_African
0.14
0.15
0.21












Neolithic_Balkan_Farmer
0.10
0.07
0.14
0.16











Indo_Chinese
0.10
0.14
0.19
0.19
0.15










SE_Asian
0.11
0.14
0.20
0.19
0.16
0.11









Papuan
0.16
0.20
0.25
0.23
0.21
0.20
0.19








E_African
0.11
0.10
0.17
0.08
0.11
0.15
0.16
0.20







Siberian
0.12
0.15
0.19
0.19
0.16
0.12
0.11
0.21
0.16






Amerindian
0.20
0.21
0.26
0.27
0.24
0.21
0.20
0.29
0.24
0.19





WHG
0.11
0.08
0.14
0.17
0.09
0.15
0.16
0.22
0.13
0.16
0.23




Andronovo
0.29
0.27
0.33
0.33
0.28
0.33
0.34
0.39
0.30
0.34
0.41
0.29



Kalash
0.08
0.07
0.14
0.16
0.09
0.14
0.14
0.20
0.12
0.15
0.22
0.10
0.28

Kurd
11-22-2015, 08:35 PM
IND
ID
S_Indian
Caucus_Hunter_Gatherer
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo
Kalash


Balochi
HGDP00086
20.05%
65.70%
0.22%
0.00%
0.25%
3.65%
0.00%
0.25%
0.67%
1.51%
0.02%
6.08%
1.60%
0.00%


Balochi
HGDP00052
19.31%
64.33%
1.39%
0.28%
1.21%
0.00%
2.23%
2.21%
0.00%
1.13%
1.47%
4.80%
1.23%
0.40%


Balochi
HGDP00066
15.94%
63.33%
0.00%
1.37%
0.00%
0.00%
0.00%
0.49%
0.91%
0.90%
1.90%
6.90%
0.00%
8.26%


Balochi
HGDP00074
17.49%
60.91%
0.00%
0.00%
0.91%
0.71%
0.00%
2.08%
3.49%
0.73%
1.65%
7.65%
0.00%
4.38%


Balochi
HGDP00054
19.84%
60.88%
3.87%
1.74%
6.49%
3.48%
0.00%
0.13%
0.00%
0.00%
0.00%
3.57%
0.00%
0.00%


Balochi
HGDP00088
11.84%
60.53%
1.67%
0.00%
1.19%
5.64%
0.00%
0.99%
4.41%
0.19%
1.77%
4.44%
1.71%
5.62%


Balochi
HGDP00080
19.71%
57.89%
3.41%
1.10%
5.28%
0.00%
0.00%
0.00%
3.49%
1.28%
1.67%
3.03%
2.18%
0.95%


Balochi
HGDP00090
25.47%
57.21%
2.57%
2.25%
0.00%
0.00%
2.02%
0.00%
0.00%
0.89%
0.00%
3.99%
0.00%
5.58%


Balochi
HGDP00062
15.30%
57.01%
4.12%
3.19%
9.33%
3.07%
0.00%
2.47%
0.49%
0.75%
0.00%
1.63%
1.84%
0.79%


Balochi
HGDP00082
16.23%
56.94%
2.35%
0.00%
0.68%
1.28%
2.60%
2.56%
2.03%
0.00%
0.66%
9.72%
0.68%
4.29%


Balochi
HGDP00068
21.13%
56.72%
0.68%
0.00%
4.39%
2.32%
4.16%
0.16%
0.00%
0.85%
0.00%
5.37%
1.69%
2.53%


Balochi
HGDP00072
21.56%
55.64%
1.80%
2.54%
0.51%
0.00%
0.69%
0.00%
2.90%
0.00%
2.12%
7.54%
1.70%
2.99%


Balochi
HGDP00064
15.84%
54.57%
1.25%
0.63%
7.99%
2.30%
0.00%
0.37%
3.95%
0.66%
0.00%
3.24%
0.80%
8.39%


Armenian
armenia191
0.00%
43.62%
2.81%
0.00%
39.90%
2.80%
0.00%
0.56%
0.27%
0.00%
1.94%
3.57%
2.34%
2.19%


Armenian
armenia106
1.36%
42.61%
2.88%
0.00%
35.87%
1.20%
0.78%
0.00%
3.03%
0.00%
0.41%
8.81%
2.19%
0.88%


Armenian
armenia86
1.16%
40.76%
2.33%
0.00%
36.73%
0.09%
1.48%
0.44%
2.62%
0.00%
0.00%
6.87%
1.95%
5.57%


Armenian
armenia293
4.65%
40.47%
1.93%
0.00%
41.75%
0.00%
1.36%
0.99%
0.00%
1.63%
0.00%
4.06%
3.14%
0.00%


Armenian
armenia279
0.32%
39.37%
0.00%
0.00%
43.13%
1.04%
0.88%
1.73%
4.64%
0.07%
0.00%
7.80%
1.01%
0.00%


Armenian
armenia91
0.28%
38.73%
0.85%
0.00%
43.89%
2.54%
0.10%
0.73%
2.09%
0.00%
1.66%
1.08%
4.38%
3.67%


Armenian
armenia139
1.21%
37.81%
2.02%
0.00%
45.80%
0.00%
0.09%
2.22%
2.67%
0.00%
1.36%
1.98%
2.25%
2.60%


Armenian
armenia102
1.46%
36.84%
1.83%
1.05%
38.09%
0.00%
0.00%
0.00%
1.89%
2.16%
1.09%
8.72%
2.86%
4.02%


Armenian
armenia162
1.21%
35.09%
1.68%
0.00%
45.34%
0.00%
1.08%
0.30%
1.80%
1.36%
0.00%
6.43%
0.19%
5.51%


Armenian
armenia176
3.98%
33.75%
2.15%
0.31%
44.40%
2.35%
0.00%
0.76%
0.00%
0.00%
0.00%
1.95%
3.48%
6.86%

pegasus
11-22-2015, 08:36 PM
They were and are still are significantly caucasian. In ADMIXTURE, they seem to have best preserved the autosomal structure modal to CHG. This corroborates Baloch historian claims that many of their tribes are decended from Kurds.

In fact many Pakistani Baloch have the last name Kurd. An internet search "kurd baloch" or "kurd balochistan" can give you more information

If that was the case then Northern Iranians and Kurds should have the highest Gerdosian component but they dont. The Brahui score even higher Gerdosian than the Baloch. I think they are an archaic people for the most part.

From the results its obvious Baloch/Brahui are not diaspora Kurds.

DMXX
11-22-2015, 08:37 PM
I think Anabasis' point is that Kurds cannot be a source population for the bulk of the CHG observed in Balochis.

As far as component proportions and preliminary interpretations go, that has to be correct, per the data you've posted. Balochis range from 47.2-65.7%, whereas your Kurdish samples are 42-48.8%.

Additionally, the Balochis lie in the same range as the Brahui and Makrani. So, the only straightforward scenarios here (bar a convoluted set of demographic events skewing component scores with time, which admittedly remains possible) are the following:

1) The NW Iranian input in modern Balochis is very limited and hasn't altered the CHG admixture extent in any meaningful sense compared with their neighbours (assuming Balochis, Brahui and Makrani have had comparable CHG proportions since the Early Neolithic)

2) The NW Iranian input was actually substantial, but the Balochi were once more CHG-rich than present (resulting in a downsizing of proportions after admixture took place)

The only other scenario that would allow for a massive Kurdish-like input in modern Balochis is if the groups ancestral to modern Kurds were once possessed far more CHG, which was reduced following subsequent admixture from other groups (e.g. EEF's).

I don't have a favoured stance on this (maintaining my usual conservatism when it comes to mapping complex ethnogenesis in the absence of obligatory data), so I'll stick with the usual "anything's possible" until the data presents itself with time.

Kurd
11-22-2015, 08:38 PM
IND
ID
S_Indian
Caucus_Hunter_Gatherer
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo
Kalash


Bulgarian
BulgarianE2
0.00%
23.72%
3.83%
0.00%
35.67%
2.38%
0.00%
1.61%
0.70%
0.00%
0.00%
30.98%
0.80%
0.31%


Bulgarian
BulgarianC1
0.00%
23.22%
5.96%
0.00%
35.32%
0.33%
0.00%
0.91%
1.97%
1.60%
0.02%
26.24%
1.57%
2.87%


Bulgarian
BulgarianB4
0.00%
21.73%
6.04%
0.00%
32.52%
2.97%
1.57%
1.49%
0.00%
0.00%
0.04%
27.88%
3.30%
2.45%


Bulgarian
BulgarianF2
0.00%
20.55%
3.80%
0.00%
35.68%
2.77%
0.00%
0.00%
1.90%
1.66%
0.30%
30.48%
0.00%
2.86%


Bulgarian
BulgarianH2
0.00%
20.24%
2.94%
0.00%
40.53%
0.74%
0.00%
1.84%
1.03%
0.00%
0.00%
25.77%
3.62%
3.27%


Bulgarian
BulgarianF1
1.90%
20.24%
5.40%
0.30%
36.89%
0.96%
1.99%
0.00%
0.58%
0.13%
0.21%
28.48%
1.37%
1.56%


Bulgarian
BulgarianA4
0.00%
19.78%
3.62%
0.00%
34.00%
1.00%
0.84%
0.00%
0.00%
1.59%
1.15%
33.94%
1.26%
2.83%


Bulgarian
BulgarianB1
0.00%
19.00%
1.40%
0.00%
39.15%
0.00%
1.85%
0.00%
0.00%
2.41%
0.00%
34.25%
0.70%
1.22%


Bulgarian
BulgarianA1
2.96%
16.59%
5.60%
0.62%
36.29%
0.00%
1.09%
0.54%
0.00%
0.97%
0.59%
30.23%
1.67%
2.85%


Czech
NA15728
0.00%
18.91%
4.94%
0.00%
26.20%
0.00%
0.00%
0.89%
0.00%
0.00%
0.42%
46.92%
1.71%
0.00%


Czech
NA15727
1.81%
17.82%
7.72%
0.00%
28.12%
0.00%
0.00%
0.00%
0.00%
0.49%
1.48%
41.14%
1.41%
0.00%


Czech
NA15731
0.16%
16.63%
2.82%
0.00%
27.12%
0.00%
0.00%
0.88%
0.00%
0.00%
1.81%
46.14%
1.85%
2.57%


Czech
NA15732
2.59%
15.78%
3.41%
0.00%
23.68%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
47.50%
1.61%
5.42%


Czech
NA15730
0.00%
15.24%
3.12%
0.00%
28.37%
0.00%
1.91%
0.00%
0.00%
1.22%
0.80%
45.22%
3.67%
0.44%


Czech
NA15733
0.71%
15.20%
5.24%
0.00%
25.97%
0.00%
0.00%
0.02%
0.45%
0.99%
1.58%
45.79%
1.15%
2.89%


Czech
NA15729
0.00%
13.34%
8.14%
0.00%
30.45%
0.00%
0.00%
1.08%
0.00%
0.00%
0.96%
41.44%
1.75%
2.84%


Czech
NA15726
2.53%
12.77%
4.41%
0.00%
30.14%
0.00%
0.00%
1.02%
0.00%
1.16%
0.00%
43.58%
4.39%
0.00%


Czech
NA15725
0.00%
12.10%
7.02%
0.00%
29.39%
1.09%
0.00%
0.38%
0.00%
0.00%
1.15%
41.76%
3.93%
3.18%


Czech
NA15724
2.12%
10.21%
10.36%
0.00%
27.72%
0.00%
0.58%
0.35%
0.77%
0.00%
0.36%
42.65%
1.64%
3.24%

alan
11-22-2015, 08:38 PM
Then the "Epigravettian" traits in these Caucasian samples must be "cultural diffusions" or "retentions"; because the stats strongly suggest that CHG refuged in Western Asia somewhere, and had no further contact with Europe.

Yeah I think its possible that they were isolated and did shelter in the northern fringes of SW Asia just south of the Caucasus but it couldnt have been anywhere near the Stans because the Gravettian at no time reached that area. So any links between a Palaeolithic epi Gravettian group in the Caucasus or their pre-LGM Gravettian ancestors in the Caucasus on the one hand and the Stans/inner area seem to almost have to:

1. Relate to later phases of flow perhaps in the Neolithic.

2. Be an illusion down to the fact that all European and north Eurasian waves go back to the Emiran of the Levant and moved out in several waves

A. c. 45000BC (Ust Ism's initial upper palaeolithic Siberian culture which ran from the Stans to NW-China/north Mongolia as well as Bohunician in Europe)

B. the Ahmerian development that came out of Emiran in the Levant c. 40000BC (which gave rise to the Aurignacian c. 40000BC, the unique Causcaus proto-Gravettian c. 38000BC, the European Gravettian c. 30000BC)

So most of the peopling of SW Asia and northern Asia goes back to what were essentially branching off of Emiran, first into an early branch c. 45000BC which included Ust Ishm's culture as well as the Bohunician. Then on a second branch the Emiran's Ahmarian development seems to have given three offshoots - Aurginacian (40000BC), Caucasus route Proto Gravettian (38000BC) and European Gravettian via Anatolia (30000BC).

Despite the names, they were coming from the same source in multiple waves. Indeed this would predict that the Ust Ishm should be 5000 years diverged from the Aurigancian while the Caucasus Proto-Gravettian is only 2000 years diverged from Aurignacian but 7000 years diverged from Ust-Ism. In theory the main Gravettian that entered Europe through Anatolia should be the most diverged as it is a late offshoot of later Ahmarian c. 30000BC.

So in reality if the CHG is descended from pre-LGM Gravettians then it should be only 2000 years diverged from Autignacians but 7000 years diverged from Ust Ishm. However, if CHG and its epigravettian background was due to a spill over of European Gravettian in or after the LGM then it should be much closer to WHG than it is to Ust Ishm.

It seems most likely to me that CHG is a surviving isolate of the proto-Gravettians who headed towards the Caucasus c. 38000BC. Some will point to an excavation or two that shown abandoning of sites in the Caucasus in the LGM but it is possible they survived in other sites in the area in a small refuge or two.

Kurd
11-22-2015, 08:40 PM
IND
ID
S_Indian
Caucus_Hunter_Gatherer
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo
Kalash


Iranian
iran2
1.20%
46.89%
6.16%
0.30%
24.61%
1.83%
0.00%
0.85%
2.97%
0.00%
2.06%
3.77%
0.00%
9.36%


Iranian
iran17
2.76%
46.63%
1.14%
1.29%
27.98%
5.05%
0.26%
1.85%
0.56%
0.89%
1.97%
4.54%
0.47%
4.60%


Iranian
iran16
6.20%
44.37%
2.68%
0.00%
25.99%
0.00%
3.01%
2.54%
2.91%
0.00%
1.76%
5.34%
3.74%
1.45%


Iranian
iran20
3.90%
42.13%
2.58%
0.53%
26.59%
3.79%
3.70%
0.00%
3.81%
0.00%
0.36%
5.37%
2.37%
4.86%


Iranian
iran3
6.94%
42.04%
3.20%
2.10%
27.33%
0.00%
0.00%
1.94%
3.32%
3.11%
0.00%
4.30%
0.18%
5.54%


Iranian
iran11
6.69%
41.28%
3.51%
0.00%
27.84%
0.00%
3.06%
2.04%
2.72%
0.00%
1.40%
6.67%
1.44%
3.34%


Iranian
iran14
6.67%
40.22%
3.07%
0.10%
27.07%
0.73%
2.25%
1.38%
0.16%
1.75%
0.63%
5.72%
2.16%
8.09%


Iranian
iran19
2.93%
35.01%
3.50%
0.00%
33.64%
1.10%
2.82%
1.20%
5.72%
0.00%
0.99%
6.06%
2.61%
4.42%


Iranian_Jew
IranianJew1832
0.00%
45.14%
0.52%
0.00%
39.78%
0.59%
3.94%
0.00%
2.97%
1.63%
0.60%
0.00%
1.16%
3.67%


Iranian_Jew
IranianJew1845
5.15%
42.36%
0.00%
0.00%
36.09%
2.16%
0.00%
1.17%
4.55%
1.19%
0.00%
1.54%
1.58%
4.20%


Iranian_Jew
IranianJew1513
0.00%
41.61%
1.89%
0.00%
42.56%
0.67%
1.29%
0.00%
6.30%
0.00%
0.54%
0.00%
1.73%
3.41%


Iranian_Jew
IranianJew1159
2.10%
41.44%
5.89%
0.00%
42.66%
3.10%
0.04%
0.00%
2.70%
0.00%
1.05%
0.00%
1.02%
0.00%


Iranian_Jew
IranianJew1557
4.87%
40.05%
4.59%
0.00%
37.53%
0.00%
2.28%
0.00%
2.66%
0.00%
0.00%
1.53%
3.46%
3.02%


Iranian_Jew
IranianJew1143
0.00%
39.58%
2.20%
0.00%
44.13%
2.51%
0.00%
0.00%
3.66%
0.00%
0.72%
0.00%
1.58%
5.61%


Iranian_Jew
IranianJew1132
2.02%
33.64%
4.39%
0.00%
48.87%
2.28%
0.00%
0.54%
2.63%
0.00%
0.00%
0.76%
2.75%
2.12%


Iranian_Jew
IranianJew1556
0.00%
33.56%
0.00%
0.00%
47.08%
3.83%
0.36%
0.00%
7.24%
0.00%
0.00%
2.53%
2.46%
2.94%


Iranian_Jew
IranianJew1409
3.03%
33.10%
3.36%
0.00%
44.32%
0.00%
2.91%
1.33%
3.56%
0.78%
0.00%
2.92%
3.75%
0.94%


Iraqi_Jew
IraqiJew4061
0.00%
44.29%
0.79%
0.00%
40.75%
1.30%
0.00%
0.82%
7.41%
0.00%
0.00%
3.40%
1.24%
0.00%


Iraqi_Jew
IraqiJew1430
0.48%
38.42%
4.61%
0.00%
45.90%
3.96%
0.00%
0.41%
4.20%
0.00%
0.19%
0.00%
1.58%
0.26%


Iraqi_Jew
IraqiJew4241
1.48%
36.28%
1.60%
0.00%
47.37%
0.00%
1.70%
0.62%
3.14%
0.00%
1.11%
0.00%
3.37%
3.33%


Iraqi_Jew
IraqiJew1417
0.00%
34.99%
4.14%
0.00%
45.86%
2.54%
0.00%
0.16%
6.98%
0.44%
0.00%
1.09%
3.48%
0.32%


Iraqi_Jew
IraqiJew4291
2.75%
33.25%
2.90%
0.66%
52.15%
1.20%
2.26%
0.00%
0.93%
0.40%
0.00%
0.00%
0.00%
3.49%

Kurd
11-22-2015, 08:45 PM
IND
ID
S_Indian
Caucus_Hunter_Gatherer
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo
Kalash


.Kurd_C1
ID001
4.03%
48.78%
0.00%
0.00%
28.08%
2.15%
0.00%
0.53%
4.06%
2.67%
0.00%
5.58%
4.13%
0.00%


.Kurd_C2
ID001
2.82%
42.19%
0.00%
0.74%
30.88%
7.95%
0.00%
0.00%
0.00%
0.33%
0.35%
8.09%
2.09%
4.55%


kurd_N
kurd1101
3.24%
42.65%
2.61%
0.63%
31.85%
0.00%
0.73%
0.00%
2.32%
3.91%
0.27%
5.70%
2.78%
3.30%


kurd_N
kurd1156
3.95%
46.91%
3.62%
0.00%
34.90%
0.00%
0.00%
0.00%
0.22%
3.42%
0.00%
3.83%
1.29%
1.87%


kurd_N
kurd1159
4.30%
45.24%
3.25%
1.30%
28.95%
0.24%
1.10%
0.21%
2.95%
0.00%
0.67%
3.73%
2.17%
5.89%


kurd_N
kurd1160
5.43%
44.79%
2.02%
0.00%
32.77%
0.00%
3.51%
0.26%
1.99%
0.00%
0.08%
3.48%
1.42%
4.25%


kurd_N
kurd1173
1.21%
43.09%
3.44%
0.00%
26.90%
4.82%
0.00%
1.09%
2.84%
0.84%
0.00%
6.39%
3.06%
6.32%


kurd_N
kurd1198
6.21%
41.98%
3.09%
0.00%
34.65%
0.12%
0.00%
2.18%
1.68%
4.57%
0.19%
3.07%
2.25%
0.00%


Pashtun_Afghan
Pashtun2_20Af
17.49%
42.35%
4.96%
0.00%
3.76%
2.30%
3.79%
0.83%
0.00%
3.34%
2.66%
12.10%
1.52%
4.91%


Pashtun_Afghan
Pashtun10_17Af
16.32%
42.30%
3.37%
0.00%
9.78%
0.00%
0.00%
3.15%
0.00%
3.04%
2.89%
11.29%
0.00%
7.86%


Pashtun_Afghan
Pashtun2_8Af
16.81%
38.05%
2.83%
0.00%
8.44%
4.02%
3.05%
0.00%
1.01%
1.18%
1.56%
12.78%
2.36%
7.90%


Pashtun_Afghan
Pashtun2_22Af
12.91%
37.73%
3.46%
0.00%
10.69%
3.42%
5.20%
2.26%
0.00%
0.62%
2.59%
8.97%
3.24%
8.93%


Pashtun_Afghan
Pashtun2_6Af
10.44%
34.07%
5.23%
0.00%
7.58%
3.83%
10.61%
0.05%
0.72%
12.90%
0.89%
5.34%
2.21%
6.13%

Kurd
11-22-2015, 08:48 PM
English
HG00131
0.70%
17.31%
3.17%
0.00%
30.34%
0.00%
0.00%
0.00%
0.00%
1.23%
1.13%
42.98%
0.96%
2.18%


English
HG00128
0.00%
15.48%
2.86%
0.00%
28.73%
0.00%
0.00%
0.00%
0.00%
0.00%
0.05%
43.71%
3.35%
5.81%


English
HG00130
0.00%
15.25%
8.20%
0.00%
32.16%
0.00%
0.00%
1.62%
0.00%
0.33%
0.49%
38.56%
1.45%
1.94%


English
HG00126
1.21%
14.61%
3.96%
0.00%
28.60%
0.00%
0.00%
1.42%
0.00%
0.51%
1.13%
48.41%
0.00%
0.15%


English
HG00232
0.00%
14.59%
6.33%
0.00%
30.95%
0.00%
0.00%
0.00%
0.00%
0.00%
0.29%
39.64%
2.63%
5.57%


English
HG00231
0.00%
14.26%
10.17%
0.00%
34.69%
1.82%
0.00%
0.72%
0.00%
0.00%
1.79%
32.91%
1.61%
2.02%


English
HG00234
0.00%
13.54%
7.18%
0.00%
30.12%
1.94%
0.00%
0.44%
0.00%
0.00%
0.15%
43.61%
0.27%
2.77%


English
HG00233
0.00%
13.48%
3.68%
0.00%
32.93%
0.00%
0.00%
0.85%
0.63%
0.33%
1.03%
41.84%
0.29%
4.93%


English
HG00160
0.00%
13.01%
9.28%
0.00%
27.52%
0.00%
0.00%
0.00%
0.00%
0.00%
1.10%
39.40%
4.75%
4.94%


English
HG00129
0.00%
9.08%
8.50%
0.00%
32.65%
0.90%
0.00%
0.35%
0.00%
0.00%
0.32%
41.68%
1.66%
4.85%


Sardinian
HGDP00672
2.91%
1.54%
1.32%
0.15%
70.91%
0.00%
0.00%
1.37%
0.00%
0.00%
0.72%
19.06%
0.30%
1.73%


Sardinian
HGDP01069
0.00%
0.69%
3.47%
0.00%
69.24%
1.81%
0.00%
0.41%
0.00%
0.00%
0.00%
23.07%
1.31%
0.00%


Sardinian
HGDP00669
2.00%
0.58%
0.86%
0.00%
68.82%
2.11%
0.00%
0.47%
0.00%
0.00%
0.00%
23.74%
0.00%
1.42%


Sardinian
HGDP01070
0.00%
0.10%
0.82%
0.76%
69.08%
2.56%
0.00%
0.80%
0.00%
0.00%
0.00%
25.87%
0.00%
0.00%


Sardinian
HGDP01073
1.29%
0.00%
0.33%
0.79%
70.06%
1.80%
0.00%
0.00%
0.00%
0.00%
0.00%
25.72%
0.00%
0.00%


Sardinian
HGDP00673
3.93%
0.00%
0.53%
0.83%
68.48%
0.00%
0.00%
0.00%
0.56%
0.00%
0.00%
22.97%
0.00%
2.69%


Sardinian
HGDP00674
0.00%
0.00%
1.29%
0.00%
67.14%
1.56%
2.06%
0.00%
0.00%
0.00%
0.00%
25.90%
0.00%
2.03%


Sardinian
HGDP00666
0.22%
0.00%
0.00%
0.00%
71.50%
0.00%
0.00%
1.09%
0.00%
0.35%
0.00%
22.77%
2.11%
1.95%


Sardinian
HGDP01067
0.00%
0.00%
0.00%
0.00%
71.95%
0.00%
0.00%
0.00%
0.00%
1.36%
0.00%
24.86%
0.00%
1.82%


Sardinian
HGDP01075
0.00%
0.00%
0.76%
1.84%
69.29%
0.00%
0.43%
0.89%
0.00%
0.00%
0.00%
25.11%
0.00%
1.66%


Sardinian
HGDP01068
1.68%
0.00%
0.00%
0.00%
72.33%
0.00%
0.72%
0.00%
0.00%
0.98%
0.00%
23.29%
0.00%
0.98%


Sardinian
HGDP01079
0.63%
0.00%
2.52%
0.00%
66.38%
0.00%
0.00%
3.08%
0.00%
0.00%
0.02%
25.32%
2.04%
0.00%


Sardinian
HGDP01077
3.78%
0.00%
1.41%
0.00%
68.44%
0.00%
0.00%
0.33%
0.00%
0.78%
0.25%
23.19%
1.80%
0.00%


Sardinian
HGDP00670
2.18%
0.00%
1.18%
0.00%
71.05%
0.00%
0.20%
0.37%
0.00%
0.00%
0.00%
24.34%
0.67%
0.00%


Sardinian
HGDP00668
1.91%
0.00%
0.00%
0.00%
67.66%
0.00%
1.77%
0.11%
0.00%
1.48%
0.37%
26.15%
0.54%
0.00%


Sardinian
HGDP01071
0.00%
0.00%
0.75%
1.39%
70.90%
0.00%
0.00%
1.83%
0.00%
2.75%
0.00%
21.85%
0.52%
0.00%

alan
11-22-2015, 08:49 PM
because the late Palaeolithic in Georgia (and Armenia) from which the DNA was found had a material culture of Gravettian origin. The Caucasus had Gravettian material also in pre-LGM times and since 38000BC. So it seems that even if the Caucasus had abandonment in between they likely sheltered nearby.

Even if the post-LGM Georgian hunters with this Gravettian material were a fresh influx from eastern Europe who came through the Caucasus or along the Black Sea shores during or after the LGM, this doesnt change the fact that Gravettian remains are unknown at any time in the area east of Caucasus in Asia. So, the idea that there was some kind of similar CHG type signal stretching from the Caucasus deep into the Stans of central Asia completely contradicts the archaeology.

I would maybe stretch to a similar signal in the the Zagros (where there was an Aurignacian whose signal was likely similar to the early proto-Gravettian in the Caucasus)/south Caspian shore but east of that you are in a different world in terms of Palaeolithic archaeology where neither Aurignacian or Gravettian ever went and where the a different early wave a few millenia earlier which included Ust Ishm went instead with no follow up waves.

That said, all of these groups who settled SW Asia, north-central Asia and Siberia dispersed from ultimate origins in SW Asia (Levant area really) and they all shared a common root there c. 45000BC. So I suspect they all started off as default basal and only diverged after dispersal into separate autosomal DNA groups over the next 10000 years.

It seems to me that all the waves going into Europe, SW Asia and north Asia c. 45000-38000BC would have been essentially basal variants. I think this is just the default of modern humans who spilled out in several waves across that area in those years. It took a long period to develop the other groups. I suspect:

1. WHG developed around 30000BC probably in Anatolia.

2. ANE developed in the middle upper palaeolithic of the area between Altai, Trans Baikal, NW China, north Mongolia c. 30000BC after 15000 years of drifting.

Kurd
11-22-2015, 08:49 PM
IND
ID
S_Indian
Caucus_Hunter_Gatherer
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo
Kalash


Turkish
Adana23113
2.29%
44.65%
4.76%
0.00%
30.41%
2.49%
1.93%
0.53%
3.72%
0.00%
2.28%
3.57%
0.85%
2.52%


Turkish
Kayseri23549
2.55%
39.74%
0.82%
0.34%
35.61%
0.99%
6.58%
0.00%
0.13%
3.68%
0.00%
6.45%
3.12%
0.00%


Turkish
Aydin18483
0.00%
37.55%
0.26%
0.00%
25.38%
3.44%
4.56%
0.00%
3.67%
8.85%
0.00%
11.15%
2.50%
2.64%


Turkish
Kayseri24075
0.00%
35.28%
0.49%
0.00%
39.21%
0.40%
4.45%
0.37%
2.67%
0.35%
1.99%
9.54%
2.85%
2.40%


Turkish
Kayseri24392
0.00%
34.41%
1.50%
0.00%
37.50%
5.90%
1.38%
0.91%
3.90%
0.89%
0.00%
8.87%
1.13%
3.61%


Turkish
Adana23117
1.79%
34.31%
1.73%
0.00%
34.09%
3.07%
1.08%
0.94%
5.44%
4.61%
1.20%
9.24%
0.07%
2.43%


Turkish
Adana23133
3.34%
34.11%
1.48%
0.00%
37.96%
1.61%
0.77%
0.11%
5.32%
2.97%
0.82%
9.04%
2.37%
0.12%


Turkish
Kayseri23271
2.41%
33.98%
4.73%
0.00%
39.82%
0.00%
3.97%
1.46%
2.35%
1.00%
0.00%
6.32%
1.43%
2.53%


Turkish
Adana23108
0.00%
33.88%
0.00%
0.42%
47.96%
0.00%
1.42%
0.00%
8.39%
0.00%
0.00%
4.26%
1.93%
1.73%


Turkish
Balikesir16653
0.63%
33.69%
0.56%
0.00%
30.26%
1.50%
7.60%
2.47%
1.53%
6.25%
0.63%
11.77%
2.93%
0.17%


Turkish
Adana23150
1.13%
33.04%
2.24%
0.00%
31.81%
0.83%
0.08%
0.60%
3.82%
4.72%
0.09%
15.36%
3.62%
2.67%


Turkish
Kayseri24032
1.18%
32.67%
2.88%
0.00%
35.45%
2.35%
0.51%
0.76%
4.19%
4.19%
0.76%
4.41%
4.09%
6.55%


Turkish
Kayseri24266
6.25%
32.48%
2.59%
0.00%
37.42%
1.99%
0.00%
0.00%
2.18%
3.47%
1.68%
9.18%
2.42%
0.33%


Turkish
Adana23144
4.75%
32.48%
1.89%
2.49%
26.86%
4.30%
1.68%
1.01%
0.00%
7.05%
0.95%
9.84%
1.80%
4.90%


Turkish
Adana23114
0.03%
32.38%
5.08%
0.00%
37.39%
3.47%
0.00%
0.71%
1.94%
1.73%
1.30%
8.05%
4.29%
3.64%


Turkish
Kayseri23967
2.45%
31.56%
0.00%
0.00%
38.20%
0.00%
3.49%
2.00%
0.00%
4.26%
1.23%
7.98%
4.83%
4.00%


Turkish
Aydin18636
0.01%
31.01%
1.74%
0.00%
32.46%
4.30%
2.78%
0.00%
4.32%
5.92%
1.19%
12.35%
2.11%
1.81%


Turkish
Balikesir16837
1.39%
30.64%
2.08%
0.00%
29.97%
1.94%
5.35%
2.76%
0.00%
6.59%
0.00%
17.02%
2.26%
0.00%


Turkish
Aydin18596
0.00%
29.83%
3.24%
0.00%
29.34%
7.25%
4.52%
0.00%
2.92%
6.07%
0.60%
8.67%
3.08%
4.48%


Turkish
Kayseri23892
2.74%
29.77%
4.85%
0.00%
38.52%
1.06%
5.23%
1.37%
2.32%
3.07%
0.60%
7.31%
3.16%
0.00%


Turkish
Adana23136
0.32%
29.35%
3.16%
0.00%
43.06%
0.04%
0.00%
0.44%
7.12%
0.00%
0.00%
8.10%
2.93%
5.48%


Turkish
Balikesir17006
1.53%
29.16%
0.00%
0.00%
31.06%
1.21%
4.30%
2.50%
0.72%
7.96%
1.20%
14.16%
3.70%
2.49%


Turkish
Adana23147
1.54%
29.16%
1.53%
0.00%
39.54%
2.06%
0.00%
0.27%
9.03%
3.24%
0.40%
6.23%
2.58%
4.41%


Turkish
Kayseri24276
4.05%
28.02%
3.73%
0.00%
42.40%
2.88%
4.09%
0.00%
2.45%
0.79%
0.00%
5.21%
1.06%
5.31%


Turkish
Aydin18112
1.29%
27.91%
5.08%
0.55%
33.53%
0.88%
4.38%
1.43%
3.79%
5.74%
0.85%
7.82%
2.75%
4.01%

Anabasis
11-22-2015, 08:51 PM
I think Anabasis' point is that Kurds cannot be a source population for the bulk of the CHG observed in Balochis.

As far as component proportions and preliminary interpretations go, that has to be correct, per the data you've posted. Balochis range from 47.2-65.7%, whereas your Kurdish samples are 42-48.8%.

Additionally, the Balochis lie in the same range as the Brahui and Makrani. So, the only straightforward scenarios here (bar a convoluted set of demographic events skewing component scores with time, which admittedly remains possible) are the following:

1) The NW Iranian input in modern Balochis is very limited and hasn't altered the CHG admixture extent in any meaningful sense compared with their neighbours (assuming Balochis, Brahui and Makrani have had comparable CHG proportions since the Early Neolithic)

2) The NW Iranian input was actually substantial, but the Balochi were once more CHG-rich than present (resulting in a downsizing of proportions after admixture took place)

The only other scenario that would allow for a massive Kurdish-like input in modern Balochis is if the groups ancestral to modern Kurds were once possessed far more CHG, which was reduced following subsequent admixture from other groups (e.g. EEF's).

I don't have a favoured stance on this (maintaining my usual conservatism when it comes to mapping complex ethnogenesis in the absence of obligatory data), so I'll stick with the usual "anything's possible" until the data presents itself with time.

Yes thatwas my point. Onthe other hand modern populations should not be told as source populations for huge component transfer like gedrosia.

Kurd i sent my raw data via mail. Could you rün your calculator for me too?

parasar
11-22-2015, 08:59 PM
This is because Stuttgart has more WHG and less Basal Eurasian. Lazardis 2014 explains how there can be MA1 ancestry in West Asia even though MA1 is more related to Europeans.

I'm not taking about admixture but Dstats. WHG, ANE, and Basal separated so early that a difference in WHG input vs Basal input will not make much of a difference. Lazaridis is talking about modern populations in West Asia that are very different from CHG.

ANE is as close to EEF as to CHG (ancient Georgians).
On the other hand ANE is closer to current Georgians than to EEF.

CHG (ancient Georgians) is as close to WHG as to ANE.
Current Georgians are closer to WHG than to MA1.

So we can see that current Georgians are quite different from CHG. This is because Georgians have ANE, WHG, and EEF, none of which can be said to be present in the CHG.
Supplementary Fig. 2. Inferred topology for ancient and modern populations http://www.nature.com/ncomms/2015/151116/ncomms9912/extref/ncomms9912-s1.pdf

DMXX
11-22-2015, 09:01 PM
So we can see that current Georgians are quite different from CHG. This is because Georgians have ANE, WHG, and EEF, none of which can be said to be present in the CHG.

Is there any evidence that EHG and EEF admixture into the South Caucasus (EHG from north, EEF from south) might have been responsible for these affinity shifts?

Arbogan
11-22-2015, 09:07 PM
IND
ID
S_Indian
Caucus_Hunter_Gatherer
EHG_Yamnaya
W_African
Neolithic_Balkan_Farmer
Indo_Chinese
SE_Asian
Papuan
E_African
Siberian
Amerindian
WHG
Andronovo
Kalash


.Kurd_C1
ID001
4.03%
48.78%
0.00%
0.00%
28.08%
2.15%
0.00%
0.53%
4.06%
2.67%
0.00%
5.58%
4.13%
0.00%


.Kurd_C2
ID001
2.82%
42.19%
0.00%
0.74%
30.88%
7.95%
0.00%
0.00%
0.00%
0.33%
0.35%
8.09%
2.09%
4.55%


kurd_N
kurd1101
3.24%
42.65%
2.61%
0.63%
31.85%
0.00%
0.73%
0.00%
2.32%
3.91%
0.27%
5.70%
2.78%
3.30%


kurd_N
kurd1156
3.95%
46.91%
3.62%
0.00%
34.90%
0.00%
0.00%
0.00%
0.22%
3.42%
0.00%
3.83%
1.29%
1.87%


kurd_N
kurd1159
4.30%
45.24%
3.25%
1.30%
28.95%
0.24%
1.10%
0.21%
2.95%
0.00%
0.67%
3.73%
2.17%
5.89%


kurd_N
kurd1160
5.43%
44.79%
2.02%
0.00%
32.77%
0.00%
3.51%
0.26%
1.99%
0.00%
0.08%
3.48%
1.42%
4.25%


kurd_N
kurd1173
1.21%
43.09%
3.44%
0.00%
26.90%
4.82%
0.00%
1.09%
2.84%
0.84%
0.00%
6.39%
3.06%
6.32%


kurd_N
kurd1198
6.21%
41.98%
3.09%
0.00%
34.65%
0.12%
0.00%
2.18%
1.68%
4.57%
0.19%
3.07%
2.25%
0.00%


Pashtun_Afghan
Pashtun2_20Af
17.49%
42.35%
4.96%
0.00%
3.76%
2.30%
3.79%
0.83%
0.00%
3.34%
2.66%
12.10%
1.52%
4.91%


Pashtun_Afghan
Pashtun10_17Af
16.32%
42.30%
3.37%
0.00%
9.78%
0.00%
0.00%
3.15%
0.00%
3.04%
2.89%
11.29%
0.00%
7.86%


Pashtun_Afghan
Pashtun2_8Af
16.81%
38.05%
2.83%
0.00%
8.44%
4.02%
3.05%
0.00%
1.01%
1.18%
1.56%
12.78%
2.36%
7.90%


Pashtun_Afghan
Pashtun2_22Af
12.91%
37.73%
3.46%
0.00%
10.69%
3.42%
5.20%
2.26%
0.00%
0.62%
2.59%
8.97%
3.24%
8.93%


Pashtun_Afghan
Pashtun2_6Af
10.44%
34.07%
5.23%
0.00%
7.58%
3.83%
10.61%
0.05%
0.72%
12.90%
0.89%
5.34%
2.21%
6.13%



This looks good, looking forward to it. I just wonder if adding a south-west asian cluster would make it optimal. It looks like neolithic balkan farmer is gobbling up all SW-type ancestry. Otherwise it looks excellent.

edit: Nevermind, I realized that east-african of this is equvalient of SW-asian ancestry.

Tomenable
11-22-2015, 09:23 PM
Currently available Ancient samples of R1b haplogroup (in total 45+ samples) and R1a haplogroup (in total 71+ samples) suggest, that both R1b M269/L23 and R1a M198/M417 could initially increase in numbers (demographic expansion) in the Volga steppe. If not counting R1b-V88 Iberians, the oldest known samples of R1a and R1b come from exactly the same prehistoric peoples and cultures - from EHGs and from Khvalynsk culture (which was the best candidate for the earliest Proto-IE speaking culture, according to Marija Gimbutas).

Roughly from that area, they could later expand territorially, roughly at the same time, but along distinct routes.

I've made two maps showing possible early expansion routes & ranges, based on available samples and their chronology:

http://www.anthrogenica.com/showthread.php?5862-Early-expansion-of-R1a-M198-M417-in-the-light-of-aDNA-evidence

R1a M198/M417:

http://s1.postimg.org/di5tzku0d/Early_expansion_of_R1a_M198.png

http://s1.postimg.org/di5tzku0d/Early_expansion_of_R1a_M198.png

R1b M269/L23:

http://s1.postimg.org/r1lbqot7h/Early_R1b_expansion.png

http://s17.postimg.org/3x1m9tbgt/Early_R1b_M269_L23_expansion.png

Kurd
11-22-2015, 09:30 PM
I think Anabasis' point is that Kurds cannot be a source population for the bulk of the CHG observed in Balochis.

As far as component proportions and preliminary interpretations go, that has to be correct, per the data you've posted. Balochis range from 47.2-65.7%, whereas your Kurdish samples are 42-48.8%.

Additionally, the Balochis lie in the same range as the Brahui and Makrani. So, the only straightforward scenarios here (bar a convoluted set of demographic events skewing component scores with time, which admittedly remains possible) are the following:

1) The NW Iranian input in modern Balochis is very limited and hasn't altered the CHG admixture extent in any meaningful sense compared with their neighbours (assuming Balochis, Brahui and Makrani have had comparable CHG proportions since the Early Neolithic)

2) The NW Iranian input was actually substantial, but the Balochi were once more CHG-rich than present (resulting in a downsizing of proportions after admixture took place)

The only other scenario that would allow for a massive Kurdish-like input in modern Balochis is if the groups ancestral to modern Kurds were once possessed far more CHG, which was reduced following subsequent admixture from other groups (e.g. EEF's).

I don't have a favoured stance on this (maintaining my usual conservatism when it comes to mapping complex ethnogenesis in the absence of obligatory data), so I'll stick with the usual "anything's possible" until the data presents itself with time.

We can definitely rule out a scenario where all CHG left the caucuses for SC Asia, because we find high levels of CHG in W Asians, albeit not quite as high as in Baloch and Brahui. What could explain this. Lets examine the facts at hand:

- Even though Kurds are geographically more distant from the location Kotias was recovered, they have CHG levels as high, if not higher than caucuses populations from that region.
- Baloch historian claims that some Baloch tribes have Kurd ancestry (the historians claim that a few Baloch tribes, forgot the names, also have non Iranic origins in Aleppo)
- A good overlap in Kurd/Baloch uniparental markers
- Linguistic associations between Balochi and Kurdi
- The last name of Kurd being used among the Baloch of Pakistan

My explanation for the current levels, is as you have correctly pointed out, is that CHG levels were higher in Kurds around 1000-2000 years ago, and have since lowered due to mixing with other groups. It seems the mixing rate coupled with the populations that the Baloch did end up mixing with tended to favor a slightly better preservation of the CHG component. Let's also remember that the Baloch tend to be more isolated in the geographical regions they inhabit.

parasar
11-22-2015, 09:35 PM
Is there any evidence that EHG and EEF admixture into the South Caucasus (EHG from north, EEF from south) might have been responsible for these affinity shifts?


It must have happened in the Bronze Age time-frame as an EHG input as well as an EEF one.

"We show that Armenian diversity can be explained by several mixtures of Eurasian populations that occurred between ~3,000 and ~2,000 BCE ... 29% of the Armenian ancestry may originate from an ancestral population best represented by Neolithic Europeans [Treemix shows this coming from the Iceman] ... The most significantly negative f3 statistics are for mixture of populations related to Sardinians and Central Asians"
http://biorxiv.org/content/biorxiv/early/2015/02/18/015396.full.pdf
BaA Data from Haak and Allentoft are not inconsistent with this scenario.

Arbogan
11-22-2015, 09:40 PM
Another question comes to mind, that I find interesting in the light of the new evidence we have. Are South-west asians(Gulf-arabs, semitic speaking desert nomads) descendents of neolithic farmers with east-african type admixture, or an entirely seperate but related west-eurasian population?

Tomenable
11-22-2015, 10:08 PM
Many people pursue a "South of Caucasus" hypothesis for the origin of either just R1b M269 or both M269 and R1a M198.

Considering that both M269 and M198 formed ca. 14 - 13 thousand years ago, I think it is probable that either one or both of them originally came from south of the Caucasus. But they most likely came as hunters, not as farmers (because it was too early for farming). And I think that L23 and M417 arose already in the steppe, most likely within the Samara & Khvalynsk cultures. There is by now rather solid evidence that demographic expansion of those lineages took place already in the steppe, and we also found aDNA samples of both R1b and R1a at first in EHGs, then in Khvalynsk culture. I think the "Southern Hypothesis" pursued by some people - which seems to be their substitute for the old Anatolian Hypothesis (which, by now, seems to be totally dead) - can't be sustained in the light of new findings.

Another nail to the coffin of this "Southern Hypothesis" was the discovery that "Teal people" could in fact be... hunters from Caucasus.

Every supporter of the "Southern Hypothesis" was expecting them to be already farmers or herders, not still hunter-gatherers. The notion that Proto-IE language first evolved south of the Caucasus, then went to the steppe is impossible to claim if we assume that the migration from the south took place in Mesolithic times. Because we know from age estimates by most of linguists, that PIE language is not so old.

So even if some R1b or R1a came from the Middle East to the north in Mesolithic times, they were not yet PIE-speakers.

It seems increasingly more probable, that Marija Gimbutas was right not just in general outline, but also in many details - it seems that Samara culture (the first guys ever who domesticated horses) and Khvalynsk culture were the earliest PIE speakers.

Gravetto-Danubian
11-22-2015, 10:12 PM
Yeah I think its possible that they were isolated and did shelter in the northern fringes of SW Asia just south of the Caucasus but it couldnt have been anywhere near the Stans because the Gravettian at no time reached that area. So any links between a Palaeolithic epi Gravettian group in the Caucasus or their pre-LGM Gravettian ancestors in the Caucasus on the one hand and the Stans/inner area seem to almost have to:

1. Relate to later phases of flow perhaps in the Neolithic.

2. Be an illusion down to the fact that all European and north Eurasian waves go back to the Emiran of the Levant and moved out in several waves

A. c. 45000BC (Ust Ism's initial upper palaeolithic Siberian culture which ran from the Stans to NW-China/north Mongolia as well as Bohunician in Europe)

B. the Ahmerian development that came out of Emiran in the Levant c. 40000BC (which gave rise to the Aurignacian c. 40000BC, the unique Causcaus proto-Gravettian c. 38000BC, the European Gravettian c. 30000BC)

So most of the peopling of SW Asia and northern Asia goes back to what were essentially branching off of Emiran, first into an early branch c. 45000BC which included Ust Ishm's culture as well as the Bohunician. Then on a second branch the Emiran's Ahmarian development seems to have given three offshoots - Aurginacian (40000BC), Caucasus route Proto Gravettian (38000BC) and European Gravettian via Anatolia (30000BC).

Despite the names, they were coming from the same source in multiple waves. Indeed this would predict that the Ust Ishm should be 5000 years diverged from the Aurigancian while the Caucasus Proto-Gravettian is only 2000 years diverged from Aurignacian but 7000 years diverged from Ust-Ism. In theory the main Gravettian that entered Europe through Anatolia should be the most diverged as it is a late offshoot of later Ahmarian c. 30000BC.

So in reality if the CHG is descended from pre-LGM Gravettians then it should be only 2000 years diverged from Autignacians but 7000 years diverged from Ust Ishm. However, if CHG and its epigravettian background was due to a spill over of European Gravettian in or after the LGM then it should be much closer to WHG than it is to Ust Ishm.

It seems most likely to me that CHG is a surviving isolate of the proto-Gravettians who headed towards the Caucasus c. 38000BC. Some will point to an excavation or two that shown abandoning of sites in the Caucasus in the LGM but it is possible they survived in other sites in the area in a small refuge or two.

I don't disagree with much here.

Some specific points, though:

* I have heard some theorising indeed that the pre-Glacial "Gravettian proper" industry might have diffused into central Europe via the Caucasus -> north Black Sea -> Austria and Bulgaria path. This might explain some of the "shared retentions" evident between post-glacial Caucasus and eastern Europe (despite lack of evident genetic contact)

* I cannot claim to yet understand the complexity of Eurasian dispersal waves as a whole. The 'black hole' in aDNA from the Middle East and India prohibits any claims of coherent theories.

* In a later post you suggest "WHG developed around 30000BC probably in Anatolia." My view is different. I'd bet Anatolia and Europe c. 30 kya would still show a basaloid, Kostenki-eque profile (pardon the whimsical wording, but you get my drift). I think what really set the formation of WHG in motion is the LGM itself, which imparted massive drift due to separation of west Eurasian populations. At least this front can be solved: all we need is some (one, even) aDNA from 30 - 20 kya from central Europe and western Russia. So I hope the promises of the skeletal samples from Dolní Věstonice (Moravia, c. 26 ky BP; being analysed by Krause) and Sunghir (Russia, c. 28 ky BP, being re-analysed by Ron Pinhasi)come to fruition in imminent future !

Gravetto-Danubian
11-22-2015, 10:17 PM
Currently available Ancient samples of R1b haplogroup (in total 45+ samples) and R1a haplogroup (in total 71+ samples) suggest, that both R1b M269/L23 and R1a M198/M417 could initially increase in numbers (demographic expansion) in the Volga steppe. If not counting R1b-V88 Iberians, the oldest known samples of R1a and R1b come from exactly the same prehistoric peoples and cultures - from EHGs and from Khvalynsk culture (which was the best candidate for the earliest Proto-IE speaking culture, according to Marija Gimbutas).

Roughly from that area, they could later expand territorially, roughly at the same time, but along distinct routes.

I've made two maps showing possible early expansion routes & ranges, based on available samples and their chronology:

http://www.anthrogenica.com/showthread.php?5862-Early-expansion-of-R1a-M198-M417-in-the-light-of-aDNA-evidence

R1a M198/M417:

http://s1.postimg.org/di5tzku0d/Early_expansion_of_R1a_M198.png

http://s1.postimg.org/di5tzku0d/Early_expansion_of_R1a_M198.png

R1b M269/L23:

http://s1.postimg.org/r1lbqot7h/Early_R1b_expansion.png

http://s17.postimg.org/3x1m9tbgt/Early_R1b_M269_L23_expansion.png

Nice maps
But I'm not sure Id share your confidence that M269 arose in Samara valley.
I think you're putting your eggs all in one basket based on the skewed selection of aDNA thus far.

You say:
* "But they (M269) most likely came as hunters, not as farmers (because it was too early for farming)."

-> how about as the pastoralists which brought the new Eneolithic economy to the steppe and north Caucasus c. 4200 BC (prior to this they were forager -fisher-hunters") ?

* "And I think that L23 and M417 arose already in the steppe, most likely within the Samara & Khvalynsk cultures. "

Well, Samara was x L23, so No. We are yet to see what specific subclades Khvalynsk were

Tomenable
11-22-2015, 10:20 PM
But we don't really need any immigration from the south to explain the presence of either R1a or R1b in the steppe.

Both CHG males, who are presumed to be "Teal people", had haplogroup J - not R1.

On the other hand, both R1a and R1b were present among EHGs, already before they acquired "Teal admixture".

The lack of R1 in CHGs seems to confirm what has been suggested time and again before - namely, that their "Teal admixture" perhaps came exclusively from women.

In wartime, prehistoric people used to capture enemy women. After conquering enemy settlements, they used to kill all men, but to take possession of their women. So a group of EHGs could either raid a group of CHGs (capturing a lot of their women in the process), or they could even entirely conquer a group of CHGs, thus acquiring their autosomal DNA, but not their Y-DNA.

There is of course also a more peaceful possibility - namely, exchanging brides between EHGs and "Teals" or CHGs, which led to the dillution of "Teal admixture" in EHGs.

Tomenable
11-22-2015, 10:28 PM
-> how about as the pastoralists which brought the new Eneolithic economy to the steppe and north Caucasus c. 4200 BC

IMO people who were local hunter-gatherers in the steppe, did domesticate local horses (Samara culture) on their own.

Nobody brought horses to the steppe, it was their natural environment. And when it comes to other domestic animals - they could come together with women, who were kidnapped by "to-become-pastoralists" steppe hunters in their raids against farming villages.

Really, know-how can spread not just through "we migrate and give you", but also through "they attack us and steal it".


Well, Samara was x L23, so No.

One Samara was xL23, another Samara could be L23. It was not a "one-person-community", but a tribe with many people.

Yes, the earliest evidence of domestication of horses comes from Samara culture.

And what advantages do horses provide (that allow for hunters to conquer farmers) - see the history of Native North Americans during the 1600s and the 1700s, when horse-riding buffalo hunters & gatherers raided and bullied sedentary farmers without horses.

Silesian
11-22-2015, 10:38 PM
..........* "And I think that L23 and M417 arose already in the steppe, most likely within the Samara & Khvalynsk cultures. "

Well, Samara was x L23, so No. We are yet to see what specific subclades Khvalynsk were

There is no right or wrong answer, however if you were to place an x on Tomenable's map to the general location of the branching - R1b-Z2103 and R1b-L51 where would you place it?
Also R1a-Z283 & R1a-93?
Finally R1a and R1b ?

Padre Organtino
11-22-2015, 10:39 PM
One important point I'd like to make is that most North Caucasians clearly carry some non-negligeable Euro ancestry through maternal side which is evident from their mtdna. So in a hypothetical scenario of bridal exchange we could see the following: two populations interact by exchanging wives. One population size is relatively big (proto NEC/NWC pops) while other is relatively small (EHGs). This will have a much larger impact on a smaller population.

Tomenable
11-22-2015, 10:47 PM
Also let's note that PIEs were strongly patriarchal and polygynous, maternal side of their ancestry did not matter to them:


https://www.youtube.com/watch?v=ErXa5PyHj4I

Gravetto-Danubian
11-22-2015, 10:58 PM
But we don't really need any immigration from the south to explain the presence of either R1a or R1b in the steppe.

yes, we don't *need* anything :) We just have to remain open to possibilities.


Both CHG males, who are presumed to be "Teal people", had haplogroup J - not R1.

correct. Strong evidence that R1 did not come from Epipalaeolithic Caucasus.


On the other hand, both R1a and R1b were present among EHGs, already before they acquired "Teal admixture".

Also correct, but what specific clades were they ?


The lack of R1 in CHGs seems to confirm what has been suggested time and again before - namely, that their "Teal admixture" perhaps came exclusively from women.

Possible. But the math is a bit odd. This would require virtually all steppe men to marry foreign women to account for the ~ 50% autosomal admixture, even if we know that the Teal accrued gradually over several hundred years.

Moreover, we know a fair whack of mtDNA from the steppe was actually of the "EHG' type - up to 50% if we allow that some mtDNA H has been found in Mesolithic Baltic. In any case, I fail to understand such practice, because EE women are bloody hot.


In wartime, prehistoric people used to capture enemy women. After conquering enemy settlements, they used to kill all men, but to take possession of their women. So a group of EHGs could either raid a group of CHGs (capturing a lot of their women in the process), or they could even entirely conquer a group of CHGs, thus acquiring their autosomal DNA, but not their Y-DNA.

Yes; but this sounds like 'stories' rather than a reality of c. 3800 BC, especially given that the steppe communities at that specific period were sparse in number, non-militarized, un-horsed and dependent on the 'higher civilizations' of Tripolye and the Caucasus for their exports and status items. So the supposition is not really born out by the observed archaeological facts. This is not to say they didn't take to occasional opportunistic raiding, but I don't think it really explains what we're seeing.



IMO people who were local hunter-gatherers in the steppe, did domesticate local horses (Samara culture) on their own.

Nobody brought horses to the steppe, it was their natural environment. And when it comes to other domestic animals - they could come together with women, who were kidnapped by "to-become-pastoralists" steppe hunters in their raids against farming villages.[/B] in EHGs.

Not quite. The earliest evidence for horse domestication comes from way to the east - not in anything we'd call Yamnaya or pre-yamnaya - but in Kazakhstan - Botai c. 3800 BC. And in turn, even earlier was in southern Uzbekistan c. 7000 YBP A PROBLEM OF THE EARLIEST HORSE DOMESTICATION. DATA FROM THE NEOLITHIC CAMP AYAKAGYTMA ‘THE SITE’, UZBEKISTAN, CENTRAL ASIA)and these two central Asian areas obviously had contacts. So what went on in Kazakh/ Uzbek can't be transplanted wishfully to areas north of the Black Sea (which some notable Professors have attempted to do- unconvincingly), because the two areas a different culturally, economically, ecologically.

Rather, the economy in incipient Yamnaya areas were fishers, hunters of different game (mostly gazelle, etc), with adoption of agro-pastoralism under the influences of Cucuteni -Tripolye and Caucasian centres- specifically goat pastoralism c. 4200 BC (early Khvalynsk), shifting to cattle c. 3800 BC, shifting to a very mobile form of cattle herding c. 3000 (ie Yamnaya proper period).

Horses did not arrived to the Black Sea steppe in any significant numbers before 3000 BC, so their relevance is a side-issue at best (Khvalynsk 38000 BC had Zero horse bones; fig 4 Multiregional Emergence of Mobile Pastoralism and Nonuniform Institutional Complexityacross Eurasia; Frachetti). What went on after 3000 BC might be a different story.

So we need to be specific and stick to observable facts, not heresay.


And what advantages do horses provide (that allow for hunters to conquer farmers) - see the history of Native North Americans during the 1600s and the 1700s, when horse-riding buffalo hunters & gatherers raided and bullied sedentary farmers without horses.

Yes but the farmers introduced those horses - already tamed - and at least at an individual level showed the Indians how to ride them. In the end, the horse-riding Indians didn't fare to well, unfortunately. So again, lets stick to facts rather than misplaced analogies; as per prev. comment.


One Samara was xL23, another Samara could be L23. It was not a "one-person-community", but a tribe with many people.

Are you talking about Samara HG ?

alan
11-22-2015, 11:28 PM
yes, we don't *need* anything :) We just have to remain open to possibilities.



correct. Strong evidence that R1 did not come from Epipalaeolithic Caucasus.



Also correct, but what specific clades were they ?



Possible. But the maths is a bit odd. This would require virtually all steppe men to marry foreign women to account for the ~ 50% autosomal admixture, even if we know that the Teal accrued gradually over several hundred years.

Moreover, we know a fair whack of mtDNA from the steppe was actually of the "EHG' type - up to 50% if we allow that some mtDNA H has been found in Mesolithic Baltic. In any case, I fail to understand such practice, because EE women are bloody hot.



Yes; but this sounds like 'stories' rather than a reality of c. 3800 BC, especially given that the steppe communities at that specific period were sparse, non-militarized and dependent on the 'higher civilizations' of Tripolye and the Caucasus. So the supposition is not really born out by the observed archaeological facts.




Well, yes, and No. The earliest evidence for horse domestication comes from way to the east - not in anything we'd call Yamnaya or pre-yamnaya - but in Kazakhstan - Botai. And in turn, even earlier was in southern Uzbekistan c. 7000 YBP A PROBLEM OF THE EARLIEST HORSE DOMESTICATION. DATA FROM THE NEOLITHIC CAMP AYAKAGYTMA ‘THE SITE’, UZBEKISTAN, CENTRAL ASIA)and these two central Asian areas obviously had contacts. So what went on in Kazakh/ Uzbek can't be transplanted wishfully to areas north of the Black Sea (which some notable Professors have attempted to do- unconvincingly).

Rather, the economy in incipient Yamnaya areas were fishers, hunters of different game (mostly gazelle, etc), with adoption of agro-pastoralism under the influences of Cucuteni -Tripolye and Caucasian centres- specifically goat pastoralism c. 4200 BC (early Khvalynsk), shifting to cattle c. 3800 BC, shifting to a very mobile form of cattle herding c. 3000 (ie Yamnaya proper period).

Horses did not arrived to the Black Sea steppe in any significant numbers before 3000 BC, so their relevance is a side-issue at best (Khvalynsk 38000 BC had Zero horse bones; fig 4 Multiregional Emergence of Mobile Pastoralism and Nonuniform Institutional Complexityacross Eurasia; Frachetti).

So we need to be specific and stick to observable facts, not Chinese whispers of heresay.



Yes but the farmers introduced those horses - already tamed - and at least at an individual level showed the Indians how to ride them. In the end, the horse-riding Indians didn't fare to well, unfortunately. So again, lets stick to facts rather than misplaced analogies; as per prev. comment.



Are you talking about Samara HG ?

The Teal flow into the steppe doesnt have literally to have been a female only one. A more realistic scenario is it involved both genders but the male superbreeders were steppe guys not Teal ones for some reason. We have great parallels for this in Ireland for example where a few chiefs and kings or especially successful royal dynasties who lived at some point 2000-1000 years ago seem to have supplied the majority of the male lineages around today without creating some sort of weird autosomal effect. Ireland has similar autosomal DNA to the rest of NW Europe despite its history of relative isolation and this seems to have been established long before these superbreeders lived. So it seems the superbreeding males didnt effect the autosomal DNA

Also should always be borne in mind that the burials we find, particularly the Kurgans, probably do not represent a normal sample of society. Indeed in almost all cases it has been shown that burials monuments, even in the Neolithic collective tombs, are selective in terms of age/sex/status and probably represented favoured elements of society. So, what may look like utter dominance may in fact just be dominance of the elite and it may have taken centuries for the yDNA of the elite to displace down to lower social strata.

DMXX
11-22-2015, 11:39 PM
A wayward thought which just occurred to me.

So far, in West Asia, we've picked up two distinct components - EEF and CHG.

EEF, from memory, is best represented by modern-day Sardinians. Several modern population-based ADMIXTURE calculators picked up a "Mediterranean" component that also peaked in Sardinians.

CHG, according to both Jones et al. and Kurd's work, peaks in the Caucasus as well as South-Central Asia. This is precisely where we've observed over the years through the ADMIXTURE "Gedrosian", "teal" or "Caucasus-Gedrosia" components (or at least CHG-like, judging by some of the formal stats).

If this tenuous bridging between old and new is maintained, then my earlier guess at a third (and final) West Asian auDNA component (represented by "SW Asian") might be the final piece of the puzzle for the region. Perhaps some Natufian, Kebaran or Zarzian culture aDNA will offer this?

Based on this, it appears as if the modern population ADMIXTURE runs weren't completely smoke and mirrors, and actually did provide reasonable evidence of prehistoric movements in Eurasia.

Alternatively, some extraneous complications are at work. Occam's Razor applies until opposing data from the region presents itself (hopefully soon).

Gravetto-Danubian
11-22-2015, 11:56 PM
If this tenuous bridging between old and new is maintained, then my earlier guess at a third (and final) West Asian auDNA component (represented by "SW Asian") might be the final piece of the puzzle for the region. Perhaps some Natufian, Kebaran or Zarzian culture aDNA will offer this?
(hopefully soon).

Yes a nice juicy proto-Natufian should seal the deal :)

ZephyrousMandaru
11-22-2015, 11:58 PM
I just have one question, seeing as how CHG is bimodal in both the Caucasus and South-Central Asia, don't we need to uncover ancient genomes from South-Central Asia in order to rule out where CHG originated?

Generalissimo
11-23-2015, 12:01 AM
The Teal flow into the steppe doesnt have literally to have been a female only one. A more realistic scenario is it involved both genders but the male superbreeders were steppe guys not Teal ones for some reason.

Why do you believe this is a more realistic scenario when we actually have evidence for female exogamy and high female mobility in Bronze Age Europe, including in the North Caucasus steppe region?

DMXX
11-23-2015, 12:05 AM
I just have one question, seeing as how CHG is bimodal in both the Caucasus and South-Central Asia, don't we need to uncover ancient genomes from South-Central Asia in order to rule out where CHG originated?

Some of the gentlemen in the Eurogenes comments section have been pouring over some formal stats regarding the exact role of CHG in South-Central Asia (among other questions, including the origin of CHG and its' relation to the hypothetical "Basal Eurasian", or Ryu's "Crown Eurasian").

Their assessment based on the outputs (which one can observe via the Z figures) is that Kotias Klde and Satsurbila didn't directly contribute to modern South-Central Asians.

Based on this, it appears that the West Asian related ancestry that predominates across South-Central Asia should be described as "CHG-like". We'll need aDNA from mesolithic cultures in northern Iran and South-Central Asia, yes.

Kurd
11-23-2015, 12:40 AM
I just have one question, seeing as how CHG is bimodal in both the Caucasus and South-Central Asia, don't we need to uncover ancient genomes from South-Central Asia in order to rule out where CHG originated?

Since the CHG levels are highest in Baloch Brahui followed by W Asian groups such as Abkhasians Adygei Georgians, and Kurds, then followed by Baloch neighbors in SC Asia such as Pashtuns, and it is modal to a mesolithic caucasian (Kotias), CHG appears to be connected to Baloch and their movements, which is a migration to their present locations from W Asia.

Gravetto-Danubian
11-23-2015, 12:51 AM
Since the CHG levels are highest in Baloch Brahui followed by W Asian groups such as Abkhasians Adygei Georgians, and Kurds, then followed by Baloch neighbors in SC Asia such as Pashtuns, and it is modal to a mesolithic caucasian (Kotias), CHG appears to be connected to Baloch and their movements, which is a migration to their present locations from W Asia.

Yes, but as DMXX mentioned above (and I on a different thread)

"
Some of the gentlemen in the Eurogenes comments section have been pouring over some formal stats regarding the exact role of CHG in South-Central Asia (among other questions, including the origin of CHG and its' relation to the hypothetical "Basal Eurasian", or Ryu's "Crown Eurasian").

Their assessment based on the outputs (which one can observe via the Z figures) is that Kotias Klde and Satsurbila didn't directly contribute to modern South-Central Asians.

Based on this, it appears that the West Asian related ancestry that predominates across South-Central Asia should be described as "CHG-like". We'll need aDNA from mesolithic cultures in northern Iran and South-Central Asia, yes. " {DMXX 2015)

Also, the Jone's paper revealed a high rate of consanguinity in the CHG. So perhaps, it is the CHG who drifted north- west (after the climactic amelioration) ? Some remained further east who later --> Steppe.

parasar
11-23-2015, 12:53 AM
Some of the gentlemen in the Eurogenes comments section have been pouring over some formal stats regarding the exact role of CHG in South-Central Asia (among other questions, including the origin of CHG and its' relation to the hypothetical "Basal Eurasian", or Ryu's "Crown Eurasian").

Their assessment based on the outputs (which one can observe via the Z figures) is that Kotias Klde and Satsurbila didn't directly contribute to modern South-Central Asians.

Based on this, it appears that the West Asian related ancestry that predominates across South-Central Asia should be described as "CHG-like". We'll need aDNA from mesolithic cultures in northern Iran and South-Central Asia, yes.

It appears to me that there are three distinct CHG-likes at least, with Y-G modal [EEF], Y-J modal [CHG] and Y-H modal separated by the LGM. Of course there will be overlaps, and we see that the Neolithic Anatolians had all three + Y-I but not any M526.
Y-L has not yet made an appearance but Y-T has in German LBK. So I think it is just a matter of time that Y-L makes an appearance too. While South Asia has Y-G, Y-J is far more common, so South Asian CHG-like will be far more like CHG than EEF.

Kurd
11-23-2015, 12:58 AM
Some of the gentlemen in the Eurogenes comments section have been pouring over some formal stats regarding the exact role of CHG in South-Central Asia (among other questions, including the origin of CHG and its' relation to the hypothetical "Basal Eurasian", or Ryu's "Crown Eurasian").

Their assessment based on the outputs (which one can observe via the Z figures) is that Kotias Klde and Satsurbila didn't directly contribute to modern South-Central Asians.

Based on this, it appears that the West Asian related ancestry that predominates across South-Central Asia should be described as "CHG-like". We'll need aDNA from mesolithic cultures in northern Iran and South-Central Asia, yes.

FWIW, my assessment as to why caucasians would appear more similar to Kotias than Baloch and other SC Asians in Dstats, is as I have previously mentioned, caucasians' non CHG admixture (Neolithic Baltic Farmer and WHG) has a lower fixation (FST) distance to CHG, than the Baloch non CHG admixture, which has lower amounts of NBF and WHG than caucasians. I have posted an FST matrix of my latest K14 run a few pages back.

DMXX
11-23-2015, 01:02 AM
It appears to me that there are three distinct CHG-likes at least, with Y-G modal [EEF], Y-J modal [CHG] and Y-H modal separated by the LGM. Of course there will be overlaps, and we see that the Neolithic Anatolians had all three + Y-I but not any M526.
Y-L has not yet made an appearance but Y-T has in German LBK. So I think it is just a matter of time that Y-L makes an appearance too. While South Asia has Y-G, Y-J is far more common, so South Asian CHG-like will be far more like CHG than EEF.

Only further aDNA will confirm the Y-DNA defining breakdowns you've kindly expressed, but I do agree with your line of reasoning. I suspect the same holds true with ANE also (there surely has to be at least two distinct clades within it, given the supreme age of Mal'ta).

Pure conjecture on my part, but I've often wondered whether the M173 and M479 (R1 and R2) nodes within R-M207 could also represent the extant post-Mal'ta ANE division (one Eurasian, the other South-Central Asian)...

Tomenable
11-23-2015, 01:06 AM
Possible. But the math is a bit odd. This would require virtually all steppe men to marry foreign women to account for the ~ 50% autosomal admixture, even if we know that the Teal accrued gradually over several hundred years.

That's not improbable over time in a polygynous society in which at least high status males have multiple wifes.

Khvalynsk samples (4700-4000 BC) were 25% Teal, while Yamnaya_Samara (3340-2620 BC) were almost 50% Teal.

We have almost 1400 years separating those two populations (which lived in the same region), and this is really a lot.


Moreover, we know a fair whack of mtDNA from the steppe was actually of the "EHG' type - up to 50% if we allow that some mtDNA H has been found in Mesolithic Baltic. In any case, I fail to understand such practice, because EE women are bloody hot.

If only up to 50% of mtDNA lineages were of EHG types, then this leaves a lot (at least 50%, possibly much more) for "Teal" women.

If before mixing of the two populations, EHGs were less numerous than "Teal people", then autosomal change without Y-DNA change was possible, especially over dozens of generations (between Khvalynsk and Yamnaya there were ~1400 years and already Khvalynsk were 25% Teal). Similar mechanisms worked among ancestors of modern populations which are mostly R1b-M269 but have not much "Steppe" autosomal.

Here is a model which shows something like this happening over just 3 generations, not dozens (imagine that red + orange = EHGs):

If we take a look at this from the point of view of EHGs, who "incorporated" Teal people into their community, then in this model EHGs transformed from being 100% autosomal EHG and 100% Y-DNA EHG, to just 42% autosomal EHG but still 75% Y-DNA EHG after 3 generations:
(and they also vastly increased their numbers in the process, from 4 to 24 - I consider all 24 to be members of the new "EHG-run" tribe)

I assumed no population growth over 3 generations for this model, just replacement fertility, but not all males reproduce:

http://s2.postimg.org/vjgr8gpux/PIE_model.png

http://s2.postimg.org/vjgr8gpux/PIE_model.png


Not quite. The earliest evidence for horse domestication comes from way to the east - not in anything we'd call Yamnaya or pre-yamnaya - but in Kazakhstan - Botai c. 3800 BC. And in turn, even earlier was in southern Uzbekistan c. 7000 YBP A PROBLEM OF THE EARLIEST HORSE DOMESTICATION. DATA FROM THE NEOLITHIC CAMP AYAKAGYTMA ‘THE SITE’, UZBEKISTAN, CENTRAL ASIA)and these two central Asian areas obviously had contacts. So what went on in Kazakh/ Uzbek can't be transplanted wishfully to areas north of the Black Sea (which some notable Professors have attempted to do- unconvincingly), because the two areas a different culturally, economically, ecologically.

Rather, the economy in incipient Yamnaya areas were fishers, hunters of different game (mostly gazelle, etc), with adoption of agro-pastoralism under the influences of Cucuteni -Tripolye and Caucasian centres- specifically goat pastoralism c. 4200 BC (early Khvalynsk), shifting to cattle c. 3800 BC, shifting to a very mobile form of cattle herding c. 3000 (ie Yamnaya proper period).

Horses did not arrived to the Black Sea steppe in any significant numbers before 3000 BC, so their relevance is a side-issue at best (Khvalynsk 38000 BC had Zero horse bones; fig 4 Multiregional Emergence of Mobile Pastoralism and Nonuniform Institutional Complexityacross Eurasia; Frachetti). What went on after 3000 BC might be a different story.

I know wikipedia is not the best choice but I'm in a hurry - it says something like this:

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

"The oldest possible archaeological indicator of a changed relationship between horses and humans is the appearance about 4800-4400 BCE of horse bones and carved images of horses in Chalcolithic graves of the early Khvalynsk culture and the Samara culture in the middle Volga region of Russia. At the Khvalynsk cemetery near the town of Khvalynsk, 158 graves of this period were excavated. Of these, 26 graves contained parts of sacrificed domestic animals, and additional sacrifices occurred in ritual deposits on the original ground surface above the graves. Ten graves contained parts of lower horse legs; two of these also contained the bones of domesticated cattle and sheep. At least 52 domesticated sheep or goats, 23 domesticated cattle, and 11 horses were sacrificed at Khvalynsk. The inclusion of horses with cattle and sheep and the exclusion of obviously wild animals together suggest that horses were categorized symbolically with domesticated animals.

At S’yezzhe, a contemporary cemetery of the Samara culture, parts of two horses were placed above a group of human graves. The pair of horses here was represented by the head and hooves, probably originally attached to hides. The same ritual—using the hide with the head and lower leg bones as a symbol for the whole animal—was used for many domesticated cattle and sheep sacrifices at Khvalynsk. Horse images carved from bone were placed in the above-ground ochre deposit at S’yezzhe and occurred at several other sites of the same period in the middle and lower Volga region. Together these archaeological clues suggest that horses had a symbolic importance in the Khvalynsk and Samara cultures that they had lacked earlier, and that they were associated with humans, domesticated cattle, and domesticated sheep. Thus, the earliest phase in the domestication of the horse might have begun during the period 4800-4400 BCE."

This is much older than evidence from Botai culture.


Are you talking about Samara HG ?

About Samara HG, as well as about Khvalynsk R1b.

Gravetto-Danubian
11-23-2015, 01:30 AM
I know wikipedia is not the best choice but I'm in a hurry - it says something like this:

I definitely agree there. If we are to debate minutii, then you should be using peer-reviewed and recent journal articles - as I have.



"The oldest possible archaeological indicator of a changed relationship between horses and humans is the appearance about 4800-4400 BCE of horse bones and carved images of horses in Chalcolithic graves of the early Khvalynsk culture and the Samara culture in the middle Volga region of Russia. At the Khvalynsk cemetery near the town of Khvalynsk, 158 graves of this period were excavated. Of these, 26 graves contained parts of sacrificed domestic animals, and additional sacrifices occurred in ritual deposits on the original ground surface above the graves. Ten graves contained parts of lower horse legs; two of these also contained the bones of domesticated cattle and sheep. At least 52 domesticated sheep or goats, 23 domesticated cattle, and 11 horses were sacrificed at Khvalynsk. The inclusion of horses with cattle and sheep and the exclusion of obviously wild animals together suggest that horses were categorized symbolically with domesticated animals.

At S’yezzhe, a contemporary cemetery of the Samara culture, parts of two horses were placed above a group of human graves. The pair of horses here was represented by the head and hooves, probably originally attached to hides. The same ritual—using the hide with the head and lower leg bones as a symbol for the whole animal—was used for many domesticated cattle and sheep sacrifices at Khvalynsk. Horse images carved from bone were placed in the above-ground ochre deposit at S’yezzhe and occurred at several other sites of the same period in the middle and lower Volga region. Together these archaeological clues suggest that horses had a symbolic importance in the Khvalynsk and Samara cultures that they had lacked earlier, and that they were associated with humans, domesticated cattle, and domesticated sheep. Thus, the earliest phase in the domestication of the horse might have begun during the period 4800-4400 BCE."

Yes it suggest a symbolic significance, but to determine accurately the nature of horse use, we need osteological analyses, sex- age distribution figures, evidence for bitting, horse aDNA etc - as a bare minimum. At the very least, from those early Khvalynsk bones - which are incomplete and scattered - "there is no correlation between horse bones and other symbols of wealth or ranking in these graves" (Mallory 1991).


That Eneolithic and Early Bronze Age peoples went to the trouble of burying horses attests to their symbolic significance, but it cannot be taken as evidence of domestication. Hares were also found in Yamnaya and Catacomb culture graves, but no one claims, on that basis, that they were domesticated."
- M Levine

Im sure more data will come in the near future, and we can pick up the discussion further then, on the right forum. :)


Btw thanks for your genetic diagram -looks good
But i think Khvalynsk is now dated 4200-3800 BC.

Tomenable
11-23-2015, 01:44 AM
Khvalynsk were 25% Teal, Yamna were 50% Teal, there were 1400 years between them.

So on average Teal was increasing by 1% per each 50-60 years, or 0,5% per generation.

This looks like it could be a steady stream of Teal flowing into the region from somewhere.

Tomenable
11-23-2015, 01:54 AM
The speed of Teal flow into the steppe between Samara HG and Khalynsk was also ca. 0,5% per generation:

1) Samara HG sample (5650-5555 BC) had ~0% Teal admixture,
2) Khvalynsk samples (4700-4000 BC) had ~25% Teal admixture,
3) Yamnaya_Samara (3340-2620 BC) had ~48% Teal admixture,

Between 1) and 2) there were 1250 years, between 2) and 3) some 1370 years.

This means that Teal admixture was increasing by 1% per 50-60 years (0,5% per generation) in both periods.

To me it looks like a steady gene flow, not like a rapid immigration. If it was indeed such a "fluent" process.

========================

Edit:


But i think Khvalynsk is now dated 4200-3800 BC.

In such case the average speed of Teal increase between Khvalynsk and Yamnaya was 1% per 45 years.

Tomenable
11-23-2015, 02:09 AM
By the way, we now treat EHGs as the "receiving end" of Teal admixture.

But what if that gene flow was not one-sided, but mutual (bride exchanging ???).

I wonder, if there was EHG admixture in Caucasus (or wherever "Teal" people lived).

Maybe at the same time when EHG became 48% "Teal", Teal became some % "EHG" ??? :)

jeanL
11-23-2015, 02:24 AM
Here is what seems to be the skull of Bichon:

6682

Source#1 (http://www.rts.ch/info/regions/neuchatel/7258440-un-squelette-neuchatelois-au-coeur-d-une-avancee-sur-l-origine-humaine.html)

6683

Source#2 (http://www.meta-jura.org/fleches/l-arc-jurassien-de-l-peu-pres-histoire-et-prehistoire-1-95.htm)

His morphology appears to be very modern.

J Man
11-23-2015, 02:34 AM
It appears to me that there are three distinct CHG-likes at least, with Y-G modal [EEF], Y-J modal [CHG] and Y-H modal separated by the LGM. Of course there will be overlaps, and we see that the Neolithic Anatolians had all three + Y-I but not any M526.
Y-L has not yet made an appearance but Y-T has in German LBK. So I think it is just a matter of time that Y-L makes an appearance too. While South Asia has Y-G, Y-J is far more common, so South Asian CHG-like will be far more like CHG than EEF.

You said..''three distinct CHG-likes at least''...What exactly do you mean by that?

parasar
11-23-2015, 04:20 AM
You said..''three distinct CHG-likes at least''...What exactly do you mean by that?

J Man,

It has been my thinking for some time that the basal side will flesh out as we get more ancient dna. Basal was only inferred in Stuttgart, while ANE and WHG were represented by actual samples that shared drift with ENA. With K14, again basal was inferred, but at least we came to know that there were multiple basal lines since K14's basal did not share drift with Stuttgart's basal but both met the definition of basal. So when I say at least three CHG-likes, I also mean at least three distinct basal lines.

Gravetto-Danubian
11-23-2015, 04:26 AM
J Man,

It has been my thinking for some time that the basal side will flesh out as we get more ancient dna. Basal was only inferred in Stuttgart, while ANE and WHG were represented by actual samples that shared drift with ENA. With K14, again basal was inferred, but at least we came to know that there were multiple basal lines since K14's basal did not share drift with Stuttgart's basal but both met the definition of basal. So when I say at least three CHG-likes, I also mean at least three distinct basal lines.

Does Kostentki share drift with a CHG like population?

Kurd
11-23-2015, 04:43 AM
Sorted by strongest signal of admixture on top



SOURCE 1
SOURCE 2
TARGET
F3
STD ERROR
Z
SNPs


Yamnaya
KOTIAS
.Kurd_SE
0.0002
0.0032
0.07
40027


Yamnaya
KOTIAS
Lezgin
0.0016
0.0010
1.65
78210


Yamnaya
KOTIAS
.Rukha
0.0027
0.0032
0.83
39764


Yamnaya
KOTIAS
Adygei
0.0040
0.0009
4.69
81822


Yamnaya
KOTIAS
Chechen
0.0046
0.0010
4.79
78028


Yamnaya
KOTIAS
Georgian
0.0054
0.0010
5.28
78523


Yamnaya
KOTIAS
.DMXX
0.0062
0.0032
1.96
46353


Yamnaya
KOTIAS
Tajik_Pomiri
0.0071
0.0010
7.07
77753


Yamnaya
KOTIAS
Iranian
0.0072
0.0010
7.05
77995


Yamnaya
KOTIAS
.Sein
0.0073
0.0033
2.21
44204


Yamnaya
KOTIAS
.Passa
0.0076
0.0035
2.16
39538


Yamnaya
KOTIAS
.Bol_Nat
0.0078
0.0033
2.34
43639


Yamnaya
KOTIAS
.Kandhari
0.0078
0.0038
2.09
39760


Yamnaya
KOTIAS
Tajik_Afghan
0.0083
0.0012
7.08
73342


Yamnaya
KOTIAS
.Dluffy
0.0090
0.0035
2.61
39632


Yamnaya
KOTIAS
Bulgarian
0.0094
0.0010
8.96
79282


Yamnaya
KOTIAS
Armenian
0.0095
0.0010
9.56
78976


Yamnaya
KOTIAS
.Jesus
0.0095
0.0033
2.87
39921


Yamnaya
KOTIAS
.Kurd_C1
0.0100
0.0035
2.87
39768


Yamnaya
KOTIAS
Hungarian
0.0103
0.0009
11.36
82719


Yamnaya
KOTIAS
.Hanna
0.0103
0.0034
3.00
44320


Yamnaya
KOTIAS
Balochi
0.0109
0.0009
11.91
82854


Yamnaya
KOTIAS
Sindhi
0.0113
0.0009
12.07
82551


Yamnaya
KOTIAS
Makrani
0.0118
0.0009
12.72
83399


Yamnaya
KOTIAS
kurd1198
0.0124
0.0036
3.47
46213


Yamnaya
KOTIAS
.Kurd_C2
0.0125
0.0035
3.58
39697


Yamnaya
KOTIAS
Brahui
0.0126
0.0009
14.27
82849


Yamnaya
KOTIAS
kurd1159
0.0131
0.0036
3.64
46022


Yamnaya
KOTIAS
Burusho
0.0158
0.0009
16.81
83051


Yamnaya
KOTIAS
.Bored
0.0164
0.0033
4.91
39748


Yamnaya
KOTIAS
.Khana
0.0166
0.0044
3.77
39577


Yamnaya
KOTIAS
.Arbogan
0.0167
0.0037
4.52
45998


Yamnaya
KOTIAS
.Kenji
0.0179
0.0037
4.80
39816


Yamnaya
KOTIAS
Iraqi_Jew
0.0182
0.0012
14.98
74176


Yamnaya
KOTIAS
.Varun
0.0184
0.0036
5.16
44728


Yamnaya
KOTIAS
Iranian_Jew
0.0186
0.0011
16.66
77550


Yamnaya
KOTIAS
kurd1160
0.0208
0.0043
4.80
45925


Yamnaya
KOTIAS
.Zephyrous
0.0263
0.0043
6.09
43787


Yamnaya
KOTIAS
kurd1173
0.0274
0.0049
5.58
45816


Yamnaya
KOTIAS
.Pak_Gujjar
0.0285
0.0047
5.99
43887


Yamnaya
KOTIAS
Kalash
0.0357
0.0012
30.40
78896


Yamnaya
KOTIAS
.Farid
0.0365
0.0057
6.43
39247


Yamnaya
KOTIAS
.Awale
0.0804
0.0042
19.03
42007


Yamnaya
KOTIAS
.Zara
0.0967
0.0096
10.10
38646


Yamnaya
KOTIAS
Ulchi
0.1305
0.0022
60.50
76765


Yamnaya
KOTIAS
Nganasan
0.1632
0.0027
61.05
70070

Kurd
11-23-2015, 04:44 AM
Sorted by strongest signal of admixture on top


SOURCE 1
SOURCE 2
TARGET
F3
STD ERROR
Z
SNPs


RISE_baAndrov
KOTIAS
.Kurd_SE
-0.00230
0.00339
-0.7
36979


RISE_baAndrov
KOTIAS
Lezgin
-0.00059
0.00123
-0.5
78021


RISE_baAndrov
KOTIAS
.Passa
-0.00006
0.00345
0.0
36488


RISE_baAndrov
KOTIAS
.Rukha
0.00051
0.00342
0.1
36766


RISE_baAndrov
KOTIAS
Adygei
0.00096
0.00116
0.8
81725


RISE_baAndrov
KOTIAS
.DMXX
0.00262
0.00334
0.8
42783


RISE_baAndrov
KOTIAS
.Bol_Nat
0.00282
0.00339
0.8
40297


RISE_baAndrov
KOTIAS
Chechen
0.00303
0.00123
2.5
77776


RISE_baAndrov
KOTIAS
.Sein
0.00305
0.00344
0.9
40863


RISE_baAndrov
KOTIAS
Georgian
0.00325
0.00127
2.6
78371


RISE_baAndrov
KOTIAS
Iranian
0.00420
0.00125
3.4
77743


RISE_baAndrov
KOTIAS
Tajik_Pomiri
0.00426
0.00127
3.4
77515


RISE_baAndrov
KOTIAS
.Hanna
0.00465
0.00350
1.3
40928


RISE_baAndrov
KOTIAS
Bulgarian
0.00487
0.00123
4.0
79085


RISE_baAndrov
KOTIAS
.Kurd_C1
0.00505
0.00361
1.4
36703


RISE_baAndrov
KOTIAS
Armenian
0.00520
0.00122
4.3
78802


RISE_baAndrov
KOTIAS
Hungarian
0.00574
0.00116
5.0
82615


RISE_baAndrov
KOTIAS
Tajik_Afghan
0.00645
0.00139
4.6
72871


RISE_baAndrov
KOTIAS
.Jesus
0.00648
0.00339
1.9
36995


RISE_baAndrov
KOTIAS
.Kandhari
0.00653
0.00388
1.7
36797


RISE_baAndrov
KOTIAS
.Kurd_C2
0.00676
0.00350
1.9
36571


RISE_baAndrov
KOTIAS
kurd_N
0.00679
0.00356
1.9
42533


RISE_baAndrov
KOTIAS
.Dluffy
0.00834
0.00358
2.3
36728


RISE_baAndrov
KOTIAS
Balochi
0.00834
0.00118
7.1
82762


RISE_baAndrov
KOTIAS
Makrani
0.00916
0.00119
7.7
83303


RISE_baAndrov
KOTIAS
Sindhi
0.00947
0.00120
7.9
82454


RISE_baAndrov
KOTIAS
Brahui
0.01000
0.00119
8.4
82762


RISE_baAndrov
KOTIAS
kurd_N
0.01003
0.00367
2.7
42517


RISE_baAndrov
KOTIAS
Burusho
0.01320
0.00121
10.9
82979


RISE_baAndrov
KOTIAS
Iraqi_Jew
0.01320
0.00139
9.5
73736


RISE_baAndrov
KOTIAS
Iranian_Jew
0.01375
0.00138
9.9
77292


RISE_baAndrov
KOTIAS
.Bored
0.01441
0.00360
4.0
36794


RISE_baAndrov
KOTIAS
.Arbogan
0.01481
0.00377
3.9
42461


RISE_baAndrov
KOTIAS
.Kenji
0.01563
0.00379
4.1
36725


RISE_baAndrov
KOTIAS
.Khana
0.01698
0.00443
3.8
36578


RISE_baAndrov
KOTIAS
.Varun
0.01780
0.00374
4.8
41486


RISE_baAndrov
KOTIAS
kurd_N
0.01871
0.00441
4.2
42422


RISE_baAndrov
KOTIAS
.Zephyrous
0.02237
0.00448
5.0
40413


RISE_baAndrov
KOTIAS
kurd_N
0.02424
0.00490
5.0
42159


RISE_baAndrov
KOTIAS
.Pak_Gujjar
0.02645
0.00484
5.5
40551


RISE_baAndrov
KOTIAS
Kalash
0.03340
0.00142
23.5
78728


RISE_baAndrov
KOTIAS
.Farid
0.03702
0.00569
6.5
36252


RISE_baAndrov
KOTIAS
.Awale
0.07401
0.00426
17.4
38924


RISE_baAndrov
KOTIAS
kurd_N
0.09253
0.00550
16.8
36421


RISE_baAndrov
KOTIAS
.Zara
0.09468
0.00967
9.8
35390


RISE_baAndrov
KOTIAS
Ulchi
0.12636
0.00237
53.4
76336


RISE_baAndrov
KOTIAS
Nganasan
0.15975
0.00280
57.1
69118

Chad Rohlfsen
11-23-2015, 05:12 AM
Eppie Jones shared the data with me, and I've begun running Admixture tests. I will post my first decent one into the crowd funding section. The first run yielded a Satsurblia component (100%).

parasar
11-23-2015, 05:22 AM
Does Kostentki share drift with a CHG like population?

I would say no, but I have limited understanding of statistical methods.
The reason I say no is because EEF and CHG share drift, while EEF's and K14's basal do not. So IMO K14's basal is one of the earlier splits while EEF's and CHG's basal maintained unity for much longer - up to 24kybp.

Chad Rohlfsen
11-23-2015, 05:24 AM
The non-basal part of CHG acts near Kostenki.

Krefter
11-23-2015, 07:09 AM
The non-basal part of CHG acts near Kostenki.

These stats say Kotias is equally close to K14 and Loschbour. Basal Eurasian IMO is exaggerated. Because of a mere 10% African most West Asians are closer(in D-stats) to Loschbour than to Saudi or BedouinB. This would not be the case if Basal Eurasian was something like 40% or 30% or 50% in West Asians.

Chimp Kotias Mbuti Kostenki14 0.3760 73.211 31070 14090 375211
Chimp Kotias Mbuti Loschbour 0.3760 74.120 36512 16557 434671

West Asians though are much closer to Loschbour than to Kostenki14. I expect the same to be true for Kotias.

Loschbour Kostenki14 BedouinA Chimp 0.0429
Loschbour Kostenki14 BedouinB Chimp 0.0444
Loschbour Kostenki14 Georgian Chimp 0.0497
Loschbour Kostenki14 Syrian Chimp 0.0494

Krefter
11-23-2015, 07:16 AM
Here is what seems to be the skull of Bichon:
............

His morphology appears to be very modern.

His skull looks like Loschbour's skull.
http://www.zlv.lu/spip/local/cache-vignettes/L374xH478/arton12822-1a5aa.jpghttp://www.anthrogenica.com/attachment.php?attachmentid=6683&d=1448245628&thumb=1

Gravetto-Danubian
11-23-2015, 08:01 AM
His skull looks like Loschbour's skull.


Yes, some definite familial resemblances, but little "evolution" (in some cases).
668466856687

Padre Organtino
11-23-2015, 08:40 AM
CHG are unlikely to have originated in South-Central Asia simply due to the fact that they obviously carry some WHG-like ancestry apart from ANE-like one (or simply have UHG admixture that is equidistant from the two). Meanwhile almost all Paleo/Meso cultures of that macro region that have been tested for DNA seem to favor ANE affinity over WHG one.
Zagreb mountains and South Caspian loo like very possible candidates though.

alan
11-23-2015, 10:10 AM
Some of the gentlemen in the Eurogenes comments section have been pouring over some formal stats regarding the exact role of CHG in South-Central Asia (among other questions, including the origin of CHG and its' relation to the hypothetical "Basal Eurasian", or Ryu's "Crown Eurasian").

Their assessment based on the outputs (which one can observe via the Z figures) is that Kotias Klde and Satsurbila didn't directly contribute to modern South-Central Asians.

Based on this, it appears that the West Asian related ancestry that predominates across South-Central Asia should be described as "CHG-like". We'll need aDNA from mesolithic cultures in northern Iran and South-Central Asia, yes.

A basic problem is that much of Europe, SW Asia and north Asia was first settled from the same Levantine Emiran source (in different phases- Emiran then its derivative Ahmarian) from 43000-38000BC. So IMO there will be at least three very similar looking branches coming off the 'generic' SW Asian ancestor. Kostenki is either an Aurignacian or proto-Gravettian guy who came through the Caucasus before 35000BC. CHG is probably some descendant of proto-Gravettians of the Caucasus area which settled about 38000BC. Ust Ism is apparently a decendent of an earlier wave who had reached Siberian by 43000BC. All the cultural roots trace to roots in the Levant Emiran-Ahmerian people in the Levant c. 45000-38000BC. So they probably look like three waves off a generic Emiran of the Levantine ancestor. In theory CHG should look slightly closer to Kostenki than Ust Ishm.

nuadha
11-23-2015, 10:43 AM
Why do you believe this is a more realistic scenario when we actually have evidence for female exogamy and high female mobility in Bronze Age Europe, including in the North Caucasus steppe region?

Wife swapping alone would not explain the huge influx of teal into the steppe.

Autosomal dna and uniparental dna shows that women tend to migrate with the men and that men/women mostly mated with their own. Thats the only way there has been such amazing autosomal turnovers in europe.

This applies to the indo europeans despite being "patriarchal".

So far, the evidence suggests that most the autosomal input of CHG into the EHG steppe favored the preservation of EHG ydna. I think subtle bias may be at play when the ydna of EHG wins out on the steppe. Perhaps a combination of migration, wife swapping, uneven population growth, and selection on the ydna.

Krefter
11-23-2015, 10:48 AM
Wife swapping alone would not explain the huge influx of teal into the steppe.

Autosomal dna and uniparental dna shows that women tend to migrate with the men and that men/women mostly mated with their own. Thats the only way there has been such amazing autosomal turnovers in europe.

This applies to the indo europeans despite being "patriarchal".

So far, the evidence suggests that most the autosomal input of CHG into the EHG steppe favored the preservation of EHG ydna. I think subtle bias may be at play when the ydna of EHG wins out on the steppe. Perhaps a combination of migration, wife swapping, uneven population growth, and selection on the ydna.

Two EHG Y DNAs(R1a-M417, R1b-L23) won out while 99.999999% went extinct. Actually one could have been EHG and one CHG.

Generalissimo
11-23-2015, 11:16 AM
Wife swapping alone would not explain the huge influx of teal into the steppe.

Yes it does. Very easily.


Autosomal dna and uniparental dna shows that women tend to migrate with the men and that men/women mostly mated with their own.

We're not discussing largescale population movements. We're talking about two populations adapted to very specific areas who interacted with each other for a long time.


So far, the evidence suggests that most the autosomal input of CHG into the EHG steppe favored the preservation of EHG ydna.

No it doesn't. You have no evidence to back up that statement.

ffoucart
11-23-2015, 01:24 PM
I don't see how one of R1a or R1b found in EHG could be of CHG origin.

Moreover, if CHG is very similar to modern Mingrelians and related populations, why there is so few of these haplogroups in Caucasus?

DMXX
11-23-2015, 02:51 PM
Sorted by strongest signal of admixture on top


Unless I'm misinterpreting the data... Surely, I can't have more Yamnaya+Kotias or Andronovo+Kotias admixture than the likes of Pamiri Tajiks?



We're not discussing largescale population movements. We're talking about two populations adapted to very specific areas who interacted with each other for a long time.


I believe it was Padre Organtino who mentioned earlier in this thread that any bidirectional wife-swapping would have affected the steppe population more so than the Neolithic agricultural communities of the Caucasus owing to population density differences. I think that adequately explains the sort of demographic outcomes we saw from whatever interactions (presumably peaceful) had between the two groups.

Kurd
11-23-2015, 03:23 PM
Unless I'm misinterpreting the data... Surely, I can't have more Yamnaya+Kotias or Andronovo+Kotias admixture than the likes of Pamiri Tajiks?



I believe it was Padre Organtino who mentioned earlier in this thread that any bidirectional wife-swapping would have affected the steppe population more so than the Neolithic agricultural communities of the Caucasus owing to population density differences. I think that adequately explains the sort of demographic outcomes we saw from whatever interactions (presumably peaceful) had between the two groups.

You have absolutely correctly interpreted it. I will give you the reasons once I get settled into my office in an hour

Kurd
11-23-2015, 05:13 PM
Unless I'm misinterpreting the data... Surely, I can't have more Yamnaya+Kotias or Andronovo+Kotias admixture than the likes of Pamiri Tajiks?



I believe it was Padre Organtino who mentioned earlier in this thread that any bidirectional wife-swapping would have affected the steppe population more so than the Neolithic agricultural communities of the Caucasus owing to population density differences. I think that adequately explains the sort of demographic outcomes we saw from whatever interactions (presumably peaceful) had between the two groups.

My various ADMIXTURE runs have shown that you do indeed have slightly more CHG admixture than Pomiri Tajiks. The fact that they may have more Yamnaya/Andronovo type admixture than you is not very relevant here, because f3 (T ;Kotias, Andronovo) output can be the same whether target T has 70% Kotias/ 20% Andronovo, or 80%Kotias/10% Andronovo. What is relevant is that whose admixture, you or Pomiri, can be better captured by an only 2 way Kotias/Andronovo model. Apparently, some of Pomiri's admixture (guessing S Asian related) can not be as easily modeled simply with Kotias and Andronovo, whereas most of your admixture components can (you have less S Asian related).

Incidentally, you may have noticed that you are better modeled with Andronovo than Yamnaya, which is expected.

Kurd
11-23-2015, 05:39 PM
I also wanted to share that I had early on noticed that Satsurbila's genome was haploid, when I saw that the allelic values on one chromo strand mirrored the other. This was done likely because of low coverage.

The accuracy of the results of certain analysis may thus be slightly less accurate than Kotias. I have noticed in higher K in Admixture, Sats splits away from Kotias, SC and W Asians, and forms a cluster with Yamnaya/Andronovo. At low K however, he clusters with Kotias and Asians.

vettor
11-23-2015, 05:50 PM
I believe it was Padre Organtino who mentioned earlier in this thread that any bidirectional wife-swapping would have affected the steppe population more so than the Neolithic agricultural communities of the Caucasus owing to population density differences. I think that adequately explains the sort of demographic outcomes we saw from whatever interactions (presumably peaceful) had between the two groups.

Or that R1 group where the first to create/have chiefs/kings and a sort of a concubine system for the women , while other haplogroups had no "chiefs/kings"

Tomenable
11-23-2015, 06:15 PM
Do you have a link to that comment by Lazaridis about CHGs ???

I can't find it again.

Edit:

OK I found it: https://twitter.com/iosif_lazaridis/status/666355572501954560

http://s28.postimg.org/be11fd19p/CHG_Lazaridis.png

Tomenable
11-23-2015, 06:44 PM
Do you have a spreadsheet for K11? Would like to compare other European results with my own family..

Me:
3 Caucas-Gedrosia 16.72

Paternal aunt:
3 Caucas-Gedrosia 16.44

Maternal grandmother:
3 Caucas-Gedrosia 18.67

Maternal grandfathers brother:
3 Caucas-Gedrosia 20.87

Seems western Norwegians score around 15-20% Teal/CHG..

Is Teal/CHG equivalent with Caucasus-Gedrosia ???

And what about calculators which count Caucasus and Gedrosia separately ???

Is Teal/CHG more similar to Caucasus, or to Gedrosia ???

alan
11-23-2015, 11:01 PM
There could have been an balanced gender influx of Teal people from the Caucasus but perhaps they had a different type of social structure from the R1b elements on the steppe. If the teal J males essentially lived a Neolithic type household based lifestyle and they interacted with superbreeder R1b herders on the steppe who took tons of wives then the net effect would be that the superbreeders would take in teal mt and autosomal DNA which growing their own R1b y lines massively while the Teal people with a more household based social structure may have essentially not have grown their y lines much if at all.

I would also point out that by Yamnaya times you are clearly talking about a light population of clans with chiefs. The pattern we see in Gaelic clans is the lines of chiefs and kings produced surviving children at epic rates in a top downwards demographic squeeze. So, let think this through. If so many of the surviving children stem from the very top of the pyramid then teal people dont have to effect the entire population. Teal genes just have to get into a few top steppe chiefs through marrying teal wives. This immediately halves those local steppe genes in those chiefs (apparently many) children and makes them half teal. This chief goes on to have 20 of these steppe-teal half-half kids and maybe so does his brother and close cousins who are also near the top of the tree. In a system of clans where the demographic push is very strongly downwards from the top this will have a major impact.

Lets put this even more specifically. A Z2103 chief (not the first Z2103 guy) and perhaps his 4 brothers and 10 close cousins (who are his lieutenants and henchmen) see value in alliance with Maykop people because of their metal knowledge. So this small bunch of 15 elite steppe males, in return for permitting 50 or so Maykop related settlers of all sexes settle up the Don and Volga and establish trade bases and eventually mines, takes two or three wives each from the daughters of the teal Maykop population. This is not a big issue for the Maykop population because it is the norm even in non-steppe Neolithic or copper age groups to marry their daughters off to live elsewhere. This happens in say 3500BC. The little group of 15 steppe men in and around the chiefship of the clan dominate reproduction in the typical top down demographic push of clans produce a way out of proportion number of the surviving children of the clan - lets say 100 children. This group then forms the core that produced the Yamnaya culture after borrowing the wheel (perhaps also from Maykop) and effectively inventing a new way of life that allowed incredible demographic expansion of the ENF-teal 50-50 group.

All a scenario like this would require is perhaps a phase or two where there was a scramble among competing R1b steppe chiefs to make useful contacts with Caucasus people and seal the deal with alliance marriages. This could have happened a number of times as the Caucasus was generally more advanced than the steppes - especially as you headed west of the Don. The Caucasus had farming of course and then later the same area had the precocious Maykop culture with its metallurgical focus. So even keeping it very simplistic, the Caucasus had at least two phases where aspirational steppe EHG groups may have wished to have contact and alliances sealed by marriages. Of course in reality those phases roll into each other so there may have been a gradual infiltration of teal genes which accelerated in the Maykop period.

alan
11-23-2015, 11:27 PM
Or to put it much more concisely, in societies where the top produces an insane amount of the surviving children, only the top needs to marry some teal women to have a massive impact on the next generation. If the tradition of marrying the daughters of their teal neighbours continued even for a century this could become solidified in the clan DNA. In mobile pastoralist clans who had founded a totally new lifestyle in the Yamnaya phase, sometimes moving into empty open areas of steppe, it hard to see how periods of relative in-breeding could have been avoided.

BTW I have simply set out to describe how it could have happened in the immediate pre-Yamnaya and early Yamnaya period of Maykop contact c. 3500-3300BC as well as likely earlier pre-Maykop phases of contact with Caucasus farmers. I suppose R1a having similar levels of teal needs explained too. However thinking about that it struck me if you turnn that on its head, you could use the fact R1a absorbed a load of apparently Caucasus derived teal to infer where R1a and teal combined. When I try and do that it makes me think a previous belief that R1a was probably located in the Middle Dnieper is probably wrong - there are not strong indicators of contact with the Caucasus in that area. In all probability the R1a-teal connection must have happened in an area where archaeology is more suggestive of links with the Caucasus which poils down to the areas of the steppe close to the Caucasus and evidence for contacts heading up the Don and Volga. As its been discovered in Khvaysk that probably proves the point. Obviously by perhaps 3000BC or a little after R1a did get itself in a position to move up into the middle Dniester-Middle Dnieper area and into the CW genesis. However, this is a most unlikely area for the R1a-EHG-Teal to have originally combined.

Chad Rohlfsen
11-23-2015, 11:30 PM
I should have the Anatolians later tonight and be able to run the first test.

Gravetto-Danubian
11-23-2015, 11:31 PM
There could have been an balanced gender influx of Teal people from the Caucasus but perhaps they had a different type of social structure from the R1b elements on the steppe. If the teal J males essentially lived a Neolithic type household based lifestyle and they interacted with superbreeder R1b herders on the steppe who took tons of wives then the net effect would be that the superbreeders would take in teal mt and autosomal DNA which growing their own R1b y lines massively while the Teal people with a more household based social structure may have essentially not have grown their y lines much if at all.

I would also point out that by Yamnaya times you are clearly talking about a light population of clans with chiefs. The pattern we see in Gaelic clans is the lines of chiefs and kings produced surviving children at epic rates in a top downwards demographic squeeze. So, let think this through. If so many of the surviving children stem from the very top of the pyramid then teal people dont have to effect the entire population. Teal genes just have to get into a few top steppe chiefs through marrying teal wives. This immediately halves those local steppe genes in those chiefs (apparently many) children and makes them half teal. This chief goes on to have 20 of these steppe-teal half-half kids and maybe so does his brother and close cousins who are also near the top of the tree. In a system of clans where the demographic push is very strongly downwards from the top this will have a major impact.

Lets put this even more specifically. A Z2103 chief (not the first Z2103 guy) and perhaps his 4 brothers and 10 close cousins (who are his lieutenants and henchmen) see value in alliance with Maykop people because of their metal knowledge. So this small bunch of 15 elite steppe males, in return for permitting 50 or so Maykop related settlers of all sexes settle up the Don and Volga and establish trade bases and eventually mines, takes two or three wives each from the daughters of the teal Maykop population. This is not a big issue for the Maykop population because it is the norm even in non-steppe Neolithic or copper age groups to marry their daughters off to live elsewhere. This happens in say 3500BC. The little group of 15 steppe men in and around the chiefship of the clan dominate reproduction in the typical top down demographic push of clans produce a way out of proportion number of the surviving children of the clan - lets say 100 children. This group then forms the core that produced the Yamnaya culture after borrowing the wheel (perhaps also from Maykop) and effectively inventing a new way of life that allowed incredible demographic expansion of the ENF-teal 50-50 group.

All a scenario like this would require is perhaps a phase or two where there was a scramble among competing R1b steppe chiefs to make useful contacts with Caucasus people and seal the deal with alliance marriages. This could have happened a number of times as the Caucasus was generally more advanced than the steppes - especially as you headed west of the Don. The Caucasus had farming of course and then later the same area had the precocious Maykop culture with its metallurgical focus. So even keeping it very simplistic, the Caucasus had at least two phases where aspirational steppe EHG groups may have wished to have contact and alliances sealed by marriages. Of course in reality those phases roll into each other so there may have been a gradual infiltration of teal genes which accelerated in the Maykop period.

It certainly sounds logical, and there is no doubt the steppe chiefs were involved in elite exchanges with more distant regions. But one fact which is omitted is that Majkop were also pastoralist and potentially patriarchical clans. I'm not sure where people are obtaining the notion that Majkop were settled farmers, and it goes against all the evidence - at least recent interpretations of it.

So it perhaps Majkop which acquired and Teal from further south and transmitted to the lower Don steppe peoples, then west and east.

So the exact 'who got what from where' hinges on what the Reich lab gets from Majkop : R1b, J and/ or G2 ?

PS: I still think we need to consider certain 'biological' actors inherent in the Y chromosome (eg its low Ne) in addition to socio-cultural practices.

alan
11-23-2015, 11:45 PM
A wayward thought which just occurred to me.

So far, in West Asia, we've picked up two distinct components - EEF and CHG.

EEF, from memory, is best represented by modern-day Sardinians. Several modern population-based ADMIXTURE calculators picked up a "Mediterranean" component that also peaked in Sardinians.

CHG, according to both Jones et al. and Kurd's work, peaks in the Caucasus as well as South-Central Asia. This is precisely where we've observed over the years through the ADMIXTURE "Gedrosian", "teal" or "Caucasus-Gedrosia" components (or at least CHG-like, judging by some of the formal stats).

If this tenuous bridging between old and new is maintained, then my earlier guess at a third (and final) West Asian auDNA component (represented by "SW Asian") might be the final piece of the puzzle for the region. Perhaps some Natufian, Kebaran or Zarzian culture aDNA will offer this?

Based on this, it appears as if the modern population ADMIXTURE runs weren't completely smoke and mirrors, and actually did provide reasonable evidence of prehistoric movements in Eurasia.

Alternatively, some extraneous complications are at work. Occam's Razor applies until opposing data from the region presents itself (hopefully soon).

Zarzian appears (from what little I have found on this) to derived ultimately from the Aurignacian groups of the Zagros. From what I understand they are clearly techologically linked to the Ahmarian of the Levant and proto-Aurignacian of Europe. So, I think they will be yet another isolate that split off early from a Basal root c. 40000BC. I generally think the several waves out of the Levant c. 45000-38000BC are all going to look very similar. IMO basal was simply the default of modern humans in Levant and where they spread from there into Europe, SW Asia and northern Asia and all other signals like WHG and ANE formed by isolation and other factors long after dispersal - probably closer to 30000BC.

My guess is Natufian will have its own distinct signal and be E yDNA dominated.However it may also derive through a chain of cultures back to the Emiran so may end up also being an basal branch off too.

Tomenable
11-24-2015, 12:07 AM
Here is a good video lecture, criticizing a study which created a model of PIE origins among Neolithic farmers:

Mismodeling Indo-European Origins: the Assault On Historical Linguistics:

https://www.youtube.com/watch?v=4jHsy4xeuoQ

http://www.geocurrents.info/cultural-geography/linguistic-geography/mismodeling-indo-european-origin-and-expansion-bouckaert-atkinson-wade-and-the-assault-on-historical-linguistics

Now some interesting fragments:

Part of the lecture about languages that they left out in their model:

https://www.youtube.com/watch?v=4jHsy4xeuoQ#t=2622

About interactions of Early Proto-Indo-Europeans with Proto-Uralics:

https://www.youtube.com/watch?v=4jHsy4xeuoQ#t=2676

And about Late Indo-European groups borrowing words from farmers:

https://www.youtube.com/watch?v=4jHsy4xeuoQ#t=2727

"(...) One thing that makes Greek [language] distinct is that it has this huuuge substrate of Non-Indo-European words. And guess what, those words are words for agriculture, for building, for statecraft. According to this model you had farmers crossing the Aegean Sea, coming into an area of Mesolithic hunter-gatherers, and then borrowing words for agriculture from those hunter-gatherers. There is the same problem with Armenian [language], probably the same problem with Germanic [language], and the same problem also with the Indic languages. (...)"

It seems that Proto-Indo-Europeans were hunters who turned into animal herders, with no agricultural vocabulary.

Not much Neolithic Near Eastern farmer influence there, at all.

So the model in which PIE emerged when two groups of hunters (EHG + CHG) mixed, explains PIE origins quite well.

alan
11-24-2015, 12:14 AM
It certainly sounds logical, and there is no doubt the steppe chiefs were involved in elite exchanges with more distant regions. But one fact which is omitted is that Majkop were also pastoralist and potentially patriarchical clans. I'm not sure where people are obtaining the notion that Majkop were settled farmers, and it goes against all the evidence - at least recent interpretations of it.

So it perhaps Majkop which acquired and Teal from further south and transmitted to the lower Don steppe peoples, then west and east.

So the exact 'who got what from where' hinges on what the Reich lab gets from Majkop : R1b, J and/ or G2 ?

PS: I still think we need to consider certain 'biological' actors inherent in the Y chromosome (eg its low Ne) in addition to socio-cultural practices.

True it is possible that teal could have originated in the south Caucasus and somehow made its way north. There are many who see the pre-Maykop north Caucasus farmers as coming from the south Caucasus so that perhaps gives the simplest explanation.

The copper age is far poor in evidence of a south Caucasus to north Caucasus flow. Maykop originated many centuries before Kura-Araxes (and the two cultures were utterly opposites in settlement types, burial traditions and apparently social structure) so its unusual advanced nature and metallurgical know how may come from the north Iran/Caspian direction as Ivanova convincingly argues.

The problem is we dont fully understand the origin of Maykop in its entirety and the role of the north Caucasus farmers who were there before it. Maykop to me looks more of coalescence of influences and development of something unique than a major A to B migration in the traditional sense.
.

tamilgangster
11-24-2015, 12:17 AM
On harappa DNA south indian tribals such as irulas score caucasian but not baloch, CHG is probably the likely source for this. Is CHG the source for th Gedrosian type ENF that dr mcninja was talking about

alan
11-24-2015, 12:19 AM
It certainly sounds logical, and there is no doubt the steppe chiefs were involved in elite exchanges with more distant regions. But one fact which is omitted is that Majkop were also pastoralist and potentially patriarchical clans. I'm not sure where people are obtaining the notion that Majkop were settled farmers, and it goes against all the evidence - at least recent interpretations of it.

So it perhaps Majkop which acquired and Teal from further south and transmitted to the lower Don steppe peoples, then west and east.

So the exact 'who got what from where' hinges on what the Reich lab gets from Majkop : R1b, J and/ or G2 ?

PS: I still think we need to consider certain 'biological' actors inherent in the Y chromosome (eg its low Ne) in addition to socio-cultural practices.

Maykop is very hard to predict in terms of yDNA. Its basically in an area with no barrier at all to the north, a major barrier to the south and another easier contact route down towards Iran. It could be a real mix.

nuadha
11-24-2015, 01:41 AM
Yes it does. Very easily.



We're not discussing largescale population movements. We're talking about two populations adapted to very specific areas who interacted with each other for a long time.

We are talking about migration, when the samara valley, far from the caucasus, went from no teal to 50% teal. Teal admixed men, must have migrated too.


No it doesn't. You have no evidence to back up that statement.

Lol, re read what I wrote. Im positive we agree on this.

Gravetto-Danubian
11-24-2015, 01:59 AM
Genetiker has analyzed that the R1b from Khvalynsk was R1b M415(xP297) and the R1a was M459 (xM198). ...

https://genetiker.wordpress.com/2015/11/24/more-y-haplogroups-for-prehistoric-eurasian-genomes/

Gravetto-Danubian
11-24-2015, 02:02 AM
True it is possible that teal could have originated in the south Caucasus and somehow made its way north. There are many who see the pre-Maykop north Caucasus farmers as coming from the south Caucasus so that perhaps gives the simplest explanation.

The copper age is far poor in evidence of a south Caucasus to north Caucasus flow. Maykop originated many centuries before Kura-Araxes (and the two cultures were utterly opposites in settlement types, burial traditions and apparently social structure) so its unusual advanced nature and metallurgical know how may come from the north Iran/Caspian direction as Ivanova convincingly argues.

The problem is we dont fully understand the origin of Maykop in its entirety and the role of the north Caucasus farmers who were there before it. Maykop to me looks more of coalescence of influences and development of something unique than a major A to B migration in the traditional sense.
.

See here 6690


It is the latest chronology of the Caucasus region c/- Ancient Metallurgy in the Caucasus From the Sixth to the Third Millennium BCE by Antoine Courcier.

It shows that there is little to speak of in the Nth Caucasus before c. 4500 BC - at which point the pre-Majkop (so-called "Meshoko" horizon begins). Im not sure if this is related to poor research or a real lack of early Neolithic finds. But it does show that there was a notable pre-Kura Araxes horizon in the Sth Caucasus (=Sioni culture), as well as the Darkveti culture in Nth Black Sea region, which preceds Meshoko in the north Caucasus.

Based on traditional archaeology, scholars have not any unequivocal ideas as to where these pre-Majkop groups came from to the North Caucasus (some argue the south, some the Black Sea, some even draw parallels to Neolithic groups in Germany !). In fact, the later, majkop phases is markedly different to Meshoko again. So we could be looking at multiple movements around the Nth Caucasus, proto-Majkop region.

So yes, Alan, I agree it is hard to predict exactly how Majkop will turn out. And IMO we also need the pre-Majkop (ie Meshoko) phase genomes also.

Chad Rohlfsen
11-24-2015, 02:48 AM
First test underway! 1990 samples x 105904 SNPs

I only lost 4 Anatolians with quality controls.

Edit: I forgot to add, 94.6% coverage in the test. It will go up as I prune more of those with less coverage.

parasar
11-24-2015, 03:02 AM
CHG are unlikely to have originated in South-Central Asia simply due to the fact that they obviously carry some WHG-like ancestry apart from ANE-like one (or simply have UHG admixture that is equidistant from the two). Meanwhile almost all Paleo/Meso cultures of that macro region that have been tested for DNA seem to favor ANE affinity over WHG one.
Zagreb mountains and South Caspian loo like very possible candidates though.

Almost everyone has some WHG like ancestry. WHG ancestry is another matter. Per Jones et al., the CHG (unlike EEF) have no WHG admixture, but only shared ancestry with WHG, and that too very early (~46.4kybp).

http://www.nature.com/ncomms/2015/151116/ncomms9912/images/ncomms9912-f2.jpg

Arame
11-24-2015, 09:28 AM
Kotias
http://s019.radikal.ru/i619/1511/96/6a5d57f4d459.png

MfA
11-24-2015, 10:06 AM
Kotias
http://s019.radikal.ru/i619/1511/96/6a5d57f4d459.png

IBD scores

Georgian 71,19
Balkarian 67,02
Lezgin 63,76
Armenian 63,46
Abkhazian 60,48
Kurd 58,69
Nogay 58,2
Ossetian 58,09
Bosnian 56,65
Cypriot 55,67
Gagauz 55,4
Croatian 55,12
Syrian 53,43
Kosovar 53,33
Adygei 52,98
Ukrainian-West-and-Center 52,72
Uttar-Pradesh-HC 52,53
Montenegrian 52,29
Iranian 52,16
UAE 52,15

Hanna
11-24-2015, 12:11 PM
IBD scores

Georgian 71,19
Balkarian 67,02
Lezgin 63,76
Armenian 63,46
Abkhazian 60,48
Kurd 58,69
Nogay 58,2
Ossetian 58,09
Bosnian 56,65
Cypriot 55,67
Gagauz 55,4
Croatian 55,12
Syrian 53,43
Kosovar 53,33
Adygei 52,98
Ukrainian-West-and-Center 52,72
Uttar-Pradesh-HC 52,53
Montenegrian 52,29
Iranian 52,16
UAE 52,15

Could you add Turkish ?

Michał
11-24-2015, 02:25 PM
So it perhaps Majkop which acquired and Teal from further south and transmitted to the lower Don steppe peoples, then west and east.

It seems that CHG was present on the NPC steppe long before Maykop was formed. Khvalynsk predates Maykop by nearly a millennium, yet the only R1b sample from Khvalynsk is not much different (autosomally) from the CHG-rich R1b samples from Yamna_Samara. This is not to say that the Maykop people did not contribute to the steppe people, as they probably did, only that this likely wasn't a decisive CHG-related input that shaped the Late PIE-speaking populations descending either from Khvalynsk itself or from a very closely related steppe culture.

Shaikorth
11-24-2015, 05:17 PM
Bichon and Loschbour combined.

http://s018.radikal.ru/i524/1511/8c/36e36c8180a7.png

CHG - WHG difference
http://s019.radikal.ru/i628/1511/a0/129b1db79342.png

CHG- Yamnaya difference
http://s018.radikal.ru/i514/1511/7a/75cf3f34c789.png

Padre Organtino
11-24-2015, 05:34 PM
What this might imply is that the people with whom EHGs mixed to form Yamnaya IEs weren't Adygo-Abkhaz speakers. Otherwise we would so more IBD sharing with them while here they are comparable to some North Euros and far below Georgians.

Polako's Kartvelian connection theory suddenly looks less wild than I though initially.

parasar
11-24-2015, 06:21 PM
There is a CHG peak among the Uttar Pradesh HC (Brahmans likely) in northern India. It is higher than in the Baloch, Makrani, Pathan, Kalash - very surprising. I suppose this would apply to me too (even though I am from the area bookended by the UP Kol and Munda, they are relatively smaller populations). WHG also shows a slight uptick (cf minimas in the Brahui and in Karnataka) among Uttar Pradesh HC.

So it appears that the predominant influence on UP Brahmans is not from the steppe (though there is some as seen from the slightly elevated WHG) despite their high R1a1.

CHG:
http://s019.radikal.ru/i619/1511/96/6a5d57f4d459.png

WHG:
http://s018.radikal.ru/i524/1511/8c/36e36c8180a7.png

Shaikorth
11-24-2015, 06:57 PM
There is a CHG peak among the Uttar Pradesh HC (Brahmans likely) in northern India. It is higher than in the Baloch, Makrani, Pathan, Kalash - very surprising. I suppose this would apply to me too (even though I am from the area bookended by the UP Kol and Munda, they are relatively smaller populations). WHG also shows a slight uptick (cf minimas in the Brahui and in Karnataka) among Uttar Pradesh HC.

So it appears that the predominant influence on UP Brahmans is not from the steppe (though there is some as seen from the slightly elevated WHG) despite their high R1a1.


Look at the CHG-Yamnaya difference above though. The peak in Uttar Pradesh high caste vanishes and Makranis start standing out instead.

Arbogan
11-24-2015, 07:33 PM
CHG - WHG difference

http://s019.radikal.ru/i628/1511/a0/129b1db79342.png

Who made these maps? How can use and Oman have higher IBD with kotias than the levant and Iran.

Gravetto-Danubian
11-24-2015, 07:47 PM
It seems that CHG was present on the NPC steppe long before Maykop was formed. Khvalynsk predates Maykop by nearly a millennium, yet the only R1b sample from Khvalynsk is not much different (autosomally) from the CHG-rich R1b samples from Yamna_Samara. This is not to say that the Maykop people did not contribute to the steppe people, as they probably did, only that this likely wasn't a decisive CHG-related input that shaped the Late PIE-speaking populations descending either from Khvalynsk itself or from a very closely related steppe culture.

True, although we don't actually know when the samples date from specifically. A very general dating range is given by the Mathieson study; and if corrected to the most recent, radiodates; the generic "Khvalynsk horizon" dates from 4200-3000. {"Khvalysnk" is used as a generic for the entire Eneolithic period between the Neolithic and Bronze age Yamnaya :4200-3000 in SW Russia; as is "Sredni Stog" in Ukraine}
So in reality; these samples could be from, say, 3500 BC; and thus after Majkpp began :)

But yes, this Khvalynsk samples possibly predate Majkop proprie dictii, but the significantly lesser amount of CHG admixture suggests that the process was just beginning at c. 42/ 4000 BC- which corresponds with near exactness the 'sudden' start of the pre-Majkop phase in the north Caucasus

So when I said Majkop I meant the " proto-Majkop " peoples of the north Caucasus. the overall conclusions are unchanged

PS:
In fact in willing to bet if/ when properly dated; the specimens from Khvalynsk date from after c.3800 Bc

PPS:
The sequence of events can be no accident: 4100 BC collapse of final Balkan copper centre
3900 BC: rose of Caucasian metallurgical centres

Shaikorth
11-24-2015, 07:49 PM
Who made these maps? How can use and Oman have higher IBD with kotias than the levant and Iran.

Srkz (https://verenich.wordpress.com/2015/11/24/%D0%BE%D1%85%D0%BE%D1%82%D0%BD%D0%B8%D0%BA%D0%B8-%D1%81%D0%BE%D0%B1%D0%B8%D1%80%D0%B0%D1%82%D0%B5%D 0%BB%D0%B8-%D0%BA%D0%B0%D0%B2%D0%BA%D0%B0%D0%B7%D0%B0-%D0%B8-%D1%8E%D0%B6%D0%BD%D1%8B%D0%B9-%D0%B3/) made them, and that's not straight on Kotias IBD but the difference between Kotias IBD and WHG IBD.

Arbogan
11-24-2015, 08:11 PM
Srkz (https://verenich.wordpress.com/2015/11/24/%D0%BE%D1%85%D0%BE%D1%82%D0%BD%D0%B8%D0%BA%D0%B8-%D1%81%D0%BE%D0%B1%D0%B8%D1%80%D0%B0%D1%82%D0%B5%D 0%BB%D0%B8-%D0%BA%D0%B0%D0%B2%D0%BA%D0%B0%D0%B7%D0%B0-%D0%B8-%D1%8E%D0%B6%D0%BD%D1%8B%D0%B9-%D0%B3/) made them, and that's not straight on Kotias IBD but the difference between Kotias IBD and WHG IBD.
Ah Thanks.

Here is the map for IBD distribution:
https://verenich.files.wordpress.com/2015/11/kotiassnpc-100ibdext.png?w=1304

IBD scoring by population:
Georgian 71,79
Abkhazian 70,75
Lezgin 68,27
Greek_Azov 67,15
Balkarian 65,02
Kurd 64,38
Ossetian 62,66
Armenian 61,98
Nogay 60,38
Bosnian 60,23
Slovenian 60,02
Chechen 59,07
Adygei 58,39
Cypriot 58,28
Turkish 55,86
Kosovar 54,64
Ukrainian-West-and-Center 54,17
Bulgarian 53,21
Slovak 53,01
Cornish 52,46
Croatian 52,21
Kumyk 51,96
Makrani 51,91
Syrian 51,78
Greek 51,68

Padre Organtino
11-24-2015, 08:29 PM
Greek_Azov, Slovenian, Bosnian and Cypriot scores are pretty interesting

vettor
11-24-2015, 08:33 PM
Bichon and Loschbour combined.

http://s018.radikal.ru/i524/1511/8c/36e36c8180a7.png

CHG - WHG difference
http://s019.radikal.ru/i628/1511/a0/129b1db79342.png

CHG- Yamnaya difference
http://s018.radikal.ru/i514/1511/7a/75cf3f34c789.png

I would really appreciate the use of Bergamo on these maps, since its used for every admixture ..............it represents Swiss, north-italian and austria ( western )

Tomenable
11-24-2015, 08:54 PM
Genetiker has analyzed that the R1b from Khvalynsk was R1b M415(xP297) and the R1a was M459 (xM198). ...

https://genetiker.wordpress.com/2015/11/24/more-y-haplogroups-for-prehistoric-eurasian-genomes/

So according to Genetiker, Khvalynsk R1b sample was not only xM269 but even xP297, and Khvalynsk R1a was xM198.

As we know Genetiker is pushing an agenda of Paleolithic Continuity of R1b in Western Europe, though.

Do you remember his claims about El Portalon cave and that - allegedly - M269 was there? Here:

https://genetiker.wordpress.com/2015/09/08/y-snp-calls-from-copper-and-bronze-age-spain/


ATP3 Pre-Beaker Copper Age 3516–3362 BC R1b1a2-M269 calls

Other researchers did not confirm that M269 call from Genetiker, as far as I know.

So I'm probably going to wait until someone else analyses Khvalynsk samples and checks if Genetiker is right.

==========================================

On the other hand, even if Genetiker is right, it is not yet a definite proof that M269 and M198 were not present in Khvalynsk.

Let's note that Khvalynsk samples are apparently from 4700-3800 BC, which might be too early for M269 to be its main lineage, because according to YFull's age estimates (if we assume that they are correct*), TMRCA of M269 is just 4400 BC.

It doesn't mean that M269 did not exist in Khvalynsk tribe(s), but it could be few in numbers (and only multiplied later).

We have a sample of R1b from Khvalynsk, but we don't know if it was already M269, or something more basal.

On the other hand R1a M198 has TMRCA of 6000 BC according to YFull, so it could be more numerous in Khvalynsk than M269.

And we have 1 sample of R1a from Khvalynsk, but again - we don't know if it was M198 or something more basal.

I'd not be surprised if it turns out, that R1a-Tarim = R1a-Khvalynsk, or a lineage descended from R1a-Khvalynsk.

====================

*According to some people, YFull notoriously underestimates age by 10% to 20%. But that's just their opinion.

Gravetto-Danubian
11-24-2015, 08:59 PM
So according to Genetiker, Khvalynsk R1b sample was not only xM269 but even xP297, and Khvalynsk R1a was xM198.

As we know Genetiker is pushing an agenda of Paleolithic Continuity of R1b in Western Europe, though.

Do you remember his claims about El Portalon cave and that - allegedly - M269 was there? Here:

https://genetiker.wordpress.com/2015/09/08/y-snp-calls-from-copper-and-bronze-age-spain/



Other researchers did not confirm that M269 call from Genetiker, as far as I know.

So I'm probably going to wait until someone else analyses Khvalynsk samples and checks if Genetiker is right.

==========================================

On the other hand, even if Genetiker is right, it is not yet a definite proof that M269 and M198 were not present in Khvalynsk.

Let's note that Khvalynsk samples are apparently from 4700-3800 BC, which might be too early for M269 to be its main lineage, because according to YFull's age estimates (if we assume that they are correct*), TMRCA of M269 is just 4400 BC.

It doesn't mean that M269 did not exist in Khvalynsk tribe(s), but it could be few in numbers (and only multiplied later).

We have a sample of R1b from Khvalynsk, but we don't know if it was already M269, or something more basal.

On the other hand R1a M198 has TMRCA of 6000 BC according to YFull, so it could be more numerous in Khvalynsk than M269.

And we have 1 sample of R1a from Khvalynsk, but again - we don't know if it was M198 or something more basal.

I'd not be surprised if it turns out, that R1a-Tarim = R1a-Khvalynsk, or a lineage descended from R1a-Khvalynsk.

====================

*According to some people, YFull notoriously underestimates age by 10% to 20%. But that's just their opinion.

One can't help but be suspicious when he insists that Papaleolithic Europe was full of R1b. Where ?
:)

parasar
11-24-2015, 09:03 PM
So according to Genetiker, Khvalynsk R1b sample was not only xM269 but even xP297, and Khvalynsk R1a was xM198.

As we know Genetiker is pushing an agenda of Paleolithic Continuity of R1b in Western Europe, though.

Do you remember his claims about El Portalon cave and that - allegedly - M269 was there? Here:

...

A few (Ted Kendall, Richard Rocca http://www.anthrogenica.com/showthread.php?1646-Genome-of-a-late-Neolithic-Iberian-farmer&p=107905&viewfull=1#post107905 ), did confirm Genetiker's read of an M269 level SNP. But while the read is confirmed, the designation is still being disputed.

Tomenable
11-24-2015, 09:05 PM
Where ?

All squeezed in one cave ??? Or maybe R1b were cremation-folks (just to piss off future researchers). :P

Tomenable
11-24-2015, 09:22 PM
It is also astonishing that so many of Corded Ware guys were M417 but xZ645, according to Genetiker:

https://genetiker.wordpress.com/2015/11/24/more-y-haplogroups-for-prehistoric-eurasian-genomes/

I1532 Germany Corded Ware R1a1a1-M417(xZ645) calls
I1538 Germany Corded Ware R1a1a1-M417(xZ645) calls
I1540 Germany Corded Ware R1a1a1-M417(xZ645) calls
I1541 Germany Corded Ware R1a1a1-M417(xZ645) calls
I1542 Germany Corded Ware R1a1a1-M417(xZ645) calls
I1544 Germany Corded Ware R1a1a1-M417(xZ645) calls

So all these lineages aren't dominant today, because most of Eurasian R1a is under Z645 (also Z93 is).

In case of M417, only rare M417* and rare Western European subclade CTS4385 aren't under Z645.

So it looks like most of Corded Ware male lineages got extinct as well.

==================================================

The remaining two Corded Ware samples from this most recent analysis by Genetiker:

I1534 Germany Corded Ware R1a1a-M198 calls
I1536 Germany Corded Ware R1a1a1-M417 calls

Potapovka:

I0246 Russia Potapovka R1a1a1-M417 calls

Timber Grave:

I0361 Russia Timber Grave R1a1a1b2a2-Z2124 calls

Gravetto-Danubian
11-24-2015, 10:27 PM
It is also astonishing that so many of Corded Ware guys were M417 but xZ645,
So it looks like most of Corded Ware male lineages got extinct as well.



Again, Im not surprised. I always doubted a straightforward equation of CWC being directly ancestral proto-Germanic and Balto-Slavic. Its always been clear from the dynamic settlement history in east-central Europe.

Krefter
11-24-2015, 11:12 PM
Again, Im not surprised. I always doubted a straightforward equation of CWC being directly ancestral proto-Germanic and Balto-Slavic. Its always been clear from the dynamic settlement history in east-central Europe.

No one has claimed Corded Ware is directly ancestral to Balto-Slavic or Germanic. That is impossible because those languages began and spread thousands of years after Corded Ware. That's the same as saying Corded Ware is directly ancestral to Russians. Corded Ware though is probably ancestral to both Balto-Slavic and Germanic languages. Corded Ware ended 4,000 years ago, then descendants of Corded Ware eventually spoke proto-Balto Slavic and proto-Germanic.IMO there must have been post or late Corded Ware expansions of R1a-Z282. Nonetheless Ra-Z282 comes from Corded Ware or his relatives.

Gravetto-Danubian
11-24-2015, 11:22 PM
No one has claimed Corded Ware is directly ancestral to Balto-Slavic or Germanic. .

Incorrect. Plenty of people here, and in academia, have claimed so. And I said proto-Germanic, not Germanic itself

http://www.anthrogenica.com/showthread.php?5775-Germanic-Admixture-in-Deutschland&p=119545#post119545

or

D Anthony p 306 "Corded Ware tumulus cemeteries in the east Carpathian foothills, a historic meeting through which dialects ancestral to the northern Indo-European languages (Germanic, Slavic, Baltic) began to spread among eastern Corded Ware groups"


Nonetheless Ra-Z282 comes from Corded Ware or his relatives

Who's "he" ?

Krefter
11-24-2015, 11:23 PM
Who's "he" ?

We're talking about Corded Ware Y DNA. So, it's he.

Gravetto-Danubian
11-24-2015, 11:27 PM
We're talking about Corded Ware Y DNA. So, it's he.

Yes, of course Z282 came from somewhere within the former CWC horizon
But the important detail is the earliest R1a - and whatever language they spoke - was in all likelihood replaced by later expansions, time (eg Z282 ? in MBA) and again (M458 and I2a1b in EMAs).

So the fact is statements like 'CWC is ancestral to Proto-Germanic and Balto-Slavic' is not particularly illuminating.

It's like saying that the earliest IE speakers in Britain spoke proto-Norse

Generalissimo
11-24-2015, 11:34 PM
But the important detail is the earliest R1a - and whatever language they spoke - was in all likelihood replaced by later expansions, time (eg Z282 ? in MBA) and again (M458 and I2a1b in EMAs).

These replacements in Y-DNA lineages were due to in-situ expansions as some clans become disproportionally more successful than others, so there's no reason to assume languages were replaced.

Tomenable
11-24-2015, 11:34 PM
Here some interesting thoughts about Proto-Indo-European language, "Out-Of-Africa" language and "Proto-Human" language:


https://www.youtube.com/watch?v=YqSGv8CnrWs

He makes some very good points which can apply also to expansion of PIEs:

https://www.youtube.com/watch?v=YqSGv8CnrWs#t=200

"(...) Let's imagine there is a tribe of people, say a 1000, they all speak a language they mutually understand, and then a 100 people leave to migrate somewhere else. That 100 take with them all the sounds of the language. Yes, they are taking away only a fraction [10%] of the genetic diversity, but not of course a fraction of the sounds. (...)"

So this is why PIE clans which migrated in distinct directions, took different Y-DNA with them, but the same language.

Here he talks about Proto-Indo-European language (from 6:00 onwards):

https://www.youtube.com/watch?v=YqSGv8CnrWs#t=360

================================================

About the "Out-Of-Africa" tribe:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3104569/

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3656025/


The “out of Africa” hypothesis proposes that a small group of Homo sapiens left Africa 80,000 years ago, spreading the mitochondrial haplotype L3 throughout the Earth.1–10 Little effort has been made to try to reconstruct the society and culture of the tribe that left Africa to populate the rest of the world.1

DMXX
11-24-2015, 11:39 PM
Firstly, a massive thank you to Srkz for continuing with his series of IBD outputs and distribution plots. They're always informative and I appreciate the effort he puts into providing us with another form of data to assist with our interpretations.

The CHG map does not offer any surprises and overlaps near-perfectly with various "West Asian" ADMIXTURE components from past runs (e.g. Dodecad K10a (http://1.bp.blogspot.com/-TK_t6TTXW20/T-pcZCZCzCI/AAAAAAAAAJ8/adNsHT6zV2A/s1600/Westasian.png)). The depreciation in Turkey and NW Iran-Azerbaijan probably comes from East Eurasian admixture in these regions. Indeed, this clearly accounts for the IBD deficiency in-between the Urals and South-Central Asia.

Bischon+Loschbour (WHG) has the usual expected pattern with very heavy segment sharing in Europe (the NE in particular) with a sharp demarcation at the traditional natural borders between Europe and Asia (the Ural and Caucasus mountains). What is interesting to note here is the trickle-down of segment affinity into South-Central Asia and various neighbouring populations (Uyghurs, Uttar Pradesh Brahmins, Iranians). Anyone familiar with the proposed development of the Indo-Iranians per the Pontic-Caspian steppe proposal will recognise it essentially traces the same route here. This is also consistent with the larger amount of WHG picked up in Eurogenes K7/8 among Sintashta and Andronovo compared to Yamnaya (whether or not it is direct WHG or by-proxy through EEF admixture is besides the point).

I've also been wondering about the status of WHG in South Asia, as some of Kurd's latest ADMIXTURE runs are showing significant (>10%) frequencies. There's something of a contradiction here (other calculators, including Eurogenes ones, didn't show this).

I'm not entirely certain what could be inferred from CHG-WHG if there is doubt (based on the f3 stats from Jones et al.) that CHG and WHG are part of the same clade relative to others. If there is a reason for this one, some guidance is much appreciated.

The CHG-Yamnaya distribution chart is highly illuminating. In short, it is displaying the IBD affinities to Kotias once the Yamnaya affinity is removed. Regions with downgraded affinities will derive more of their Kotias-like admixture from Yamnaya than another Kotias-like population. Reverse applies for upgraded affinities.
The situation in Asia looks quite mixed. It's clear that some populations (Uttar Pradesh Brahmins notably, also Iranians, Kalash, Pamiris and others) received a "top-up" in their total affinity to Kotias via a Yamnaya-like source. The rest of the Near-East and South Asia seems to be independent of this.
The Caucasus is also interesting. Note that Georgians still display a very high affinity, whereas that's downgraded in the North Caucasus. This could suggest rather limited Yamnaya gene-flow towards the South Caucasus, whereas the north received a fair bit (in keeping with various theories about the source of "teal" in the past).
Within Europe, one will observe substantial downgrading in the north. This clearly implies (as Davidski's concluded independently) that the majority of N/NE Europe's CHG-related admixture came directly from steppe groups.

If Srkz is reading this, it would be quite illuminating for those of us interested in the IBD situation in Asia if a combined WHG+Yamnaya+Andronovo+Sintashta chart could be created. Would give us a clear visual regarding what to expect when it comes to steppe admixture in Asia.

Gravetto-Danubian
11-24-2015, 11:42 PM
These replacements in Y-DNA lineages were due to in-situ expansions as some clans become disproportionally more successful than others, so there's no reason to assume languages were replaced.

Yes there is
Languages were continually evolving and being replaced - even if by very close "cousins"- until the most recent stage of prehistory (pre -Roman period) . That's because pre-literate societies lacked the apparatus to "conservatize" language; as well as that chiefdoms were constantly competeting, ousting and replacing; merging and then dissolving again

So I'm not proposing that Balto- Slavic or Germanic 'arrived' from Tanzania in 400 BC; but there is no 'proto-Germanic' to speak of in 2500 BC; nor is there any distinct ancestor to which it can be linked to in any linear fashion

alan
11-25-2015, 12:07 AM
There is another possibility about CHG. AFAIK the early wing of proto-Gravettian that went to the Caucasus also went through the Caucasus into south Russia. So perhaps this element would retreat north and south depending on the climate oscillation but remains essentially on the same sort of longitude either side. AFAIK the archaeology of both the Caucasus and south Russia for most of the Palaeolithic after a certain point in time and after the LGM is Gravettian derived until the Younger Dryas anyway. So perhaps CHG was an element that was always coming in and out of Russia and a minority of it survived into the Mesolithic. I am not saying continuous presence necessarily - just that some CHG was on scene in the mix among hunters north of the Caucasus as well as in Georgia etc and in the north it mixed in with WHG and ANE elements in the Mesolithic. There could have been pockets here and there where it was higher.

https://instaar.colorado.edu/uploads/people/155/hoffecker-2012-eshe-2.pdf

alan
11-25-2015, 12:38 AM
Firstly, a massive thank you to Srkz for continuing with his series of IBD outputs and distribution plots. They're always informative and I appreciate the effort he puts into providing us with another form of data to assist with our interpretations.

The CHG map does not offer any surprises and overlaps near-perfectly with various "West Asian" ADMIXTURE components from past runs (e.g. Dodecad K10a (http://1.bp.blogspot.com/-TK_t6TTXW20/T-pcZCZCzCI/AAAAAAAAAJ8/adNsHT6zV2A/s1600/Westasian.png)). The depreciation in Turkey and NW Iran-Azerbaijan probably comes from East Eurasian admixture in these regions. Indeed, this clearly accounts for the IBD deficiency in-between the Urals and South-Central Asia.

Bischon+Loschbour (WHG) has the usual expected pattern with very heavy segment sharing in Europe (the NE in particular) with a sharp demarcation at the traditional natural borders between Europe and Asia (the Ural and Caucasus mountains). What is interesting to note here is the trickle-down of segment affinity into South-Central Asia and various neighbouring populations (Uyghurs, Uttar Pradesh Brahmins, Iranians). Anyone familiar with the proposed development of the Indo-Iranians per the Pontic-Caspian steppe proposal will recognise it essentially traces the same route here. This is also consistent with the larger amount of WHG picked up in Eurogenes K7/8 among Sintashta and Andronovo compared to Yamnaya (whether or not it is direct WHG or by-proxy through EEF admixture is besides the point).

I've also been wondering about the status of WHG in South Asia, as some of Kurd's latest ADMIXTURE runs are showing significant (>10%) frequencies. There's something of a contradiction here (other calculators, including Eurogenes ones, didn't show this).

I'm not entirely certain what could be inferred from CHG-WHG if there is doubt (based on the f3 stats from Jones et al.) that CHG and WHG are part of the same clade relative to others. If there is a reason for this one, some guidance is much appreciated.

The CHG-Yamnaya distribution chart is highly illuminating. In short, it is displaying the IBD affinities to Kotias once the Yamnaya affinity is removed. Regions with downgraded affinities will derive more of their Kotias-like admixture from Yamnaya than another Kotias-like population. Reverse applies for upgraded affinities.
The situation in Asia looks quite mixed. It's clear that some populations (Uttar Pradesh Brahmins notably, also Iranians, Kalash, Pamiris and others) received a "top-up" in their total affinity to Kotias via a Yamnaya-like source. The rest of the Near-East and South Asia seems to be independent of this.
The Caucasus is also interesting. Note that Georgians still display a very high affinity, whereas that's downgraded in the North Caucasus. This could suggest rather limited Yamnaya gene-flow towards the South Caucasus, whereas the north received a fair bit (in keeping with various theories about the source of "teal" in the past).
Within Europe, one will observe substantial downgrading in the north. This clearly implies (as Davidski's concluded independently) that the majority of N/NE Europe's CHG-related admixture came directly from steppe groups.

If Srkz is reading this, it would be quite illuminating for those of us interested in the IBD situation in Asia if a combined WHG+Yamnaya+Andronovo+Sintashta chart could be created. Would give us a clear visual regarding what to expect when it comes to steppe admixture in Asia.

This makes a lot of sense to me. It has seemed clear for a while that Teal spread only partly with Yamnaya and also spread independently of Yamnaya. In some areas it one or the other and in other areas its both. The geography of this makes complete sense of the model of two waves of Teal, only one of which was carried by Yamnaya and related IE groups.

I have to add I felt the same about ANE. Some of it looks like its been carried through EHG element within Yamnaya but some of it looks like it spread independent of Yamnaya. Well actually we know for a fact some ANE spread independently of Yamnaya from Scandi hunters.

Personally I think all the components basically formed out of basal types once they had spread out and formed distinct groups for 10000 years or more. Then in the LGM this patterning got very intensified. These groups however came to mix again in variety of ways and times especially in the post-Younger Dryas phase. IMO EHG is some sort of mix of European Gravettian WHG types, east Gravettians of CHG type and ANE Siberians but this mixing and the formation of EHG is a post-Younger Dryas, basically Mesolithic thing. What is holding us back from concluding this is the weird pattern of ancient DNA sampling where we tend to have early upper Paleolithic hunters or the later Mesolithic ones around 6000-4000BC which misses out a crucial period in between. The lack of ancient DNA showing the changes in genetic makeup of hunters of c. 10000-6000BC is a particular problem as I think that was a melting pot period of different hunter groups in eastern and north-east Europe (I realise the Caucausus samples are an exception but they were off the beaten track of the mixing bowl).

Gravetto-Danubian
11-25-2015, 12:46 AM
There is another possibility about CHG. AFAIK the early wing of proto-Gravettian that went to the Caucasus also went through the Caucasus into south Russia. So perhaps this element would retreat north and south depending on the climate oscillation but remains essentially on the same sort of longitude either side. AFAIK the archaeology of both the Caucasus and south Russia for most of the Palaeolithic after a certain point in time and after the LGM is Gravettian derived until the Younger Dryas anyway. So perhaps CHG was an element that was always coming in and out of Russia and a minority of it survived into the Mesolithic. I am not saying continuous presence necessarily - just that some CHG was on scene in the mix among hunters north of the Caucasus as well as in Georgia etc and in the north it mixed in with WHG and ANE elements in the Mesolithic. There could have been pockets here and there where it was higher.

https://instaar.colorado.edu/uploads/people/155/hoffecker-2012-eshe-2.pdf

If I have understood correctly the guys at Eurogenes (RK, Matt, Dave), then here is what a preliminary 'Eurasian family tree' might look like:

(pardon the roughness of it, and there is still work to be done on South Asians. naturally, the tree will get more 'bushy' when we get aDNA from more areas (Middle East, South Asia, SEA).

6702

Ryukendo
11-25-2015, 09:48 AM
Hi, I am Ryukendo from the Eurogenes blog. Just in case any of you are wondering, here are the D-stats with which the tree is constructed:

________1)
Chimp Dai LBK_EN Kotias -0.0048 -1.133 303595
Chimp Papuan LBK_EN Kotias -0.008 -1.726 303595
Dai and and Papuan are closer to LBK_EN than to Kotias, but this is not significant.

________2)
Chimp Karitiana Kotias LBK_EN 0.0022 0.483 303595
Chimp Nganasan Kotias LBK_EN 0.0056 1.286 303595
Chimp Ket Kotias LBK_EN 0.0074 1.49 92303
Chimp Yakut Kotias LBK_EN 0.0043 1.055 303595
Chimp Ulchi Kotias LBK_EN 0.0045 1.07 303595
Chimp Lahu Kotias LBK_EN 0.0036 0.845 303595
There is not a significant pattern between ANE ancestry in mostly ENA populations and similarity to Kotias or LBK_EN; but once again all populations prefer LBK_EN nonsignificantly.

________3)
Chimp Kostenki14 LBK_EN Kotias -0.0101 -1.744 226149
The hypothesis that Kostenki forms a clean outgroup to Kotias and LBK_EN is not rejected; i.e. Kostenki might form a clean outgroup to Kotias and LBK_EN.

LBK_EN Kostenki14 Chimp Kotias -0.0329 -5.366 226149
Kotias Kostenki14 Chimp LBK_EN -0.0228 -3.763 226149
Kotias share a large amount of drift with LBK_EN to the exclusion of Kostenki.

________4)
The following stats are for tree-building:

__ i. Building a basic tree with Dai, Ust Ishim, Loschbour and Kostenki.

Chimp Ust-Ishim Dai Loschbour -0.0013 -0.232 466345
The tree of (Chimp)(Ust-Ishim(Dai, Loschbour)) is well supported. Loschbour does not have ancestry from outside the (Ust-Ishim(Other populations, Dai)) clade.

Chimp Ust-Ishim Kostenki14 Loschbour 0.0032 0.405 391567
The tree of (Chimp)(Ust-Ishim(Kostenki (Loschbour))) is well supported. Kostenki and Loschbour do not have ancestry from outside the clade formed by (Ust-Ishim(Other populations, Dai)).

I will call the clade (Ust Ishim(Dai, Kostenki, Loschbour)) conventional Crown Eurasian.

__ ii. Discovering ancestry outside of (Loschbour, Dai) in Kostenki, LBK_EN and Kotias

For Kostenki:

Chimp Dai Kostenki14 Loschbour 0.0183 2.887 392946
The tree of (Chimp)(Dai(Kostenki, Loschbour)) is rejected. The above statistic implies either Loschbour<---->Dai, in which case:

the tree is (Ust Ishim, ((Kostenki, WHG) Dai) and Dai contributes to WHG,

OR

Kostenki has ancestry from outside (Loschbour, Dai) which I will call Basal 'K'. Since we know from 4) i. that Kostenki does not have ancestry from outside (Ust Ishim(Other populations, Dai)), in other words Kostenki is 100% conventional Crown Eurasian as far as we can tell, I conclude that for this scenario Kostenki must have ancestry from Loschbour as well as from position K in:
(Ust Ishim(K(Loschbour, Dai)))

The two scenarios for Kostenki cannot be distinguished for now.

for LBK_EN:

Chimp Ust-Ishim LBK_EN Loschbour 0.0151 2.692 279178
Unlike the case of Kostenki, the tree of (Ust-Ishim(LBK_EN( Loschbour)) is rejected. The above statistic implies either Loschbour<---->Ust Ishim, which we know is very unlikely because Ust Ishim is equidistant from literally everyone in the paper where its genome was presented, OR it implies that LBK_EN contains ancestry from outside of conventional crown Eurasian clade (Ust Ishim, (other Eurasians, Dai)). I will call this ancestry 'B', as it was the first Basal Eurasian ghost that was postulated in Lazaridis, and it comes from the following position:

(B(Ust Ishim(Loschbour, Dai))))

The following stat, which has been replicated over and over in many places with similar EEF and East Eurasian populations, is automatically explained:
Chimp Dai LBK_EN Loschbour 0.0147 3.14 298813

for Kotias:

Chimp Ust-Ishim Kotias Loschbour 0.0305 4.665 404058
Like the case of LBK_EN and unlike the case for Kostenki, the tree of (Ust-Ishim(Kotias( Loschbour)) is rejected; much, much more strongly rejected than (Ust-Ishim(LBK_EN(Loschbour)) where we have 0.0151, Z=2.691. This can imply extensive Ust-Ishim<----->Loschbour, which I think is unlikely as talked about previously. OR it implies that either Kotias has even more ancestry from 'B' than LBK_EN does, or that he has ancestry from some population even more divergent than 'B'.

The following stats are automatically explained:
Chimp Dai Kotias Loschbour 0.0171 3.166 434671
Dai and Loschbour share drift to the exclusion of Kotias.

Chimp Loschbour Kostenki14 Kotias -0.0348 -4.554 343553
Loschbour and Kostenki share drift to the exclusion of Kotias.

Chimp MA1 Kostenki14 Kotias -0.0171 -2.142 271405
MA-1 and Kostenki share drift to the exclusion of Kotias.

Chimp Ust-Ishim Kostenki14 Kotias -0.0273 -3.758 373910
Ust Ishim and Kostenki share drift to the exclusion of Kotias.

Lets investigate the Basal in Kotias. For simplicity lets assume that Kotias either got its ancestry from outside the conventional Crown Eurasian clade (what we lump under the term 'Basal Eurasian') 100% from 'B', or 100% from a even more diverged population, 'C'(for Caucasus in the CHG acronym):
(C(B(Ust Ishim(Kostenki, Dai, WHG)))

From section 2) we know that Kotias and LBK_EN share drift to the exclusion of Kostenki. This is unexpected if the tree has the above shape, because if LBK_EN is WHG+'B', and Kotias is some other crown Eurasian+'C', Kotias will be close to equidistant between LBK_EN and Kostenki. However Kotias is strongly biased towards LBK_EN. From this, I deduce that CHG having 100% of its Basal ancestry from 'B', i.e. 100% the same Basal Eurasian as LBK_EN, is closer to the truth than it having 100% of its ancestry from 'C', (though of course some proportion of B and another proportion of even more diverged ancestry, with B predominant, is likely closer to the truth, because we cannot exclude the possibility of even more diverged ancestry in Kotias entirely).

Now lets investigate the Crown Eurasian in Kotias.

Chimp Kotias Ust-Ishim Kostenki14 0.0729 9.925 373910
This tells us that the Crown Eurasian in Kotias favours Kostenki over Ust Ishim. From the paper itself, we know that the crown Eurasian in Kotias favours West Eurasians over Dai significantly, and WHG over MA-1 and EHG over WHG non-significantly. We cannot exclude the possibility of crown eurasian from either Kostenki, Dai, MA-1, WHG, or EHG in Kotias completely, though crown eurasian in Kotias is likely to be dominated by West Eurasian ancestry (ENA ancestry is very unlikely for geographical reasons IMO), and unlikely to be dominated by any of Kostenki-like, MA1-like, WHG-like or EHG-like ancestries respectively.

Chimp Yoruba LBK_EN Kotias -0.005 -1.454 303595
West Eurasian input in Africans is closer to LBK_EN than to Kotias, but this is not significant.

Chimp Mota LBK_EN Kotias -0.0058 -1.194 276854
Mota approaches neither LBK_EN nor Kotias. This makes it unlikely that there was Mota-like ancestry in Kotias, unless it appears in LBK_EN as well in approximately the same proportion.

_______ 5.

Final Tree, either
Chimp (C(B(Ust Ishim(K(Loschbour, Dai)))))
Kostenki = Loschbour+K
LBK_EN = WHG+B
Kotias = Crown Eurasian (probably 100% from Kostenki/WHG/ANE, no ENA)+B with possibility from C

OR
Chimp (C(B(Ust Ishim(Kostenki, Loschbour, Dai))))
With Loschbour = Kostenki+Dai,
and the other two the same.


End of analysis.



Hopefully this will prove useful for futher analyses of the CHG genome and of SC Asian and S Asian ancestries. From the tree model in this analysis, it is possible to use stats (thanks to Davidski) to show that highly ASI/Low-caste populations in S India nevertheless have some amount of CHG ancestry, despite having low levels of IBS sharing with Kotias and sharing no more with Kotias than with LBK_EN in a head-on comparison.

Tomenable
11-25-2015, 10:04 AM
Was one of CHGs J2 ???: http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

Gravetto-Danubian
11-25-2015, 10:40 AM
Was one of CHGs J2 ???: http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

Off top of my head ; Satsurblia was J1b, maybe Kotias was J2. But I don't think the current pattern seen of J2 can be attributed to Mesolithic movements .

sweuro
11-25-2015, 12:59 PM
Was one of CHGs J2 ???: http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png
There is many areas of France with near 10% of J2, Serbia should be in the 5-10% range (not 1-5%),

DMXX
11-25-2015, 02:35 PM
Decent map (judging from the URL, the authors are Turkish).

The entire western fringe of Iran could theoretically be >50% J2-M172 (one old paper found that to be the frequency of "J-12f2"). As this SNP is upstream of J1-M267 and J2-M172, one cannot be certain how much of the "J-12f2" can be ascribed to either clade, aside from the strong likelihood (based on the existing trend across Iran) that J2 will predominate. In light of that, western Iran could potentially be the next "hotspot" for J2 after the highly isolated Caucasus locales. But one can forgive the authors for not including that to their chart (isn't feasible anyway). As someone who's created frequency maps in the past, I'm well aware of the fact that not everyone can be pleased!

YFull has J1-M267 and J2-M172's MRCA's at 18.4kya and 27.8kya respectively (http://www.yfull.com/tree/J/). We now have aDNA from one Mesolithic HG that's confirmed Y-DNA J. I agree with G-D that the bulk of modern J likely descends from those early Neolithic farming communities that enjoyed the survival advantages conferred from subsistence agriculture, but I'm sure we'll find a couple more offbeat Y-DNA J aDNA results (as well as some isolated branches, a la Soqotra's "J*" cases) in the coming years. :)

Ryukendo
11-25-2015, 03:26 PM
Alan, while we are on the topic of ancient populations such as CHG, I have a question for you. I have followed your archaeological explanations with interest :) and you seem to envision a scenario where most Eurasians today, or at least those in North and West Eurasia, derive their ancestry from a short pulse outward from the Middle East in a comparatively recent period.

However, I wonder how this can be reconciled with the picture we get from autosomal genetics, where the deepest split in Eurasia is not between different groups of ancestry that converge in the Middle East, but between the Middle East, where Basal Eurasian is found, vs everywhere else. How can the archaeology be reconciled with a paleolithic genetic landscape where the population just north of the Cauccasus in Siberia, EHG, has much, much more shared drift, and therefore share much more shared population history, with East Asian and Oceanian populations such as Onge, Papuan and Dai than with their CHG neighbours just across the mountain range?

Is it really possible to reconcile an archaeological scenario favouring expansion from the Levant, with the genetic trend in Paleolithic North and West Eurasia towards greater and greater similarity with present-day East Asians and Oceanians? Ust-Ishim and Oase, the oldest genomes, are very far from present-day East Asians (Even though they are already closer to East Asians than they are to Basal-carrying populations such as CHG, which we know from Satsurblia must have been widespread in the Middle East since a very ancient, pre-farming period), then Kostenki turns out even closer to East Asians than Ust_Ishim or Oase, and then WHG, ANE, and EHG are yet another step closer to East Asians/Oceanians than Kostenki.

Later on, when a population from the Middle East does show up in Europe, its overall effect is to decrease affinity to East Asians, and decrease affinity to even Ust_Ishim, implying that, by 45 kya in the middle of the Aurignacian, Middle Easterners were already separated for a long time from the common population that gave rise to Hunter Gatherers all across Eurasia. How can this be explained?

Are there any archaeological correlates which fit the autosomal scenario?

ADW_1981
11-25-2015, 03:45 PM
There is many areas of France with near 10% of J2,

Highly doubtful. Source? You might get one off, small datasets that misrepresent the overall picture from a small section of a village. I am extremely skeptical that would be the reality even on a provincial level.

Chad Rohlfsen
11-25-2015, 03:47 PM
If I have understood correctly the guys at Eurogenes (RK, Matt, Dave), then here is what a preliminary 'Eurasian family tree' might look like:

(pardon the roughness of it, and there is still work to be done on South Asians. naturally, the tree will get more 'bushy' when we get aDNA from more areas (Middle East, South Asia, SEA).

6702


You have one mistake on your tree. I said at the beginning, and Dstats show that it is Crown WEST Eurasian in CHG, not Crown Eurasian, as stats show CHG is significantly closer to WHG, EHG, and ANE, than to Dai. Other than that, it looks good.

J Man
11-25-2015, 04:39 PM
Was one of CHGs J2 ???: http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

http://www.haplogruplar.com/wp-content/uploads/2015/07/J2-Y-DNA-Haplogroup-Map-J2-M172-Map-J2-Haplogrubu-Haritasi-v3.png

Yes the Mesolithic CHG sample from Kotias Klde is J2a.

J Man
11-25-2015, 04:42 PM
Decent map (judging from the URL, the authors are Turkish).

The entire western fringe of Iran could theoretically be >50% J2-M172 (one old paper found that to be the frequency of "J-12f2"). As this SNP is upstream of J1-M267 and J2-M172, one cannot be certain how much of the "J-12f2" can be ascribed to either clade, aside from the strong likelihood (based on the existing trend across Iran) that J2 will predominate. In light of that, western Iran could potentially be the next "hotspot" for J2 after the highly isolated Caucasus locales. But one can forgive the authors for not including that to their chart (isn't feasible anyway). As someone who's created frequency maps in the past, I'm well aware of the fact that not everyone can be pleased!

YFull has J1-M267 and J2-M172's MRCA's at 18.4kya and 27.8kya respectively (http://www.yfull.com/tree/J/). We now have aDNA from one Mesolithic HG that's confirmed Y-DNA J. I agree with G-D that the bulk of modern J likely descends from those early Neolithic farming communities that enjoyed the survival advantages conferred from subsistence agriculture, but I'm sure we'll find a couple more offbeat Y-DNA J aDNA results (as well as some isolated branches, a la Soqotra's "J*" cases) in the coming years. :)

It will be interesting to see if any J2 turns up among the very first Pre-Potter Neolithic A and Pre-Pottery Neolithic B farmers of the Fertile Crescent once and if they are tested.

alan
11-25-2015, 05:08 PM
Alan, while we are on the topic of ancient populations such as CHG, I have a question for you. I have followed your archaeological explanations with interest :) and you seem to envision a scenario where most Eurasians today, or at least those in North and West Eurasia, derive their ancestry from a short pulse outward from the Middle East in a comparatively recent period.

However, I wonder how this can be reconciled with the picture we get from autosomal genetics, where the deepest split in Eurasia is not between different groups of ancestry that converge in the Middle East, but between the Middle East, where Basal Eurasian is found, vs everywhere else. How can the archaeology be reconciled with a paleolithic genetic landscape where the population just north of the Cauccasus in Siberia, EHG, has much, much more shared drift, and therefore share much more shared population history, with East Asian and Oceanian populations such as Onge, Papuan and Dai than with their CHG neighbours just across the mountain range?

Is it really possible to reconcile an archaeological scenario favouring expansion from the Levant, with the genetic trend in Paleolithic North and West Eurasia towards greater and greater similarity with present-day East Asians and Oceanians? Ust-Ishim and Oase, the oldest genomes, are very far from present-day East Asians (Even though they are already closer to East Asians than they are to Basal-carrying populations such as CHG, which we know from Satsurblia must have been widespread in the Middle East since a very ancient, pre-farming period), then Kostenki turns out even closer to East Asians than Ust_Ishim or Oase, and then WHG, ANE, and EHG are yet another step closer to East Asians/Oceanians than Kostenki.

Later on, when a population from the Middle East does show up in Europe, its overall effect is to decrease affinity to East Asians, and decrease affinity to even Ust_Ishim, implying that, by 45 kya in the middle of the Aurignacian, Middle Easterners were already separated for a long time from the common population that gave rise to Hunter Gatherers all across Eurasia. How can this be explained?

Are there any archaeological correlates which fit the autosomal scenario?

I suppose the difference is there are older groups in east Asia from a separate wave pre-dating the basal 45000-4000BC moves into Europe and north Asia. IMO EHG is a hybrid that took place in the Europe steppes sort of area c. 9500-8000BC when Siberian ANE mixed with WHG. This mix simply didnt happen in the Levant and west Anatolia until after the farmers headed to Europe.

The archaeologicy would indicated multiple pulses from basically the same area in north Levant c. 45000BC (Emiran/Bohunician/earliest Siberian upper Paleolithic), then 40000BC (Ahmarian, Aurignacian), 38000BC (later Ahmarian, Caucauses/south Russian proto-Gravettian/east Gravettian), 30000BC (late Ahmarian, classic Gravettian). The archaeology seems reasonably clear on this.

Megalophias
11-25-2015, 05:26 PM
The entire western fringe of Iran could theoretically be >50% J2-M172 (one old paper found that to be the frequency of "J-12f2"). As this SNP is upstream of J1-M267 and J2-M172, one cannot be certain how much of the "J-12f2" can be ascribed to either clade, aside from the strong likelihood (based on the existing trend across Iran) that J2 will predominate. In light of that, western Iran could potentially be the next "hotspot" for J2 after the highly isolated Caucasus locales.

From Grugni et al:
Khuzestan Arabs - 25% J2 (58% J)
Lorestan Lurs - 18% J2 (24% J)
Kurdestan Kurds - 24% J2 (29% J)
Azerbaijan Assyrians - 16% J2 (33% J)
Gharbi Azeris - 22% J2 (27% J)

So around 15-25% J2, which is almost entirely J2a. Most have little J1 except of course the Arabs who have loads of J1a1a2b-P58, and the Assyrians who have a good deal of J1*. The J2a seems to be quite diverse (within the limited number of SNPs tested), with significant amounts of J2a1a1a1a1-M47 (old J2a1a), J2a1a1a1b1-M67 (old J2a1b), J2a1a1b1-L24 (old J2a1h), J2a1a*-Page55 (old J2a1*), and J2a*(xJ2a1).

Unfortunately full sequences are hard to come by, all I know of is an unspecified Iranian J2a1a1b1a3-F3133 (subclade of L24) and an Assyrian J2a1a1a1a*-Z6065 (sister branch to M47) in Karmin et al.

Ryukendo
11-25-2015, 05:46 PM
I suppose the difference is there are older groups in east Asia from a separate wave pre-dating the basal 45000-4000BC moves into Europe and north Asia. IMO EHG is a hybrid that took place in the Europe steppes sort of area c. 9500-8000BC when Siberian ANE mixed with WHG. This mix simply didnt happen in the Levant and west Anatolia until after the farmers headed to Europe.

The archaeologicy would indicated multiple pulses from basically the same area in north Levant c. 45000BC (Emiran/Bohunician/earliest Siberian upper Paleolithic), then 40000BC (Ahmarian, Aurignacian), 38000BC (later Ahmarian, Caucauses/south Russian proto-Gravettian/east Gravettian), 30000BC (late Ahmarian, classic Gravettian). The archaeology seems reasonably clear on this.

Hi alan, the issue remains that WHG, ANE and EHG alike are far more similar to East Asians and Oceanians than they are to Middle Eastern HGs such as Satsurblia, to Anatolia Neolithic, or present day Middle Easterners. In fact the statistics indicate that the common ancestral population of East Asians, Oceanians, Onge, WHG, ANE and EHG had broken off from ancient or modern Middle Easterners earlier than the time of Ust ishim. The population stayed isolated from Middle Easterners for some time, with the result that their descendants incl. WHG, East Asians etc. share much more drift with, and are more similar to, each other than they have drift with/ are similar to ancient or modern middle easterners. And this is ignoring the fact that kostenki did have to become closer to East Asians and Oceanians to create the ANE, WHG and EHG that we are so familiar with today.

Perhaps the transmission of cultures cannot be used to draw inferences about population movements in this time period, though judging from more recent periods this would be highly atypical. Or perhaps the interpretation in archaeology is biased by taphonomic factors, e.g. extremely poor preservation in much of India, S China or Southeast Asia compared to exceptional preservation in the Middle East.

sweuro
11-25-2015, 06:21 PM
Highly doubtful. Source? You might get one off, small datasets that misrepresent the overall picture from a small section of a village. I am extremely skeptical that would be the reality even on a provincial level.
What sample size you used ?

For starters, the study of Bekada et al. 2013 has a sample of 776 for France, and it has 6.7% of J2.
The study of King et al. 2011 has 10% J2 for the sample of Provence.
The study of Ramos-Luis et al. 2013 has the following percentages : Alsace : 8.8%, Midi-Pyrenees : 7.5%, Auvergene : 7.9%, Provence-Alpes-Cote-D'azur: 6.7%

Gravetto-Danubian
11-25-2015, 11:01 PM
I suppose the difference is there are older groups in east Asia from a separate wave pre-dating the basal 45000-4000BC moves into Europe and north Asia. IMO EHG is a hybrid that took place in the Europe steppes sort of area c. 9500-8000BC when Siberian ANE mixed with WHG. This mix simply didnt happen in the Levant and west Anatolia until after the farmers headed to Europe.

The archaeologicy would indicated multiple pulses from basically the same area in north Levant c. 45000BC (Emiran/Bohunician/earliest Siberian upper Paleolithic), then 40000BC (Ahmarian, Aurignacian), 38000BC (later Ahmarian, Caucauses/south Russian proto-Gravettian/east Gravettian), 30000BC (late Ahmarian, classic Gravettian). The archaeology seems reasonably clear on this.

Further to Alan's & RKs point, even WHG shows marked separation from BE/ EF; which his odd given that the earliest AMH 'transitional' industries (Bacho-Kirian, Bohunician) are linked to the Near East (as Alan aptly highlights), as is the proto-Aurignacian. I can imagine a couple of confounding/ explanatory factors:

1) flow from ENA to the west (perhaps riding in with the movement of R* lineages northwest) helped differentiate north Eurasian into the 'clustered-cline' seen in WHG - SHG - EHG

2) a yet un-appreciated secondary dispersal point of UP industries (especially Gravettian) to Europe and north Eurasia. Ie not from the Middle Ease, but somewhere in NW Iran/ south Caucasus

3) turnover in the Near East itself, with loss of proto-north Eurasian signatures.

4) the LGM had the most affect to separated the southern Levant from proto-North Eurasians, whilst the proto-NEs had ongoing contact. In turn, the north Eurasians lost many 'basal' signatures due to drift

(?)

Gravetto-Danubian
11-25-2015, 11:07 PM
Hi alan, the issue remains that WHG, ANE and EHG alike are far more similar to East Asians and Oceanians than they are to Middle Eastern HGs such as Satsurblia, to Anatolia Neolithic, or present day Middle Easterners. In fact the statistics indicate that the common ancestral population of East Asians, Oceanians, Onge, WHG, ANE and EHG had broken off from ancient or modern Middle Easterners earlier than the time of Ust ishim. The population stayed isolated from Middle Easterners for some time, with the result that their descendants incl. WHG, East Asians etc. share much more drift with, and are more similar to, each other than they have drift with/ are similar to ancient or modern middle easterners. And this is ignoring the fact that kostenki did have to become closer to East Asians and Oceanians to create the ANE, WHG and EHG that we are so familiar with today.

Perhaps the transmission of cultures cannot be used to draw inferences about population movements in this time period, though judging from more recent periods this would be highly atypical. Or perhaps the interpretation in archaeology is biased by taphonomic factors, e.g. extremely poor preservation in much of India, S China or Southeast Asia compared to exceptional preservation in the Middle East.

Very good points RK. The Genetic data does not quite Gel with the archaeological evidence - hence my initial posting of the tree based on yours and Davidski's analysis in reply to Alan's ideas.

As per my reply to Alan above, point (2) might be most pertinent ?

parasar
11-25-2015, 11:31 PM
You have one mistake on your tree. I said at the beginning, and Dstats show that it is Crown WEST Eurasian in CHG, not Crown Eurasian, as stats show CHG is significantly closer to WHG, EHG, and ANE, than to Dai. Other than that, it looks good.

The tree looks fine to me. The closeness has to explained by a TreeMix type tree with not only splits but subsequent, more recent interactions following the splits.

Edit: There is an additional basal split that goes into K14 I believe.

Chad Rohlfsen
11-26-2015, 12:41 AM
The tree looks fine to me. The closeness has to explained by a TreeMix type tree with not only splits but subsequent, more recent interactions following the splits.

Edit: There is an additional basal split that goes into K14 I believe.

It's not Crown Eurasian. If it were, it would be equally related to Dai and WHG, which it obviously is not.

DMXX
11-26-2015, 08:25 AM
From Grugni et al:
Khuzestan Arabs - 25% J2 (58% J)
Lorestan Lurs - 18% J2 (24% J)
Kurdestan Kurds - 24% J2 (29% J)
Azerbaijan Assyrians - 16% J2 (33% J)
Gharbi Azeris - 22% J2 (27% J)

So around 15-25% J2, which is almost entirely J2a. Most have little J1 except of course the Arabs who have loads of J1a1a2b-P58, and the Assyrians who have a good deal of J1*. The J2a seems to be quite diverse (within the limited number of SNPs tested), with significant amounts of J2a1a1a1a1-M47 (old J2a1a), J2a1a1a1b1-M67 (old J2a1b), J2a1a1b1-L24 (old J2a1h), J2a1a*-Page55 (old J2a1*), and J2a*(xJ2a1).

Unfortunately full sequences are hard to come by, all I know of is an unspecified Iranian J2a1a1b1a3-F3133 (subclade of L24) and an Assyrian J2a1a1a1a*-Z6065 (sister branch to M47) in Karmin et al.

Thanks for compiling this. As you're probably aware, Grugni et al. is the best region-specific breakdown we have of Iranian Y-DNA. Nice to see it cited from the get go. :)

I was referring earlier to Quintana-Murci et al., a paper from 2004 (link (https://www.familytreedna.com/pdf/Quintana-Murci-Iran.pdf)). The paper showcased frequency data for J-12f2 (their "HG 9") and R1a1-SRY1532 (their "HG 3"). Please note J-12f2 exceeds 50% in their Caspian samples. The peak within Iran was among "Zagros" populations (text does not clearly state which populations these are unfortunately). I'll have to dig through Roewer et al. again to get a feel for the (predicted) J subclades among Gilakis and Mazandaranis (Iran's main Caspian populations).

A side note, but this was also the paper that refuted the oft-repeated conclusion of Wells et al. concerning the near-absence of Y-DNA R1a1a in Iran. The general patterns observed there are corroborated by Grugni et al.

alan
11-26-2015, 10:14 AM
Hi alan, the issue remains that WHG, ANE and EHG alike are far more similar to East Asians and Oceanians than they are to Middle Eastern HGs such as Satsurblia, to Anatolia Neolithic, or present day Middle Easterners. In fact the statistics indicate that the common ancestral population of East Asians, Oceanians, Onge, WHG, ANE and EHG had broken off from ancient or modern Middle Easterners earlier than the time of Ust ishim. The population stayed isolated from Middle Easterners for some time, with the result that their descendants incl. WHG, East Asians etc. share much more drift with, and are more similar to, each other than they have drift with/ are similar to ancient or modern middle easterners. And this is ignoring the fact that kostenki did have to become closer to East Asians and Oceanians to create the ANE, WHG and EHG that we are so familiar with today.

Perhaps the transmission of cultures cannot be used to draw inferences about population movements in this time period, though judging from more recent periods this would be highly atypical. Or perhaps the interpretation in archaeology is biased by taphonomic factors, e.g. extremely poor preservation in much of India, S China or Southeast Asia compared to exceptional preservation in the Middle East.

The bamboo belt is a particular problem because all through it flint appears to have been largely abandoned except the crudest tools and bamboo used instead.

alan
11-26-2015, 10:27 AM
Further to Alan's & RKs point, even WHG shows marked separation from BE/ EF; which his odd given that the earliest AMH 'transitional' industries (Bacho-Kirian, Bohunician) are linked to the Near East (as Alan aptly highlights), as is the proto-Aurignacian. I can imagine a couple of confounding/ explanatory factors:

1) flow from ENA to the west (perhaps riding in with the movement of R* lineages northwest) helped differentiate north Eurasian into the 'clustered-cline' seen in WHG - SHG - EHG

2) a yet un-appreciated secondary dispersal point of UP industries (especially Gravettian) to Europe and north Eurasia. Ie not from the Middle Ease, but somewhere in NW Iran/ south Caucasus

3) turnover in the Near East itself, with loss of proto-north Eurasian signatures.

4) the LGM had the most affect to separated the southern Levant from proto-North Eurasians, whilst the proto-NEs had ongoing contact. In turn, the north Eurasians lost many 'basal' signatures due to drift

(?)

I think a lot of that may be right. The key problem remains of course we just have a tiny scattering of ancient DNA from Eurasia covering the period 45000-6000BC with an ice age and younger dryas in between to mess it all up.

Iran is an interesting point. It had Emiran type stuff passing through it to the stans and into Siberia c, 45000BC and then it also had Aurignacian in Zagros after 40000BC and proto-Gravettian in the Caucasus next door after 38000BC. So a mixed population could have arisen in that sort of area and if the samples are not early enough and dont straddle both sides of these waves we wont see the process.

As you noted a four wave model from Levant perhaps 45000, 40000, 38000 and 30000BC doesnt mean the Levant/Anatolia was genetically static. In fact there is reason to believe that by 30000BC a population around the Levant/Anatolia (apparently a subset of the Ahmarian) had become WHG-like despite surely having originally been basal.

Krefter
11-26-2015, 12:02 PM
Hi alan, the issue remains that WHG, ANE and EHG alike are far more similar to East Asians and Oceanians than they are to Middle Eastern HGs such as Satsurblia, to Anatolia Neolithic, or present day Middle Easterners.

They are much closer to Middle Easterners than to East Asians. The Middle East isn't and wasn't 100% Basal Eurasian. Plus Basal Eurasian is overrated, I doubt anyone is more than 25%, if it is a real type of ancestry.

Heber
11-26-2015, 12:35 PM
A new paper from Mike Hammers Lab. High levels of J1 (M267) and R1b-L23* found in Daghestan and Caucasus highland populations-

Coevolution of genes and languages and high levels of population structure among the highland populations of Daghestan
Tatiana M Karafet, Kazima B Bulayeva, Johanna Nichols, Oleg A Bulayev, Farida Gurgenova, Jamilia Omarova, Levon Yepiskoposyan, Olga V Savina, Barry H Rodrigue and Michael F Hammer

http://www.nature.com/jhg/journal/vaop/ncurrent/suppinfo/jhg2015132s1.html

rms2
11-26-2015, 01:02 PM
A new paper from Mike Hammers Lab. High levels of J1 (M267) and R1b-L23* found in Daghestan and Caucasus highland populations-

Coevolution of genes and languages and high levels of population structure among the highland populations of Daghestan
Tatiana M Karafet, Kazima B Bulayeva, Johanna Nichols, Oleg A Bulayev, Farida Gurgenova, Jamilia Omarova, Levon Yepiskoposyan, Olga V Savina, Barry H Rodrigue and Michael F Hammer

http://www.nature.com/jhg/journal/vaop/ncurrent/suppinfo/jhg2015132s1.html

That's not ancient y-dna though, is it?

Ryukendo
11-26-2015, 03:41 PM
They are much closer to Middle Easterners than to East Asians. The Middle East isn't and wasn't 100% Basal Eurasian. Plus Basal Eurasian is overrated, I doubt anyone is more than 25%, if it is a real type of ancestry.

But that is because of recent gene flow from W Eurasian HGs into Middle Easterns, which of course pulls them together. It is not the other way round, which would be necessary to make the archaeological scenario work.

The rest of the points are not necessarily true. From f4 ratio estimation in Lazaridis et al, Stuttgart is ~44% from outside the clade formed by Onge, Loschbour and MA-1; aka he is 44% Basal Eurasian+K, both of which reduce closeness to East Asians and European Hunter Gatherers, and his ancestors in Anatolia Neolithic were no doubt even more Basal still, judging by their position on the PCA. Kotias is yet again more Basal than EEF. Most middle Easterns are further south of the line formed by EEFs annd Kotias in a PCA, and Stuttgart can be modeled as Armenian+WHG(!), so MEs are almost certainly more Basal than these ancient pops.

The stats you posted before do not actually show that the Middle East is mostly UHG.

Nevertheless, no matter how little Basal Eurasian Middle Easterners have, this does not invalidate the fact that Basal Eurasian ancestry much have existed somewhere in Eurasia in isolation from Crown Eurasians by the time of Ust Ishim; neither does it invalidate the fact that a 100% Basal Eurasian population must have existed at some period. Formal statistics do not produce inferences as flighty as ADMIXTURE clusters. When two populations do not share drift to the magnitude that Basal and Crown do not, two separate ancestral groups 100% their own thing must have really existed.

Kurd
11-26-2015, 06:23 PM
Fixation distances; WHG, EHG, CHG vs Anatolian Farmers & E Asians taken from Eurasia 11 Admixture run. WHG, EHG, & CHG have more overlapping allele frequencies with Anatolian Farmers, who tend to be closer to basal than Stuttgart & Iceman because of the latter's elevated WHG, than WHG, EHG, & CHG have with E Asians. I will do qpADM and Dstat analysis in a couple of days.



FST/ EURASIA K11
ANATOLIAN FARMERS
E ASIANS/SIBERIAN
AMERINDIAN


WHG
0.109
0.142
0.23


EHG
0.123
0.155
0.23


CHG
0.087
0.129
0.22


ANATOLIAN FARMERS
0
0.154
0.25


E ASIAN/SIBERIAN
0.154
0
0.18





POP
Papuan
Neolthic_Anatolian_Farmer
S_Indian
EHG
Kalash
E_Asian
Amerindian
E_African/SW Asian
CHG
W_African
WHG


Stuttgart
0.0%
57.3%
0.0%
0.0%
0.0%
0.0%
0.0%
4.9%
13.1%
0.0%
24.7%


Iceman
0.0%
57.9%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
2.3%
2.0%
37.8%


BedouinA
0.7%
23.2%
0.8%
0.8%
1.0%
0.5%
0.6%
26.1%
35.7%
5.8%
4.9%


Georgian
1.0%
20.7%
1.0%
5.2%
4.1%
0.6%
0.9%
3.7%
54.6%
0.1%
8.3%


Lithuanian
0.3%
12.0%
0.4%
17.6%
5.7%
0.6%
0.8%
0.3%
12.4%
0.0%
49.8%

Padre Organtino
11-26-2015, 06:47 PM
If this admixture test does not lie we are seeing a certain substructure among Neolithic Euro farmers with some of them clearly carrying a non-negligeable amount of CHG-liek ancestry.

Krefter
11-26-2015, 09:21 PM
But that is because of recent gene flow from W Eurasian HGs into Middle Easterns, which of course pulls them together. It is not the other way round, which would be necessary to make the archaeological scenario work.

It wasn't recent. 13,000 years ago and CHG had lots of it, maybe over 50%. For all we know WHG and Kostienki14 came from West Asia. You were writing that WHG is closer to East Asians than to modern Middle Easterners which isn't true.

The rest of the points are not necessarily true. From f4 ratio estimation in Lazaridis et al, Stuttgart is ~44% from outside the clade formed by Onge, Loschbour and MA-1; aka he is 44% Basal Eurasian+K, both of which reduce closeness to East Asians and European Hunter Gatherers, and his ancestors in Anatolia Neolithic were no doubt even more Basal still, judging by their position on the PCA. Kotias is yet again more Basal than EEF. Most middle Easterns are further south of the line formed by EEFs annd Kotias in a PCA, and Stuttgart can be modeled as Armenian+WHG(!), so MEs are almost certainly more Basal than these ancient pops.


The stats you posted before do not actually show that the Middle East is mostly UHG.

How can we explain Bedouin being slightly closer to Loschbour than to CHG, Iranian? Or all West Asians, except a few, being closer to WHG than to Syrians? To me it looks like Middle Easterners are mostly UHG, and any significant exotic admixture(African, East Asian, South Asian) makes them closer to Loschbour than to each other.

Ryukendo
11-27-2015, 07:09 AM
It wasn't recent. 13,000 years ago and CHG had lots of it, maybe over 50%. For all we know WHG and Kostienki14 came from West Asia. You were writing that WHG is closer to East Asians than to modern Middle Easterners which isn't true.

The rest of the points are not necessarily true. From f4 ratio estimation in Lazaridis et al, Stuttgart is ~44% from outside the clade formed by Onge, Loschbour and MA-1; aka he is 44% Basal Eurasian+K, both of which reduce closeness to East Asians and European Hunter Gatherers, and his ancestors in Anatolia Neolithic were no doubt even more Basal still, judging by their position on the PCA. Kotias is yet again more Basal than EEF. Most middle Easterns are further south of the line formed by EEFs annd Kotias in a PCA, and Stuttgart can be modeled as Armenian+WHG(!), so MEs are almost certainly more Basal than these ancient pops.



How can we explain Bedouin being slightly closer to Loschbour than to CHG, Iranian? Or all West Asians, except a few, being closer to WHG than to Syrians? To me it looks like Middle Easterners are mostly UHG, and any significant exotic admixture(African, East Asian, South Asian) makes them closer to Loschbour than to each other.


Krefter, you have been misinterpreting the stats that Davidski gave you. All Middle Easterners favour each other very consistently over Loschbour, except for Saudi, Bedouin and Syrian, and of course they screw up the stats because of African admixture.

Stats with Chimp as outgroup and African in 2nd or 3rd postion, e.g.
Chimp Kotias LBK_EN Mota -0.3051 -66.356 276854

routinely produce Z scores and D-distances literally an order of magnitude greater than the intra-eurasian variation we are used to seeing, e.g.

Chimp Dai Kotias Loschbour 0.0171 3.166 434671

They branched off that far back, and their drift paths are so long, that the length of their drift path *10% is more than sufficient to screw up the results.

There is no real evidence that Middle Easterners are mostly UHG.

Basal Eurasian and Crown Eurasian were already distinct more than 45 ky ago. In that scale 13kya is recent.

vettor
11-27-2015, 07:23 AM
That's not ancient y-dna though, is it?

no

but interesting that Dagestan has a clear majority of K mtDna over all the other areas

I can tell because the 4 x T-P327 ( ydna ) in Palestine are from one family and noted in Isogg T as a private marker

Krefter
11-27-2015, 08:01 AM
Krefter, you have been misinterpreting the stats that Davidski gave you. All Middle Easterners favour each other very consistently over Loschbour, except for Saudi, Bedouin and Syrian, and of course they screw up the stats because of African admixture.

Stats with Chimp as outgroup and African in 2nd or 3rd postion, e.g.
Chimp Kotias LBK_EN Mota -0.3051 -66.356 276854

I know it's African admixture that makes other West Asians more distant from each other than Loschbour. But South Asian admixture makes a similar affect as African Admixture. Most West Asians are very slightly closer to Iranians in D-stats than to Loshbour. That's good evidence West Asians are mostly UHG. South Asian isn't nearly as distant from Loschbour as Africans, yet it still makes a big affect,


Basal Eurasian and Crown Eurasian were already distinct more than 45 ky ago. In that scale 13kya is recent.

But You were saying UHG ancestry came to the Middle East recently. There's nothing to refute or support this.

Ryukendo
11-27-2015, 08:28 AM
Krefter, where are you getting your information? Except for those three populations mentioned previously, all middle easterners are much closer to Iranian than Loschbour is.

Chimp Iranian Loschbour Syrian -0.007 -1.908 503209
Chimp Iranian Loschbour Cypriot 0.0171 4.789 503209
Chimp Iranian Loschbour Saudi -0.0043 -1.161 503209
Chimp Iranian Loschbour Georgian 0.0214 6.05 503209
Chimp Iranian Loschbour BedouinB -0.0086 -2.381 503209
Chimp Iranian Loschbour Assyrian 0.0228 5.146 113202
Chimp Iranian Loschbour Turkish_Istanbul 0.0146 4.156 503209
Chimp Iranian Loschbour Lezgin 0.0194 5.48 503209

If two populations do not share drift, ie are 'split', then by definition there is no gene flow. Furthermore crown eurasian did not come into the middle east until afyer the kostenki-likes were replaced by loschbour and mal'ta-likes at the earliest.

K33
11-27-2015, 06:23 PM
Another question comes to mind, that I find interesting in the light of the new evidence we have. Are South-west asians(Gulf-arabs, semitic speaking desert nomads) descendents of neolithic farmers with east-african type admixture, or an entirely seperate but related west-eurasian population?Levantines (and Arabs in general) in this test score much higher for the CHG component than one would expect. The old ANE K7 calculator for example appeared to attribute >75% of Levantine autosomal DNA to the "Early Neolithic Farmer" component and only ~ 15% combined between WHG/ANE-- implying a virtually unbroken continuity between Fertile Crescent/Anatolian-type farmers and modern Levantines.

However in the new test the CHG component cannibalizes a considerable proportion of the ENF (even accounting for the apparent overlap between ENF and modern East African, which seems less pronounced in the ANE_K7 calc due to its more archaic EA sample). Obviously this effect was even more pronounced in Kurds, who had large portions of their ancestry scored as "Neolithic Farmer" in the old tests simple because CHG didn't "fit" the WHG or EHG models.

Re: Levantines, averaging the Syrian and Lebanese components from the spreadsheet (https://drive.google.com/file/d/0B-ObXiVfL-RzOEpDQ2U3MTJzNTA/view) produces an ENF/CHG/EastAfrican mixture like this: 24%/37%/18%. (Even Bedouins and Saudis surprisingly score ~35% CHG)

The combined EHG/WHG is still only about 10% for Levantines, so the CHG certainly wasn't contributed by Indo-Aryans like the Mittani etc (Mittani's likely closest approximation, the Sintashta sample, came up ~ 69% EHG/WHG & only 22% CHG), but rather directly from the CHG refugia descendants themselves.

While its not always wise making simple associations between autosomal data and y-dna haplogroups, with the finding of J2 in both CHG samples one can't but help correlate ENF's spread with haplogroup G2a, CHG with haplogroup J2, and EA with haplogroup E1b1. G2a is typically associated with the spread of agriculture, while J2 hunter-gatherers appear to have gradually adopted herding.

It seems likely that J2-bearing pastoralist-raiders (descendants of the CHG refugia population) descended from their mountains throughout the Near East after the LGM and engaged in a protracted back-and-forth battle with farmers-- perhaps lasting for multiple millenia throughout the Neolithic-- before gradually at least partially converting to agriculture and integrating with the local ENF population. This early, pre-Bronze Age dispersal would explain the ubiquitous ~35% CHG admixture observed throughout Mesopotamia, the Levant, Anatolia, and even the Arabian penninsula.

The bottom line is that we need many more genomes from the early, middle, late neolithic and bronze age Middle East to track the spread of ENF and CHG admixture.

Agamemnon
11-27-2015, 06:31 PM
Levantines (and Arabs in general) in this test score much higher for the CHG component than one would expect. The old ANE K7 calculator for example appeared to attribute >75% of Levantine autosomal DNA to the "Early Neolithic Farmer" component and only ~ 15% combined between WHG/ANE-- implying a virtually unbroken continuity between Fertile Crescent/Anatolian-type farmers and modern Levantines.

However in the new test the CHG component cannibalizes a considerable proportion of the ENF (even accounting for the apparent overlap between ENF and modern East African, which seems less pronounced in the ANE_K7 calc due to its more archaic EA sample). Obviously this effect was even more pronounced in Kurds, who had large portions of their ancestry scored as "Neolithic Farmer" in the old tests simple because CHG didn't "fit" the WHG or EHG models.

Re: Levantines, averaging the Syrian and Lebanese components from the spreadsheet (https://drive.google.com/file/d/0B-ObXiVfL-RzOEpDQ2U3MTJzNTA/view) produces an ENF/CHG/EastAfrican mixture like this: 24%/37%/18%. (Even Bedouins and Saudis surprisingly score ~35% CHG)

The combined EHG/WHG is still only about 10% for Levantines, so the CHG certainly wasn't contributed by Indo-Aryans like the Mittani etc (Mittani's likely closest approximation, the Sintashta sample, came up ~ 69% EHG/WHG & only 22% CHG), but rather directly from the CHG refugia descendants themselves. The timeline for this partial replacement of the ENF population (presumably from people originating in the Caucasus, or possibly the Iranian plateau) is unclear.... the other possibility is that farming never really did dominate the Near East as rapidly or as completely as once thought, despite the obvious advantage in population growth induced by agriculture.

While its not always wise making simple associations between autosomal data and y-dna haplogroups, with the finding of J2 in both CHG samples one can't but help correlate ENF's spread with haplogroup G2a, CHG with haplogroup J2, and EA with haplogroup E1b1. G2a is typically associated with the spread of agriculture, while J2 hunter-gatherers appear to have gradually adopted herding.

It's possible that J2-bearing pastoralist-raiders (descendants of CHG refugia) descended from their mountains after LGM and engaged in a protracted back-and-forth battle with farmers-- perhaps lasting for multiple millenia throughout the Neolithic-- before gradually at least partially converting to agriculture and integrating with the local ENF population. The bottom line is that we need many more genomes from the early, middle, late neolithic and bronze age Middle East to track the spread of ENF and CHG admixture.

Actually, only Kotias was J2a, Satsurblia was J1.

Megalophias
11-27-2015, 06:44 PM
It seems likely that J2-bearing pastoralist-raiders (descendants of the CHG refugia population) descended from their mountains throughout the Near East after the LGM and engaged in a protracted back-and-forth battle with farmers-- perhaps lasting for multiple millenia throughout the Neolithic-- before gradually at least partially converting to agriculture and integrating with the local ENF population.

Or that ENF and CHG are just two of many related Middle Eastern ancestral populations, but we lack the necessary reference samples to tell them all apart. Or that CHG-like people were among the first farmers in the eastern Fertile Crescent.

vettor
11-27-2015, 06:53 PM
Levantines (and Arabs in general) in this test score much higher for the CHG component than one would expect. The old ANE K7 calculator for example appeared to attribute >75% of Levantine autosomal DNA to the "Early Neolithic Farmer" component and only ~ 15% combined between WHG/ANE-- implying a virtually unbroken continuity between Fertile Crescent/Anatolian-type farmers and modern Levantines.

However in the new test the CHG component cannibalizes a considerable proportion of the ENF (even accounting for the apparent overlap between ENF and modern East African, which seems less pronounced in the ANE_K7 calc due to its more archaic EA sample). Obviously this effect was even more pronounced in Kurds, who had large portions of their ancestry scored as "Neolithic Farmer" in the old tests simple because CHG didn't "fit" the WHG or EHG models.

Re: Levantines, averaging the Syrian and Lebanese components from the spreadsheet (https://drive.google.com/file/d/0B-ObXiVfL-RzOEpDQ2U3MTJzNTA/view) produces an ENF/CHG/EastAfrican mixture like this: 24%/37%/18%. (Even Bedouins and Saudis surprisingly score ~35% CHG)

The combined EHG/WHG is still only about 10% for Levantines, so the CHG certainly wasn't contributed by Indo-Aryans like the Mittani etc (Mittani's likely closest approximation, the Sintashta sample, came up ~ 69% EHG/WHG & only 22% CHG), but rather directly from the CHG refugia descendants themselves.

While its not always wise making simple associations between autosomal data and y-dna haplogroups, with the finding of J2 in both CHG samples one can't but help correlate ENF's spread with haplogroup G2a, CHG with haplogroup J2, and EA with haplogroup E1b1. G2a is typically associated with the spread of agriculture, while J2 hunter-gatherers appear to have gradually adopted herding.

It seems likely that J2-bearing pastoralist-raiders (descendants of the CHG refugia population) descended from their mountains throughout the Near East after the LGM and engaged in a protracted back-and-forth battle with farmers-- perhaps lasting for multiple millenia throughout the Neolithic-- before gradually at least partially converting to agriculture and integrating with the local ENF population. This early, pre-Bronze Age dispersal would explain the ubiquitous ~35% CHG admixture observed throughout Mesopotamia, the Levant, Anatolia, and even the Arabian penninsula.

The bottom line is that we need many more genomes from the early, middle, late neolithic and bronze age Middle East to track the spread of ENF and CHG admixture.

Is the issue really that the ancient populace in the area are a completely different race and that the arabic race entered these areas after the demise of the western Roman empire . I cannot recall any mention of arabs by Roman historians in regards to north-africa or the levant ................maybe someone can link me more info

K33
11-27-2015, 07:06 PM
Or that CHG-like people were among the first farmers in the eastern Fertile Crescent.It's possible, although historically hunter-gatherers' spiritual horizon ranks raiding and herding as superior to farming, and even after contact with farming methods and peoples HG's and pastoralists (the Yamnaya, Mongols, Comanche, etc) remain reluctant to overahaul their value systems barring an admixture event. In the latter two of these examples, even the overwhelming force of modernity hasn't induced them to become agriculturalists or truly sedentary.

More fundamentally, from what I understand of the early Neolithic genomes found in Iraq, they bear very close resemblance to the Neolithic Anatolian farmer... on the other hand, even prior to the finding of CHG, the Caucasus were commonly posited as the origin of sheep and goat herding based on the region's topography and y-dna markers.

J Man
11-27-2015, 08:03 PM
It's possible, although historically hunter-gatherers' spiritual horizon ranks raiding and herding as superior to farming, and even after contact with farming methods and peoples HG's and pastoralists (the Yamnaya, Mongols, Comanche, etc) remain reluctant to overahaul their value systems barring an admixture event. In the latter two of these examples, even the overwhelming force of modernity hasn't induced them to become agriculturalists or truly sedentary.

More fundamentally, from what I understand of the early Neolithic genomes found in Iraq, they bear very close resemblance to the Neolithic Anatolian farmer... on the other hand, even prior to the finding of CHG, the Caucasus were commonly posited as the origin of sheep and goat herding based on the region's topography and y-dna markers.

Sheep and goat herding or any animal domestication did not originate in the Caucasus. Animal domestication originated in the Fertile Crescent maybe in the Zagros region more specifically.

Tomenable
11-27-2015, 08:19 PM
the Sintashta sample, came up ~ 69% EHG/WHG & only 22% CHG

That's actually pretty close to Khvalynsk samples.

ZephyrousMandaru
11-27-2015, 08:54 PM
Krefter, where are you getting your information? Except for those three populations mentioned previously, all middle easterners are much closer to Iranian than Loschbour is.

Chimp Iranian Loschbour Syrian -0.007 -1.908 503209
Chimp Iranian Loschbour Cypriot 0.0171 4.789 503209
Chimp Iranian Loschbour Saudi -0.0043 -1.161 503209
Chimp Iranian Loschbour Georgian 0.0214 6.05 503209
Chimp Iranian Loschbour BedouinB -0.0086 -2.381 503209
Chimp Iranian Loschbour Assyrian 0.0228 5.146 113202
Chimp Iranian Loschbour Turkish_Istanbul 0.0146 4.156 503209
Chimp Iranian Loschbour Lezgin 0.0194 5.48 503209

If two populations do not share drift, ie are 'split', then by definition there is no gene flow. Furthermore crown eurasian did not come into the middle east until afyer the kostenki-likes were replaced by loschbour and mal'ta-likes at the earliest.

Do you think that when we do discover an ancient genome from Southwest Asia, that this will decrease CHG and CHG-like ancestry in some Middle Easterners substantially considering that CHG still contains a good amount of Basal Eurasian?

K33
11-27-2015, 09:03 PM
Sheep and goat herding or any animal domestication did not originate in the Caucasus. Animal domestication originated in the Fertile Crescent maybe in the Zagros region more specifically.The important point is that sheep/goat herding didn't seem to originate in the lowland valleys, where agriculture took hold and where the ENFs seemed to dominate, but in the mountains. The northern entrance of the Zagros range would indeed seem an attractive migration route for mountainous CHG peoples into greater Mesopotamia, the Iranian plateau, and Balochistan.

Gravetto-Danubian
11-27-2015, 10:09 PM
The important point is that sheep/goat herding didn't seem to originate in the lowland valleys, where agriculture took hold and where the ENFs seemed to dominate, but in the mountains. The northern entrance of the Zagros range would indeed seem an attractive migration route for mountainous CHG peoples into greater Mesopotamia, the Iranian plateau, and Balochistan.

..... and north via the Caucasus to the steppe, taking "Teal" (? and M269) with them

ZephyrousMandaru
11-27-2015, 10:13 PM
The important point is that sheep/goat herding didn't seem to originate in the lowland valleys, where agriculture took hold and where the ENFs seemed to dominate, but in the mountains. The northern entrance of the Zagros range would indeed seem an attractive migration route for mountainous CHG peoples into greater Mesopotamia, the Iranian plateau, and Balochistan.

You do realize that CHG is a composite? I highly doubt the Basal Eurasian ancestry in CHG is native to the Caucasus or even that all of the CHG present in Non-Caucasians is reflective of in situ migrations, we have still yet to uncover ancient genomes from the lowland valleys. You can't draw conclusions about migrations from one or two genomes, we still need ancient genomes from South-Central Asia or elsewhere from the Middle East before we can deduce who migrated where. Formal stats demonstrated that Satsurblia and Kotias didn't contribute directly to South-Central Asians and if that's the case, then it's likely there was no direct gene flow to Middle Easterners either. Which implies that there may be an alternate source by which CHG in these populations was acquired and it very well could be a population in the lowlands.

Gravetto-Danubian
11-27-2015, 10:15 PM
Levantines (and Arabs in general) in this test score much higher for the CHG component than one would expect. The old ANE K7 calculator for example appeared to attribute >75% of Levantine autosomal DNA to the "Early Neolithic Farmer" component and only ~ 15% combined between WHG/ANE-- implying a virtually unbroken continuity between Fertile Crescent/Anatolian-type farmers and modern Levantines.

However in the new test the CHG component cannibalizes a considerable proportion of the ENF (even accounting for the apparent overlap between ENF and modern East African, which seems less pronounced in the ANE_K7 calc due to its more archaic EA sample). Obviously this effect was even more pronounced in Kurds, who had large portions of their ancestry scored as "Neolithic Farmer" in the old tests simple because CHG didn't "fit" the WHG or EHG models.

Re: Levantines, averaging the Syrian and Lebanese components from the spreadsheet (https://drive.google.com/file/d/0B-ObXiVfL-RzOEpDQ2U3MTJzNTA/view) produces an ENF/CHG/EastAfrican mixture like this: 24%/37%/18%. (Even Bedouins and Saudis surprisingly score ~35% CHG)

The combined EHG/WHG is still only about 10% for Levantines, so the CHG certainly wasn't contributed by Indo-Aryans like the Mittani etc (Mittani's likely closest approximation, the Sintashta sample, came up ~ 69% EHG/WHG & only 22% CHG), but rather directly from the CHG refugia descendants themselves.

While its not always wise making simple associations between autosomal data and y-dna haplogroups, with the finding of J2 in both CHG samples one can't but help correlate ENF's spread with haplogroup G2a, CHG with haplogroup J2, and EA with haplogroup E1b1. G2a is typically associated with the spread of agriculture, while J2 hunter-gatherers appear to have gradually adopted herding.

It seems likely that J2-bearing pastoralist-raiders (descendants of the CHG refugia population) descended from their mountains throughout the Near East after the LGM and engaged in a protracted back-and-forth battle with farmers-- perhaps lasting for multiple millenia throughout the Neolithic-- before gradually at least partially converting to agriculture and integrating with the local ENF population. This early, pre-Bronze Age dispersal would explain the ubiquitous ~35% CHG admixture observed throughout Mesopotamia, the Levant, Anatolia, and even the Arabian penninsula.

The bottom line is that we need many more genomes from the early, middle, late neolithic and bronze age Middle East to track the spread of ENF and CHG admixture.

Fair summary; but I dont think prehistoric reality was a Kung-Fu movie or a John Wayne western; with 'clashes of civilisations' and blood feuds till the death ;)

War was predicated on the basis of local conflicts over local key loco-regional strategic routes, lands/ pastures and other resources (odd vendetta and grievance excepted)

These Battles would have been conducted by men numbering in mere scores (~20-30 men; with usually low fatality rates). They engaged whoever required- whether a fellow forager, pastoralist or farmer; and irrespective of which haplogroup their cheek swab came back as

K33
11-28-2015, 12:34 AM
Fair summary; but I dont think prehistoric reality was a Kung-Fu movie or a John Wayne western; with 'clashes of civilisations' and blood feuds till the death ;)

War was predicated on the basis of local conflicts over local key loco-regional strategic routes, lands/ pastures and other resources (odd vendetta and grievance excepted)

These Battles would have been conducted by men numbering in mere scores (~20-30 men; with usually low fatality rates). They engaged whoever required- whether a fellow forager, pastoralist or farmer; and irrespective of which haplogroup their cheek swab came back asWell of course intra-tribal warfare occurred among all peoples including the CHG, but the mountainous terrain clearly shielded them from admixture. Otherwise, tribes like the Kurds and Balochis would quite literally no longer exist as distinct today.

K33
11-28-2015, 12:43 AM
You do realize that CHG is a composite? I'm not quite sure what you mean by this... the paper states CHG diverged from WHG ~ 45kya and from ENF ~25kya.


The distinctness of CHG can be clearly seen on a principal component analysis (PCA) plot11 loaded on contemporary Eurasian populations1, where they fall between modern Caucasian and South Central Asian populations in a region of the graph separated from both other hunter gatherer and EF samples (Fig. 1a). Clustering using ADMIXTURE software12 confirms this view, with CHG forming their own homogenous cluster (Fig. 1b). The close genetic proximity between Satsurblia and Kotias is also formally supported by D-statistics13, indicating the two CHG genomes form a clade to the exclusion of other pre-Bronze Age ancient genomes (Supplementary Table 2; Supplementary Note 3), suggesting continuity across the Late Upper Palaeolithic and Mesolithic periods.
http://www.nature.com/ncomms/2015/151116/ncomms9912/images/ncomms9912-f1.jpg

Generalissimo
11-28-2015, 12:55 AM
I'm not quite sure what you mean by this... the paper states CHG diverged from WHG ~ 45kya and from ENF ~25kya.

The paper's wrong. Lots of papers that get published are wrong. Nothing new in that.

Gravetto-Danubian
11-28-2015, 12:56 AM
Well of course intra-tribal warfare occurred among all peoples including the CHG, but the mountainous terrain clearly shielded them from admixture. Otherwise, tribes like the Kurds and Balochis would quite literally no longer exist as distinct today.

Not really. Even mountain people admix. The idea that people from mountainous regions are unadmixed - whilst partly true- mostly stems from Romantic stories rather than fact. Anyhow, I'm sure aDNA will prove me correct.In fact, we already know from the but handful of samples from the region.

And highland herders are inexricably linked to lowland agriculture. They were just certain sectors of a larger community.

Kurd
11-28-2015, 02:40 AM
The motivation for these stats was to check CHG's position vs. ENF's position with regards to ANE. If the assumption is that CHG's affinity to ANE is limited to any drift it shares with ANE, and there was no subsequent ANE input after CHG's split from Basal, then CHG & ENF should have approximately the same affinity to MA1. I say approximately because realistically CHG would appear a little more similar to MA1 than ENF, for the simple reason that the CHG genomes are older than the ENF genomes, and would thus be a little closer to the split of Basal from the rest of the non African groups, resulting in a little less non-shared drift with MA1, and thus should appear a little more similar than ENF to MA1,

I used 2 out groups to gauge absolute similarity to MA1, because for a Dstat of the form D(W, MA1, TARGET, OUT), W will affect the outcome if there is any similarity with the target, that is above and beyond any similarity the other targets share with W.

I first tried using Mbuti as outgroup W, but what I found that some targets had more affinity with Mbuti than others, not because there had been any genefow between Mbuti and the targets subsequent to OOA, but simply because some targets were older, and thus had less non-shred drift with Mbuti. So I opted for chimp gorilla outgroups.

Unfortunately my 2nd Karelian sample did not work out because it only had about 30K overlapping markers. I removed any samples with less than 130K marker overlap, because I found out they were at a disadvantage. Case at point Pashtun_Afghan had only 60K markers overlap, and because of this its affinity to MA1 was much less than Pathan. In fact its position was similar to Bedouin. So for an accurate comparison it is imperative that all samples have similar marker overlaps.

My conclusions based on this table are:

1- CHG shows more similarity to MA1 than ENF, and although it is likely that it is due to actual ANE input post split, there is a small possibility it may be due to CHG being older, and thus having less non-shared drift than ENF with regards to MA1.

2- S and SC Asians have recieved an ANE boost from the steppe post CHG, otherwise they would have appeared much farther from MA1 than CHG because they have much more non-shared drift with MA1 than CHG due to the 9000 or 10000 years that separate the two.

3- Kalash's position is interesting. It is closer to MA1 than other SC Asians, which begs the question why?

4- Somalis Ethiopian Jews and Yoruba were included for member Awale.

The table is sorted with the most similarity to MA1 on top.

Kurd
11-28-2015, 02:41 AM
I had to split the table in 2 because of post size requirements



OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
MA1
Samara_HG
Gorilla
-0.9653
-100
121501


Chimp
MA1
Karelia_HG
Gorilla
-0.9648
-100
199299


Chimp
MA1
Karitiana
Gorilla
-0.964
-100
205707


Chimp
MA1
Samara_Eneolithic
Gorilla
-0.9637
-100
147672


Chimp
MA1
Loschbour
Gorilla
-0.9634
-100
203847


Chimp
MA1
Scythian_IA
Gorilla
-0.9633
-100
190666


Chimp
MA1
Pima
Gorilla
-0.9632
-100
205707


Chimp
MA1
Yamnaya
Gorilla
-0.9631
-100
205050


Chimp
MA1
Srubnaya
Gorilla
-0.963
-100
203617


Chimp
MA1
Poltavka
Gorilla
-0.9629
-100
201554


Chimp
MA1
Lithuanian
Gorilla
-0.9627
-100
205707


Chimp
MA1
RISE_baAndrov
Gorilla
-0.9627
-100
205244


Chimp
MA1
Kalash
Gorilla
-0.9626
-100
205707


Chimp
MA1
Estonian
Gorilla
-0.9625
-100
205707


Chimp
MA1
SATSURBILA
Gorilla
-0.9623
-100
149251


Chimp
MA1
Lezgin
Gorilla
-0.9623
-100
205707


Chimp
MA1
Burusho
Gorilla
-0.9622
-100
205707


Chimp
MA1
KOTIAS
Gorilla
-0.9622
-100
181609


Chimp
MA1
Pathan
Gorilla
-0.9619
-100
205707


Chimp
MA1
Tajik_Pomiri
Gorilla
-0.9619
-100
205707


Chimp
MA1
Norwegian
Gorilla
-0.9619
-100
205707


Chimp
MA1
Punjabi
Gorilla
-0.9617
-100
205707

Kurd
11-28-2015, 02:43 AM
OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
MA1
Bulgarian
Gorilla
-0.9616
-100
205707


Chimp
MA1
Brahui
Gorilla
-0.9615
-100
205707


Chimp
MA1
Nganasan
Gorilla
-0.9615
-100
205707


Chimp
MA1
Balochi
Gorilla
-0.9614
-100
205707


Chimp
MA1
Kostenki14
Gorilla
-0.9613
-100
193820


Chimp
MA1
Georgian
Gorilla
-0.9612
-100
205707


Chimp
MA1
Armenian
Gorilla
-0.9612
-100
205707


Chimp
MA1
Potapovka
Gorilla
-0.9611
-100
105850


Chimp
MA1
Iceman
Gorilla
-0.9609
-100
204502


Chimp
MA1
Iranian
Gorilla
-0.9608
-100
205707


Chimp
MA1
Dai
Gorilla
-0.9608
-100
205707


Chimp
MA1
Sardinian
Gorilla
-0.9606
-100
205707


Chimp
MA1
LBK_EN
Gorilla
-0.9605
-100
205450


Chimp
MA1
Stuttgart
Gorilla
-0.9603
-100
202373


Chimp
MA1
Anatolia_Neolithic
Gorilla
-0.9599
-100
204242


Chimp
MA1
Syrian
Gorilla
-0.9598
-100
205707


Chimp
MA1
Saudi
Gorilla
-0.9596
-100
205707


Chimp
MA1
Ust_Ishim
Gorilla
-0.9593
-100
205307


Chimp
MA1
BedouinB
Gorilla
-0.959
-100
205707


Chimp
MA1
BedouinA
Gorilla
-0.9588
-100
205707


Chimp
MA1
Papuan
Gorilla
-0.9584
-100
205707


Chimp
MA1
Ethiopian_Jew
Gorilla
-0.9526
-100
205707


Chimp
MA1
Somali
Gorilla
-0.9504
-100
205707


Chimp
MA1
Yoruba
Gorilla
-0.9408
-100
205707

K33
11-28-2015, 03:46 AM
@Kurd, thanks for creating the calculator btw. Any estimate on the release of the updated one?

By the way, could you use the old Mota Cave sample (kit #M261275) for the East African component next time? I actually ran this ancient kit through the current K11 calculator, and this is how it scored:

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

In other words, the "East African" component in the K11 CHF calculator has so much Eurasian admixture that an actual pure East African sample matches SSA more closely... by using Mota next time instead, a lot of this East African in peoples' results should migrate to other more suitable reference groups like ENF, CHG, etc. It would also be more consistent from a timeframe perspective, since the test is purportedly meant to measure archaic components in particular...

Kurd
11-28-2015, 04:13 AM
Sorted with the most similarity to EHG on top. Transition SNPs excluded. Kotias is more accurate as Satsurbila is haploid and has less overlapping SNPs. Note position of Iceman and Kotias vs Neolithic farmers.




OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
Karelia_HG
Samara_HG
Gorilla
-0.9673
-100
165909


Chimp
Karelia_HG
Samara_Eneolithic
Gorilla
-0.9655
-100
201809


Chimp
Karelia_HG
Loschbour
Gorilla
-0.964
-100
275104


Chimp
Karelia_HG
Poltavka
Gorilla
-0.9638
-100
273358


Chimp
Karelia_HG
Yamnaya
Gorilla
-0.963
-100
277295


Chimp
Karelia_HG
Bichon
Gorilla
-0.9628
-100
186747


Chimp
Karelia_HG
Lithuanian
Gorilla
-0.9627
-100
277597


Chimp
Karelia_HG
Srubnaya
Gorilla
-0.9627
-100
275322


Chimp
Karelia_HG
Estonian
Gorilla
-0.9625
-100
277597


Chimp
Karelia_HG
Scythian_IA
Gorilla
-0.9625
-100
259708


Chimp
Karelia_HG
Norwegian
Gorilla
-0.9622
-100
277597


Chimp
Karelia_HG
Potapovka
Gorilla
-0.9622
-100
144458


Chimp
Karelia_HG
Iceman
Gorilla
-0.9621
-100
276050


Chimp
Karelia_HG
RISE_baAndrov
Gorilla
-0.962
-100
276964


Chimp
Karelia_HG
Karitiana
Gorilla
-0.9617
-100
277597


Chimp
Karelia_HG
Bulgarian
Gorilla
-0.9615
-100
277597


Chimp
Karelia_HG
Pima
Gorilla
-0.9615
-100
277597


Chimp
Karelia_HG
SATSURBILA
Gorilla
-0.9613
-100
201431


Chimp
Karelia_HG
Tajik_Pomiri
Gorilla
-0.9612
-100
277597


Chimp
Karelia_HG
Lezgin
Gorilla
-0.9611
-100
277597


Chimp
Karelia_HG
Kalash
Gorilla
-0.9609
-100
277597


Chimp
Karelia_HG
Pathan
Gorilla
-0.9606
-100
277597


Chimp
Karelia_HG
Sardinian
Gorilla
-0.9606
-100
277597


Chimp
Karelia_HG
KOTIAS
Gorilla
-0.9605
-100
244839


Chimp
Karelia_HG
Georgian
Gorilla
-0.9605
-100
277597


Chimp
Karelia_HG
LBK_EN
Gorilla
-0.9604
-100
277529


Chimp
Karelia_HG
Burusho
Gorilla
-0.9603
-100
277597


Chimp
Karelia_HG
Armenian
Gorilla
-0.9603
-100
277597


Chimp
Karelia_HG
Anatolia_Neolithic
Gorilla
-0.9601
-100
275648


Chimp
Karelia_HG
Nganasan
Gorilla
-0.9601
-100
277597


Chimp
Karelia_HG
Iranian
Gorilla
-0.96
-100
277597


Chimp
Karelia_HG
Brahui
Gorilla
-0.96
-100
277597


Chimp
Karelia_HG
Balochi
Gorilla
-0.9599
-100
277597


Chimp
Karelia_HG
Punjabi
Gorilla
-0.9599
-100
277597


Chimp
Karelia_HG
Stuttgart
Gorilla
-0.9595
-100
273101


Chimp
Karelia_HG
Syrian
Gorilla
-0.959
-100
277597


Chimp
Karelia_HG
Kostenki14
Gorilla
-0.9589
-100
261463


Chimp
Karelia_HG
Saudi
Gorilla
-0.9587
-100
277597


Chimp
Karelia_HG
BedouinB
Gorilla
-0.9585
-100
277597


Chimp
Karelia_HG
Dai
Gorilla
-0.9584
-100
277597


Chimp
Karelia_HG
BedouinA
Gorilla
-0.9582
-100
277597


Chimp
Karelia_HG
Ust_Ishim
Gorilla
-0.957
-100
277029


Chimp
Karelia_HG
Papuan
Gorilla
-0.9567
-100
277597


Chimp
Karelia_HG
Ethiopian_Jew
Gorilla
-0.951
-100
277597


Chimp
Karelia_HG
Somali
Gorilla
-0.9492
-100
277597


Chimp
Karelia_HG
Yoruba
Gorilla
-0.9382
-100
277597

Gravetto-Danubian
11-28-2015, 04:25 AM
Kurd

So you're again seeing something eastern in Oetzi

Kurd
11-28-2015, 04:37 AM
Kurd

So you're again seeing something eastern in Oetzi

He has always looked more similar than Stuttgart LBK and now Anatolian Farmers to WHG, EHG and now CHG., ever since I have focused on him in the last few months

Kurd
11-28-2015, 04:48 AM
Sorted with most similar to IA Scythians on top. Iranians and Pathans are both in the middle of the table. I took an interest in Scythians because of all the stories I have heard since childhood that Some Kurds and Iranians are Scythian derived



OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
Scythian_IA
Iceman
Gorilla
-0.963
-100
264265


Chimp
Scythian_IA
Bichon
Gorilla
-0.962
-100
178097


Chimp
Scythian_IA
Samara_HG
Gorilla
-0.9619
-100
160378


Chimp
Scythian_IA
Loschbour
Gorilla
-0.9618
-100
263486


Chimp
Scythian_IA
KOTIAS
Gorilla
-0.9616
-100
234522


Chimp
Scythian_IA
Yamnaya
Gorilla
-0.9616
-100
265449


Chimp
Scythian_IA
Srubnaya
Gorilla
-0.9616
-100
265442


Chimp
Scythian_IA
Lithuanian
Gorilla
-0.9615
-100
265791


Chimp
Scythian_IA
Potapovka
Gorilla
-0.9614
-100
140949


Chimp
Scythian_IA
Estonian
Gorilla
-0.9613
-100
265791


Chimp
Scythian_IA
Lezgin
Gorilla
-0.9613
-100
265791


Chimp
Scythian_IA
Norwegian
Gorilla
-0.9612
-100
265791


Chimp
Scythian_IA
Samara_Eneolithic
Gorilla
-0.9611
-100
196357


Chimp
Scythian_IA
Poltavka
Gorilla
-0.9611
-100
263901


Chimp
Scythian_IA
Bulgarian
Gorilla
-0.9611
-100
265791


Chimp
Scythian_IA
Tajik_Pomiri
Gorilla
-0.961
-100
265791


Chimp
Scythian_IA
Armenian
Gorilla
-0.9609
-100
265791


Chimp
Scythian_IA
Kalash
Gorilla
-0.9606
-100
265791


Chimp
Scythian_IA
Georgian
Gorilla
-0.9606
-100
265791


Chimp
Scythian_IA
LBK_EN
Gorilla
-0.9606
-100
265673


Chimp
Scythian_IA
Stuttgart
Gorilla
-0.9604
-100
261515


Chimp
Scythian_IA
Pathan
Gorilla
-0.9603
-100
265791


Chimp
Scythian_IA
Iranian
Gorilla
-0.9603
-100
265791


Chimp
Scythian_IA
Brahui
Gorilla
-0.9603
-100
265791


Chimp
Scythian_IA
Sardinian
Gorilla
-0.9603
-100
265791


Chimp
Scythian_IA
Karitiana
Gorilla
-0.9603
-100
265791


Chimp
Scythian_IA
Burusho
Gorilla
-0.9602
-100
265791


Chimp
Scythian_IA
Anatolia_Neolithic
Gorilla
-0.9601
-100
265792


Chimp
Scythian_IA
Balochi
Gorilla
-0.96
-100
265791


Chimp
Scythian_IA
RISE_baAndro
Gorilla
-0.96
-100
265156


Chimp
Scythian_IA
SATSURBILA
Gorilla
-0.9599
-100
193446


Chimp
Scythian_IA
Nganasan
Gorilla
-0.9598
-100
265791


Chimp
Scythian_IA
RISE_baArm
Gorilla
-0.9597
-100
177683


Chimp
Scythian_IA
Punjabi
Gorilla
-0.9596
-100
265791


Chimp
Scythian_IA
Syrian
Gorilla
-0.9594
-100
265791


Chimp
Scythian_IA
Pima
Gorilla
-0.9593
-100
265791


Chimp
Scythian_IA
Saudi
Gorilla
-0.9591
-100
265791


Chimp
Scythian_IA
BedouinB
Gorilla
-0.9588
-100
265791


Chimp
Scythian_IA
BedouinA
Gorilla
-0.9586
-100
265791


Chimp
Scythian_IA
Dai
Gorilla
-0.9586
-100
265791


Chimp
Scythian_IA
Kostenki14
Gorilla
-0.9579
-100
250149


Chimp
Scythian_IA
Papuan
Gorilla
-0.9568
-100
265791


Chimp
Scythian_IA
Ust_Ishim
Gorilla
-0.9563
-100
265256


Chimp
Scythian_IA
Ethiopian_Jew
Gorilla
-0.95
-100
265791


Chimp
Scythian_IA
Somali
Gorilla
-0.9498
-100
265791


Chimp
Scythian_IA
Yoruba
Gorilla
-0.9394
-100
265791

Kale
11-28-2015, 06:33 AM
[SIZE=3]I say approximately because realistically CHG would appear a little more similar to MA1 than ENF, for the simple reason that the CHG genomes are older than the ENF genomes, and would thus be a little closer to the split of Basal from the rest of the non African groups, resulting in a little less non-shared drift with MA1, and thus should appear a little more similar than ENF to MA1,

Age of the samples really doesn't play a major factor in d-stats. I'm sure if you had millions of years of drift it would, but not a few thousand. Mutation is just as likely to revert any derived SNP back to its ancestral state as it is to mutate an ancestral version to somebody else's derived version. If you don't believe me have fun chewing on these conundrums.

Chimp Ust-Ishim Kostenki14 Loschbour 0.0032 0.405 391567
30,000 year difference between K14 and Loschbour, must be some massive Usty<>WHG flow.
Chimp Ust_Ishim Samara_HG MA1 0.0041 0.411 6650 6705 148789
17,000 year difference, also some massive Usty<>EHG flow
Chimp Ust_Ishim Austroasiatic Kostenki14_UP 0.0032 0.423 122830
37,000 year difference...ginormous Usty<>ENA flow
Ju_Hoan_North Mota Ust-Ishim Han 0.0000 -0.010 13610 13611 297165
Usty also mixed it up with Mota.
MbutiPygmy Hadza Karitiana Ust'-Ishim -0.0032 -1.030
And of course the Hadza

Usty sure got around didn't he! lol

Kurd
11-28-2015, 02:53 PM
Age of the samples really doesn't play a major factor in d-stats. I'm sure if you had millions of years of drift it would, but not a few thousand. Mutation is just as likely to revert any derived SNP back to its ancestral state as it is to mutate an ancestral version to somebody else's derived version. If you don't believe me have fun chewing on these conundrums.

Chimp Ust-Ishim Kostenki14 Loschbour 0.0032 0.405 391567
30,000 year difference between K14 and Loschbour, must be some massive Usty<>WHG flow.
Chimp Ust_Ishim Samara_HG MA1 0.0041 0.411 6650 6705 148789
17,000 year difference, also some massive Usty<>EHG flow
Chimp Ust_Ishim Austroasiatic Kostenki14_UP 0.0032 0.423 122830
37,000 year difference...ginormous Usty<>ENA flow
Ju_Hoan_North Mota Ust-Ishim Han 0.0000 -0.010 13610 13611 297165
Usty also mixed it up with Mota.
MbutiPygmy Hadza Karitiana Ust'-Ishim -0.0032 -1.030
And of course the Hadza

Usty sure got around didn't he! lol


I do appreciate the feedback as it does stimulate thinking, but unfortunately there are so many flaws in your reasoning. Here are just a few:

First, with you have correctly pointed out that a mutation is likely to revert a derived SNP back to ancestral, but you have forgotten one very important thing, humans DNA is not comprised of a single SNP. Even with just a bi-allelic locus (for the sake of simplicity 0 or 1), and just a million loci, do you realize how many combinations you can have. 01011111011110111000000011111............ The probability of 2 genomes reverting back to being similar once a few of these sites start mutating, is practically 0, which is about the probability of us reverting back to ape (even in the absence of natural selection pressures).

Now let's look at the stats you presented:

1- Chimp Ust-Ishim Kostenki14 Loschbour 0.0032 0.405 391567

What immediately caught my eye was the number of SNPs. I suspected that transition SNPs had not been filtered out, otherwise the number would have been lower. So I ran my own Dstats to check. Transition SNPs need to be filtered out because they are susceptible to change during sequencing and amplification due to cytosine deamination (you probably also remember the cotamination issues raised by I believe Reich when Kotenski14 was sequenced). I wanted to gauge absolute affinity to Ust, so I used 2 outgroups, since I wanted include other samples in the comparison so that you can see the pattern that emerged



OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
MA1
Ust_Ishim
Gorilla
-0.9593
-100
205307


Chimp
LaBrana1
Ust_Ishim
Gorilla
-0.9593
-100
271054


Chimp
Han
Ust_Ishim
Gorilla
-0.9586
-100
286939


Chimp
Dai
Ust_Ishim
Gorilla
-0.9586
-100
286939


Chimp
Papuan
Ust_Ishim
Gorilla
-0.9584
-100
286939


Chimp
KOTIAS
Ust_Ishim
Gorilla
-0.9582
-100
252477


Chimp
Kostenki14
Ust_Ishim
Gorilla
-0.9579
-100
269687


Chimp
Bichon
Ust_Ishim
Gorilla
-0.9578
-100
191261


Chimp
Yamnaya
Ust_Ishim
Gorilla
-0.9575
-100
285906


Chimp
Karelia_HG
Ust_Ishim
Gorilla
-0.957
-100
277029


Chimp
Stuttgart
Ust_Ishim
Gorilla
-0.957
-100
282363


Chimp
Loschbour
Ust_Ishim
Gorilla
-0.9566
-100
284422


Chimp
LBK_EN
Ust_Ishim
Gorilla
-0.9559
-100
286504


Chimp
BedouinB
Ust_Ishim
Gorilla
-0.9555
-100
286939


Chimp
Anatolia_Neolithic
Ust_Ishim
Gorilla
-0.955
-100
284891


Chimp
Yoruba
Ust_Ishim
Gorilla
-0.94
-100
286939



The observable patterns are:

1- Any genomes with either substantial African or Basal Eurasian are furthest from Ust, with the more modern ones further down the list
2- Ust prefers E Eurasian shifted genomes because of slightly more shared drift, and older genomes because of less non-shared drift (La Brana). K14 would have been even higher up the list if it were not for some of the Basal Eurasian input he had. Loschbour is further down the list from LaBrana because he has more non-shared drift (younger). He is down the list from K14 because apparently his lack of Basal input, was not enough to counteract his younger age relative to K14.
3- Kotias is further up the list from LBK, Stuttgart, and Bedouin, because he probably has some ANE input.

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

Kurd
11-28-2015, 03:16 PM
Age of the samples really doesn't play a major factor in d-stats. I'm sure if you had millions of years of drift it would, but not a few thousand. Mutation is just as likely to revert any derived SNP back to its ancestral state as it is to mutate an ancestral version to somebody else's derived version. If you don't believe me have fun chewing on these conundrums.

Chimp Ust-Ishim Kostenki14 Loschbour 0.0032 0.405 391567
30,000 year difference between K14 and Loschbour, must be some massive Usty<>WHG flow.
Chimp Ust_Ishim Samara_HG MA1 0.0041 0.411 6650 6705 148789
17,000 year difference, also some massive Usty<>EHG flow
Chimp Ust_Ishim Austroasiatic Kostenki14_UP 0.0032 0.423 122830
37,000 year difference...ginormous Usty<>ENA flow
Ju_Hoan_North Mota Ust-Ishim Han 0.0000 -0.010 13610 13611 297165
Usty also mixed it up with Mota.
MbutiPygmy Hadza Karitiana Ust'-Ishim -0.0032 -1.030
And of course the Hadza

Usty sure got around didn't he! lol

Continuing with stats you presented:

Chimp Ust_Ishim Samara_HG MA1 0.0041 0.411 6650 6705 148789
17,000 year difference, also some massive Usty<>EHG flow

Unfortunately you got this totally wrong. This one actually shows MA1 more similar than Samara, which is corroborated by the table I posted above, MA1 tops the list, again because he is older than Samara, which goes to prove my point.

Also, FYI, based on all the experience I have running Dstats, it is a very bad idea to mix samples with 148,789 SNPs with samples of 300,000 SNPs, as those are at a disadvantage. Case in point, I did some Dstats yesterday with various SC Asian and W Asian groups vs EHG. My Pashtun and Afghan Tajik samples only had 60K overlapping SNPs, whereas the Bedouin Pathan and other samples had around 200K overlapping SNPs with EHG. Pathan was towards the top of a 50 sample list with regards to similarity to EHG, whereas Tajik and Pashtun where at the bottom with Bedouin, because of their low SNPs.


Chimp Ust_Ishim Austroasiatic Kostenki14_UP 0.0032 0.423 122830
37,000 year difference...ginormous Usty<>ENA flow

You got this wrong too. It actually shows K14 more similar than Austroasiatic. Again the stat would have been even more in favor of K14, had it not been for K14's partial Basal Eurasian makeup.


Ju_Hoan_North Mota Ust-Ishim Han 0.0000 -0.010 13610 13611 297165
Usty also mixed it up with Mota.

This one is totally insignificant. Check the D and Z score, but if anything it goes to prove my point, even though as we all know modern E Asians are relatively close to Ust (check table above), this closeness was negated by Mota's age.



MbutiPygmy Hadza Karitiana Ust'-Ishim -0.0032 -1.030
And of course the Hadza

You got this wrong too. Actually it shows the opposite of what you are implying, it shows that Karitiana are more similar to Hadza, than Ust is to Hadza.


You may want to brush up on how to read Dstats and the significance of the Z score, as some of the stats you posted are insignificant (Z<3)

Kurd
11-28-2015, 04:11 PM
Posted at the request of some members who want to compare to Scythian IA. I had to break it up into 3 different groups as for example, some members are V3 genotyped and thus have more overlap with Scythian, than those who are V4. Combining the 3 would not be an accurate comparison. Sorted by most similar on top. Those with the most SNP overlap receive the highest advantage.



OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
Scythian_IA
.Pak_Gujjar
Gorilla
-0.9513
-100
109499


Chimp
Scythian_IA
.Hanna
Gorilla
-0.9509
-100
110765


Chimp
Scythian_IA
.Varun
Gorilla
-0.9506
-100
110665


Chimp
Scythian_IA
.Sein
Gorilla
-0.9504
-100
109553


Chimp
Scythian_IA
.Bol_Nat
Gorilla
-0.9498
-100
108661


Chimp
Scythian_IA
.Zephyrous
Gorilla
-0.9493
-100
110823



There is a 3-way tie here:



OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
Scythian_IA
.Kandhari
Gorilla
-0.9477
-100
79853


Chimp
Scythian_IA
.Bored
Gorilla
-0.9476
-100
79938


Chimp
Scythian_IA
.Rukha
Gorilla
-0.9468
-100
79889


Chimp
Scythian_IA
.Kurd_C2
Gorilla
-0.9468
-100
79934


Chimp
Scythian_IA
.Kurd_SE
Gorilla
-0.9468
-100
79911


Chimp
Scythian_IA
.Kenji
Gorilla
-0.9467
-100
79786


Chimp
Scythian_IA
.Zara
Gorilla
-0.9464
-100
79957


Chimp
Scythian_IA
.Dluffy
Gorilla
-0.9463
-100
79804


Chimp
Scythian_IA
.Kurd_C1
Gorilla
-0.946
-100
79868


Chimp
Scythian_IA
.Khana
Gorilla
-0.9447
-100
79832


Chimp
Scythian_IA
.Jesus
Gorilla
-0.9444
-100
79756


Chimp
Scythian_IA
.Passa
Gorilla
-0.9442
-100
79661


Chimp
Scythian_IA
.Farid
Gorilla
-0.9435
-100
79793


Chimp
Scythian_IA
.Awale
Gorilla
-0.935
-100
79879




OUTGROUP
POP1
TARGET
OUTGROUP
D
Z
SNP


Chimp
Scythian_IA
.Arbogan
Gorilla
-0.9474
-100
93461


Chimp
Scythian_IA
.DMXX
Gorilla
-0.9454
-100
93528

Kurd
11-28-2015, 05:02 PM
I realize that many don't know how to properly read Dstats, so I am posting some instructions straight from the Dstat Readme file in ADMIXTOOLS for reference::

DOCUMENTATION OF D-statistics (qpDstat):

The 4-population test, implemented here as D-statistics, is also a formal test for admixture based on a four taxon 4 statistic, which can provide some information about the direction of gene flow.
For any 4 populations (W, X, Y, Z), qpDstat computes the D-statistics as -
num = (w − x)(y − z )
den = (w + x − 2wx)(y + z − 2yz )

D = num/ den

The output of qpDstat is informative about the direction of gene flow. So for 4 populations (W, X, Y, Z) as follows -
If the Z-score is +ve, then the gene flow occured either between W and Y or X and Z
If the Z-score is -ve, then the gene flow occured either between W and Z or X and Y.

So we make Z or sometimes W and Z outgroups to rule out geneflow W and Y or W and Z etc