PDA

View Full Version : Eurogenes _ ANE K7



Pages : [1] 2 3

ZephyrousMandaru
09-02-2014, 05:22 PM
There's a new test that will available soon from the Eurogenes Genetic Ancestry Project called K=6. The following labels will be applied to the components.

Ancient North Eurasian (ANE)
Ancestral South Indian (ASI)
Western Hunter-Gatherer (WHG)
Mediterranean (similar to EEF)
Middle Eastern
East Eurasian
Sub-Saharan African

Here's a preview PCA of what's to come.

https://docs.google.com/file/d/0B9o3EYTdM8lQT2VkZVFLSVY1ejg/edit

I created this thread in advance to post our results in.

Dr_McNinja
09-04-2014, 02:53 AM
Is it going to be on GEDmatch?

ZephyrousMandaru
09-04-2014, 03:39 AM
Is it going to be on GEDmatch?

I think so.

Sein
09-04-2014, 03:39 PM
David is still working on this test, but he has posted ANE levels for most Eurasian populations in his data-set. I must say, the results are the most accurate I've seen yet. I really think he has latched onto solid ANE estimates.

http://polishgenes.blogspot.com/2014/09/ancient-north-eurasian-ane-admixture.html

Edit: On average, it seems that the Chechens are around 24%-25% ANE-admixed, the Lezgins are around 26%-27% ANE-admixed, the Di Cristofaro Pashtuns (from Afghanistan) are around 32% ANE-admixed (the ones that don't show evidence of non-Pashtun ancestry), the HGDP Pashtuns (either from eastern Afghanistan or FATA, but probably FATA) are around 33%-34% ANE-admixed, and the Burusho are around 34%-35% ANE-admixed. Quite solid.

Dr_McNinja
09-04-2014, 04:31 PM
I wonder how Indian results will look and if ASI will screw with the results

DMXX
09-04-2014, 04:32 PM
ANE population averages from that spreadsheet for West Asian groups in order of decreasing frequency, plus two select additions (Turkmen, Uyghur).

The Turkmen having a greater score than Uyghurs, a population who are situated closer to Siberia and have a larger ancestral derivation from the steppelands, reads as a pretty good indication that some sort of genetic substratum in South-Central Asia is boosting the ANE scores beyond what one would normally expect. This could be something which resembles it (i.e. a later prehistoric Eurasian hunter-gatherer which became more specific to this area), or some other ancestral component modal elsewhere that's interfering (ASI?). The only alternative would have us reconsider our preconceived notions of what the involved groups are, genetically, but I find this a difficult prospect to accept, given the likes of Turkmen or Uzbek do technically fall on one genetic cline beginning in West Asia.

I have a suspicion this may also extend (albeit possibly to a lesser extent) to Iranians and Kurds. I'm having a hard time accepting they have comparable ANE as the Uyghur, but this may again have something to do with a preconceived notion on my part.

Also quite surprising to see the Lebanese Muslim samples are ~3% more ANE than their Christian counterparts, given European admixture via the Crusades has been frequently implicated in both discussion boards and select studies.

We ultimately need more data from various parts of Eurasia to make the ANE signal more distinct and settle some of the conflicting, yet reasonable, suppositions many of us have of the populations involved.



Turkmen 20.8
Iranian 18.9
Uyghur 18.7
Kurdish 18.5
Azeri 17.9
Turkish 16.1
Armenian 14.5
Assyrian 12.9
Leb_Muslim 9.3
Syrian 8.8
Druze 7.9
Leb_Christian 6.5
Palestinian 5
Bedouin negligible
Saudi negligible

bored
09-04-2014, 04:42 PM
I wonder how Indian results will look and if ASI will screw with the results

Is this because ASI is quite hard to separate out from ANE? I'm also very curious to see South Asian ANE results.

parasar
09-04-2014, 05:51 PM
... I'm also very curious to see South Asian ANE results.
The Piramalai Kallar (TN) have ~30% ANE.


Piramalai PK6415 0.310205
Piramalai PK6422 0.301029
Piramalai PK6442 0.301389
Piramalai PK6458 0.293844
Piramalai PK6459 0.301784

Dr_McNinja
09-04-2014, 06:24 PM
The Piramalai Kallar (TN) have ~30% ANE.Do you think that's realistic? Would it indicate ANE is just very old and/or it has some kind of relationship with ASI?

bored
09-04-2014, 06:24 PM
The Piramalai Kallar (TN) have ~30% ANE.

Very interesting! It seems that ANE is elevated all across South Asia, even deep into South India, which is a bit surprising considering the region's geographical distance from South Central Asia (where ANE peaks in the region). I suppose it is also possible that some of the ASI is leaking into ANE?

evon
09-04-2014, 06:38 PM
David, where are those Norwegian samples from?

Sein
09-05-2014, 02:39 AM
ANE population averages from that spreadsheet for West Asian groups in order of decreasing frequency, plus two select additions (Turkmen, Uyghur).

The Turkmen having a greater score than Uyghurs, a population who are situated closer to Siberia and have a larger ancestral derivation from the steppelands, reads as a pretty good indication that some sort of genetic substratum in South-Central Asia is boosting the ANE scores beyond what one would normally expect. This could be something which resembles it (i.e. a later prehistoric Eurasian hunter-gatherer which became more specific to this area), or some other ancestral component modal elsewhere that's interfering (ASI?). The only alternative would have us reconsider our preconceived notions of what the involved groups are, genetically, but I find this a difficult prospect to accept, given the likes of Turkmen or Uzbek do technically fall on one genetic cline beginning in West Asia.

I have a suspicion this may also extend (albeit possibly to a lesser extent) to Iranians and Kurds. I'm having a hard time accepting they have comparable ANE as the Uyghur, but this may again have something to do with a preconceived notion on my part.

Also quite surprising to see the Lebanese Muslim samples are ~3% more ANE than their Christian counterparts, given European admixture via the Crusades has been frequently implicated in both discussion boards and select studies.

We ultimately need more data from various parts of Eurasia to make the ANE signal more distinct and settle some of the conflicting, yet reasonable, suppositions many of us have of the populations involved.



Turkmen 20.8
Iranian 18.9
Uyghur 18.7
Kurdish 18.5
Azeri 17.9
Turkish 16.1
Armenian 14.5
Assyrian 12.9
Leb_Muslim 9.3
Syrian 8.8
Druze 7.9
Leb_Christian 6.5
Palestinian 5
Bedouin negligible
Saudi negligible


But the Uyghur are around 50% ENA, and their non-ENA admixed ancestors were unlikely to be predominantly ANE, so their results do make sense.

We have to maintain a holistic picture. The formal tests do demonstrate that ANE admixture is highest in South Asia (f3 statistics, shared drift), as do methods used in the literature (TreeMix). Not to mention other forms of analysis (PCA, MA1's ADMIXTURE results, IBS). So the ADMIXTURE output isn't unique/alone. When all methods point in a single direction, consistently, we need to reexamine our priors.

ASI was once a good candidate, but it seems ASI admixture decreases ANE affinity. If South Asians didn't have ASI, they would be much more ANE-shifted. Local West Eurasian-hunter ancestry was also a good possibility, but the poor WHG-affinity in South Asia (and eastern West Asia, as well as Central Asia) makes this exceedingly unlikely.

Regardless, the importance of Y-DNA haplogroup R in South Asia, as well as South Asian-specific mtDNA U lineages, is excellent corroboration from the uniparental data, as far as a peak in ANE ancestry is concerned.

The big question right now concerns what set of pre-historical processes can account for the high ANE-admixture in South Asia. Reich et al. want to pin this on the Indo-European dispersal into South Asia (they mention heavy ANE-admixture in South Asia, in one of their comments concerning the Kalash-centered component). Personally, I find this unlikely. I think we will find that the Burushaski-Yeniseian connection is of great relevance. At the end of the day, it could involve both events dating to the Mesolithic (who knows, maybe even the Upper Paleolithic), and events dating to the Bronze Age. Also, South Asia has been at the receiving end for countless southwestern leaning movements from the steppe, which must have had a cumulative genetic effect (across South Central Asia and the Sub-Continent), if we ignore any punctuated events.

Edit: When it comes to Lebanese Muslims in comparison to Christians, I would note that Europeans aren't really rich in ANE-admixture, especially in the regions from which the supposed Crusader ancestors of Levantine Christians originated. And that's even if we concede European admixture for Levantine Christians. Going by what I know concerning Near Eastern genetics, I would assume Levantine Christians are less genetically cosmopolitan in comparison to Muslims.

Generalissimo
09-05-2014, 03:21 AM
Lebanese Christians don't really have that much European admixture. The reason they're relatively distinct from Lebanese Muslims is because they have remained more isolated during historic times, much like the Druze and various Jewish groups. As a result these groups now have less ANE than the Muslims, and this is why they're actually more similar to Mediterranean Europeans.

It's basically the West Asian version of the Sardinian effect, which results in a more western shift due to less ANE, and thus a seemingly more European outcome.

DMXX
09-05-2014, 05:44 AM
But the Uyghur are around 50% ENA, and their non-ENA admixed ancestors were unlikely to be predominantly ANE, so their results do make sense.


That is a good point, you're right. I suppose if we remove most of the ENA and "double up" the remaining components, the ANE would be in the 35-40% range, which would make sense given the history of the region.

This is precisely what I referred to regarding the "preconceived notions" we have regarding certain groups in the region. If we go by established history, the Uyghurs should contain a substantial amount of ANE. A quick whistle-stop tour of Xinjiang's history to provide us a narrative of how the Uyghurs came into being:

The earliest evidence of humans in Xinjiang was by (paleolithic from memory) mobile Eurasian hunter-gatherers whose flint tools bore a resemblance to those found in places such as Afghanistan (this may be one of our pre-IE ANE clues). Then we have intrusions from the north via Afanasievo, which was a hybrid culture of Yamnaya and local hunter-gatherer practices and the likely speakers of Tocharian languages. Then, we observe a flow of interactions originating from western populations (Andronovo, later Pamiri populations), which continues up until the Medieval era via various speakers of Iranic, providing us with two more links between these parts of Central Asia. Then we find the first interactions with East Asia via the Han Chinese, who initiated the Silk Road's founding by sending a trade convoy towards Parthia. After many decades of cosmopolitan life and a significant population density was reached across Central Asia, the speakers of Uyghur Turkish arrived around the 9th century and this became the new identity of the local population.

With the above in mind, most who follow the historical narrative would suspect the Uyghurs and other groups in the region with a large background of steppe ancestry to have numbers greater than those observed in places such as Iran or Turkmenistan. I do suspect the early Turks had some ANE themselves, but any estimates would be entirely speculative at best.

I suppose the overarching lesson here is to keep in mind the additional components, such as ENA in this instance.

evon
09-05-2014, 08:37 AM
It is worth noting that the Tarim Basin (Xinjiang province) where Uighur's mostly live, have had a a long and stable influx of South Asian peoples, mainly from North India and North Pakistan, especially during the 1-200CE period. So that could explain their elevated ANE if ANE is elevated in North India ect..

MasterRoshi
09-06-2014, 02:13 AM
I wonder how Indian results will look and if ASI will screw with the results


You have to know, ANE played a role in forming ASI itself. What means some genes in ASI are ANE related. In South Asia some of the ASI might "screw" with the results but I doubt that is the case in North and West Asia. I strongly assume that the few percent ASI in Iranians(4%) and Kurds(1%) is actually related to ANE and came together with the entire ANE package.

Alanson
09-06-2014, 03:36 AM
Why does ANI peaks in the Karitiana people? They are a Native American tribe?

parasar
09-06-2014, 07:25 PM
Why does ANI peaks in the Karitiana people? They are a Native American tribe?

I think you mean ANE.
MA1/ANE and the Karitiana share a common Siberian ancestry.

parasar
09-06-2014, 07:33 PM
You have to know, ANE played a role in forming ASI itself. What means some genes in ASI are ANE related. In South Asia some of the ASI might "screw" with the results but I doubt that is the case in North and West Asia. I strongly assume that the few percent ASI in Iranians(4%) and Kurds(1%) is actually related to ANE and came together with the entire ANE package.

That is very doubtful.
ASI and ANE may form a clade going back 40000 years but I doubt that there is any ANE in populations such the Ongee.

MasterRoshi
09-06-2014, 09:43 PM
That is very doubtful.
ASI and ANE may form a clade going back 40000 years but I doubt that there is any ANE in populations such the Ongee.


Davidski made a break up of ANE using calculator components. And ANE turned out 27% Caucasus_Gedrosia like, 33% North European like, 26% Amerindian like and 13% ASI like.

ASI is a relatively modern component. And ANE played a role in forming this component. Just like ANE played a role in forming the Amerindian, Caucasus-Gedrosia and North European component.

ASI is like Southeast Asian/Austranesian with ANE admixture. And I assume most of this genes showing up in Kurds and Iranians as "ASI" is actually ANE related part.

http://www.eupedia.com/forum/attachment.php?attachmentid=6152&d=1387915415

Sein
09-07-2014, 03:22 AM
Davidski made a break up of ANE using calculator components. And ANE turned out 27% Caucasus_Gedrosia like, 33% North European like, 26% Amerindian like and 13% ASI like.

ASI is a relatively modern component. And ANE played a role in forming this component. Just like ANE played a role in forming the Amerindian, Caucasus-Gedrosia and North European component.

ASI is like Southeast Asian/Austranesian with ANE admixture. And I assume most of this genes showing up in Kurds and Iranians as "ASI" is actually ANE related part.

http://www.eupedia.com/forum/attachment.php?attachmentid=6152&d=1387915415

So you aren't really referring to ASI (which is a theoretical construct used to describe the ENA ancestry unique to South Asia and it's environs). Rather, you are referring to the ADMIXTURE cluster modal in South India. That cluster is a mix of ANE+ASI+Near Eastern. Calling it ASI leads to confusion.

jesus
09-07-2014, 03:30 AM
When is it going to be available ?

ZephyrousMandaru
09-07-2014, 05:14 AM
When is it going to be available ?

Hopefully sometime early this week.

MasterRoshi
09-07-2014, 11:36 AM
So you aren't really referring to ASI (which is a theoretical construct used to describe the ENA ancestry unique to South Asia and it's environs). Rather, you are referring to the ADMIXTURE cluster modal in South India. That cluster is a mix of ANE+ASI+Near Eastern. Calling it ASI leads to confusion.


C6 with 11,03% is labeled as Ancestral South Indian. So I am quite sure this is at least by large majority ASI. And wasn't Onge used as refference population for ASI by Reich too? I am not sure thats why asking.

parasar
09-07-2014, 02:07 PM
Davidski made a break up of ANE using calculator components. And ANE turned out 27% Caucasus_Gedrosia like, 33% North European like, 26% Amerindian like and 13% ASI like.

ASI is a relatively modern component. And ANE played a role in forming this component. Just like ANE played a role in forming the Amerindian, Caucasus-Gedrosia and North European component.

ASI is like Southeast Asian/Austranesian with ANE admixture. And I assume most of this genes showing up in Kurds and Iranians as "ASI" is actually ANE related part.

How much ANE is there in populations such as the Ongee?
Ancestral ASI is OoF - Basal West Eurasian, which split happened ~50000 ybp. From the Ancestral ASI side populations such as ASI, WHG, ANE, East Asian emanate. But even this split happened ~40000 ybp. So the Ongee for example would have barely higher affinity to more geographically distant WHG than to EEF.

parasar
09-07-2014, 03:17 PM
I think you mean ANE.
MA1/ANE and the Karitiana share a common Siberian ancestry.
http://www.sciencemag.org/content/345/6201/1106.full

...these Eurasians lived in Siberia 24,000 years ago. Some of their descendants then crossed Beringia to the New World more than 14,000 years ago. Others reached Scandinavia—but not central or Western Europe—by 8000 years ago, explaining the presence of their DNA in Scandinavian hunter-gatherers. Then, sometime after 7000 years ago, people carrying this DNA entered central and Western Europe.

Hok
09-08-2014, 08:14 AM
Let's go!

http://bga101.blogspot.nl/2014/09/eurogenes-ane-k7.html

evon
09-08-2014, 08:36 AM
http://bga101.blogspot.nl/2014/09/eurogenes-ane-k7.html


Ancestral South Eurasian (ASE): this is a really basal cluster that peaks in tribal groups of Southeast Asia. It's probably very similar in some ways to the Ancestral South Indian (ASI) component described by Reich et al. a few years ago.

Western European/Unknown Hunter-Gatherer (WHG-UHG): this essentially looks like a West Eurasian forager component, and includes the forager-like stuff carried by Neolithic farmers (Oetzi the Iceman has 40% of it).

Early Neolithic Farmer (ENF): I'd say that this is the component of the earliest Neolithic farmers from the Fertile Crescent.

Dont have time to do all of my relatives just now, but here is mine so far...

Mine:
15.64% ANE
2.72% ASE
68.31% WHG-UHG
0.44% East_Eurasian
0.29% West_African
0.64% East_African
11.95% ENF

StarDS9
09-08-2014, 09:01 AM
My results

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

19.35% ANE
4.59% ASE
10.98% WHG-UHG
0.71% East_Eurasian
1.05% West_African :confused:
0.40% East_African
62.91% ENF

MfA
09-08-2014, 09:17 AM
My results

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

19.35% ANE
4.59% ASE
10.98% WHG-UHG
0.71% East_Eurasian
1.05% West_African :confused:
0.40% East_African
62.91% ENF

.....

Stephen1986
09-08-2014, 09:19 AM
I'll have to wait for this to be available on Gedmatch as I can't download R as I now use a Chromebook rather than a laptop.

DMXX
09-08-2014, 09:22 AM
My results. A bit confused by the African as well.

18.17% ANE
5.66% ASE
10.58% WHG-UHG
1.96% East_Eurasian
0.12% West_African
1.63% East_African
61.89% ENF

MfA
09-08-2014, 09:49 AM
This is my fellow Ezidi Kurdish friend's results, just like mine


----------------------------
FINAL ADMIXTURE PROPORTIONS:
----------------------------
17.95% ANE
4.54% ASE
8.00% WHG-UHG
0.00% East_Eurasian
0.52% West_African
0.30% East_African
68.69% ENF

Dr_McNinja
09-08-2014, 10:07 AM
There's a 95% genotyping rate with 23andMe V3 and FTDNA FF files. Only 86% with 23andMe V4.

DMXX
09-08-2014, 10:22 AM
What was the rationale behind combining WHG and UHG ancestry into a single component? Was there some sort of technical difficulty preventing a distinction?

Also, is it possible some actual ANE has ended up in this WHG-UHG component?

Generalissimo
09-08-2014, 10:28 AM
What was the rationale behind combining WHG and UHG ancestry into a single component? Was there some sort of technical difficulty preventing a distinction?

It's the same component.

Minor African signals can be put down to Basal Eurasian ancestry.


Also, is it possible some actual ANE has ended up in this WHG-UHG component?

Not really.

Dr_McNinja
09-08-2014, 10:36 AM
I'm about to post several South Asian results, we're all averaging several % of WHG-UHG (whopping 13% in Haryana Jatts). The amount of WHG-UHG in this calculator directly corresponds to varying rates of "NE-Euro" component in other calculators.

What do you suppose WHG-UHG stands for in South Asians who should be 0% WHG from what I've heard? It looks here like it's complementary to ENF (high WHG-UHG correlates with lower ENF among South Asians or Indians/Pakistanis east of Punjab).

Dr_McNinja
09-08-2014, 10:52 AM
~95-96% genotyping rate for both 23andMe V3 and FTDNA FF... Unless otherwise mentioned, everyone I'm testing is getting 95-96% genotyping rate too... 23andMe V4 is around 86%

Me and my family:


(my family are all Punjabi Jatt Muslim, but from Central/Eastern Punjab as much as I know)

HRP0349, Me (23andMe V3):

29.23% ANE
16.80% ASE
5.30% WHG-UHG
5.66% East_Eurasian
0.41% West_African
3.15% East_African
39.44% ENF

HRP0349, Me (FTDNA):

29.25% ANE
16.84% ASE
5.36% WHG-UHG
5.64% East_Eurasian
0.42% West_African
3.15% East_African
39.34% ENF

HRP0374 (My mother)

29.89% ANE
16.75% ASE
5.44% WHG-UHG
4.02% East_Eurasian
0.22% West_African
2.94% East_African
40.74% ENF

My father (FTDNA, 97.8% genotyping rate)

29.02% ANE
16.46% ASE
5.76% WHG-UHG
5.48% East_Eurasian
0.87% West_African
2.80% East_African
39.61% ENF

Grandmother (mother's mother) (FTDNA, genotyping rate 97.5%) (Might be half or 1/4 Western Punjab)

30.09% ANE
17.00% ASE
4.83% WHG-UHG
3.82% East_Eurasian
0.51% West_African
2.92% East_African
40.82% ENF

Other South Asian members:


HRP0393 (Haryana Jatt) (23andMe V4)

30.16% ANE
15.03% ASE
13.37% WHG-UHG
3.00% East_Eurasian
0.04% West_African
3.06% East_African
35.33% ENF

---

HRP0341 (Sapporo) (Punjabi Jatt Sikh)

30.70% ANE
14.79% ASE
6.24% WHG-UHG
3.08% East_Eurasian
0.01% West_African
2.84% East_African
42.33% ENF

---

HRP0402 (Punjabi Jatt Muslim, Gondal) (Western Punjab) (23andMe V4)

30.08% ANE
16.10% ASE
4.94% WHG-UHG
1.90% East_Eurasian
0.05% West_African
3.72% East_African
43.22% ENF

---

soulblighter (99.4% genotyping rate) (South Indian)

28.36% ANE
20.79% ASE
0.30% WHG-UHG
6.69% East_Eurasian
0.05% West_African
4.53% East_African
39.28% ENF

---

HRP0315 (Everest) (Nepal Brahmin)

28.59% ANE
17.75% ASE
7.83% WHG-UHG
7.95% East_Eurasian
0.23% West_African
2.56% East_African
35.09% ENF

---

HRP0353 (Punjabi Gujjar Muslim) (Western Punjab)

30.94% ANE
17.20% ASE
2.31% WHG-UHG
1.57% East_Eurasian
0.50% West_African
2.52% East_African
44.95% ENF

---

HRP0350 (Punjabi Jatt Muslim) (Western Punjab)

29.10% ANE
16.11% ASE
4.11% WHG-UHG
4.25% East_Eurasian
0.37% West_African
1.97% East_African
44.10% ENF

---

HRP0282 (Sein) (Pashtun Pakistan)

29.26% ANE
13.41% ASE
5.02% WHG-UHG
4.35% East_Eurasian
0.46% West_African
1.71% East_African
45.79% ENF

---

Half Afghan Panjsheri Tajik, Half Afghan Pashtun (23andMe V4)

29.38% ANE
10.28% ASE
9.35% WHG-UHG
3.25% East_Eurasian
0.17% West_African
1.56% East_African
46.01% ENF

---

HRP0370 (Afghan Pashtun Popalzai from Kandahar)

26.42% ANE
9.32% ASE
11.34% WHG-UHG
3.07% East_Eurasian
0.15% West_African
2.11% East_African
47.59% ENF

Generalissimo
09-08-2014, 10:53 AM
I'm about to post several South Asian results, we're all averaging several % of WHG-UHG (whopping 13% in Haryana Jatts). The amount of WHG-UHG in this calculator directly corresponds to varying rates of "NE-Euro" component in other calculators.

What do you suppose WHG-UHG stands for in South Asians who should be 0% WHG from what I've heard? It looks here like it's complementary to ENF (high WHG-UHG correlates with lower ENF among South Asians or Indians/Pakistanis east of Punjab).

No one ever said South Asians were 0% WHG-UHG. In fact, I seem to remember that Patterson posted something at Dienekes' blog talking about the WHG in South Asia being lumped in with the signal that looks like admix from Stuttgart.

And what the Lazaridis study says is that Near Easterns don't have any more WHG-UHG than Stuttgart. That's not really surprising, since Stuttgart is a European with admixture from European hunter-gatherers.

Another way to call WHG-UHG is West Eurasian, btw, which is what it's called in the ADMIXTUREGRAPH schematic in Laz et al.

Dr_McNinja
09-08-2014, 11:03 AM
Some things I noticed:

Overall much more consistent results for South Asians using these components than the popular calculators.

My family's ASE results are close to everyone else's from the region and ASE doesn't correlate exactly to South Indian in Harappa. In fact, ASE is completely different to South Indian results in my family (my grandmother having the most, my mother having more than my father, etc, see Spreadsheet in my sig for other calculator results for most of these people).

I think I detect a bit of a relationship (no more than 1-2%, maybe even 1% or less) between ANE and East Eurasian in some people. Otherwise this ANE component is very stable and well done by Davidski.

There is potentially a connection between ANE and WHG-UHG going by HRP0370's results. Their ANE seems lower than what everyone else is getting, but their higher WHG-UHG seems more likely to have come from their lower ANE than their normal ENF.

In others (such as Haryanvi Jatts), it seems like WHG-UHG tracks ENF, since that is the major difference between them and their closest relatives, Punjabi Jatt Sikhs.

East Eurasian and WHG-UHG seem like the most fluctuating components which leads me to believe they're linked to the others for South Asians. In my case for example I have lower WHG-UHG than both parents and higher East Eurasian than both parents. Also have higher ASE than both parents, which is interesting since my father has higher S-Indian than both me and my mom in Harappa but here he has the lowest ASE. ANE is really the most stable component because even my ENF is lower than both my parents so it seems like ANE is acting like an anchor in the middle and has connections to the other components (particularly WHG-UHG, East Eurasian, and possibly ENF). In my family among my parents/grandparent also highest/lowest ANE correlates with highest/lowest ASE.

Sangarius
09-08-2014, 11:33 AM
Me
Grandpa
Grandma


ANE
15.24%
14.73%
14.41%


ASE
3.53%
1.71%
1.36%


WHG-UHG
14.17%
8.50%
8.62%


East_Eurasian
3.68%
0.00%
0.00%


West_African
0.02%
0.00%
0.00%


East_African
0.00%
0.00%
0.00%


ENF
63.36%
75.05%
75.59%

NK19191
09-08-2014, 11:38 AM
Here is mine:


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

19.02% ANE
6.00% ASE
9.08% WHG-UHG
0.63% East_Eurasian
0.35% West_African
1.08% East_African
63.84% ENF

Dr_McNinja
09-08-2014, 11:41 AM
Minor African signals can be put down to Basal Eurasian ancestry.



Not really.
So they would be closest to ENF?

Dr_McNinja
09-08-2014, 11:49 AM
Here is mine:


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

19.02% ANE
6.00% ASE
9.08% WHG-UHG
0.63% East_Eurasian
0.35% West_African
1.08% East_African
63.84% ENF

Well that's interesting, your WHG-UHG is higher than some of the South Asians I've posted, but whereas WHG-UHG correlates to relative NE-Euro amounts in those South Asians, yours doesn't manifest as higher European admixture in the other calculators (you had 6.51% in K36 compared to 10-11% European for those South Asians with 5-6% WHG-UHG and 19% European for those with 13% WHG-UHG). DMXX had 10.58% WHG-UHG going with 10.20% European in K36.

Instead it looks like the ANE came into your K36 North Caucasian component.

I suppose it could mean ANE proportion is filtered through WHG-UHG-like proportion for what ADMIXTURE translates it into...

Generalissimo
09-08-2014, 11:51 AM
So they would be closest to ENF?

East African is often interchangeable with ENF for many people, because it's a hybrid cluster.

ANE is quite unique actually, and only really gets mixed up with ASE, but again not by much. There's probably a buffer of 1-2% for most people.

Most of these clusters are really basal, so it's actually a miracle that they still give such good signals via genotype data.

Dr_McNinja
09-08-2014, 11:54 AM
HRP0363 Khetran Baloch

29.82% ANE
12.98% ASE
2.89% WHG-UHG
0.51% East_Eurasian
0.71% West_African
2.95% East_African
50.14% ENF

Dr_McNinja
09-08-2014, 11:58 AM
East African is often interchangeable with ENF for many people, because it's a hybrid cluster.

ANE is quite unique actually, and only really gets mixed up with ASE, but again not by much. There's probably a buffer of 1-2% for most people.

Most of these clusters are really basal, so it's actually a miracle that they still give such good signals via genotype data.Is East Eurasian closer to ASE then? And which would you say WHG-UHG is closer to out of all of these components, ENF?

The ANE numbers look real good honestly, even with the 23andMe V4 kits (~86% genotyping rate).

Generalissimo
09-08-2014, 12:07 PM
Is East Eurasian closer to ASE then? And which would you say WHG-UHG is closer to out of all of these components, ENF?

The ANE numbers look real good honestly, even with the 23andMe V4 kits (~86% genotyping rate).

Yeah, East Eurasian and ASE are closest to each other, but only a bit closer than to ANE and WHG-UHG.

WHG-UHG is closest to ENF, probably because the only difference between them is more "Basal Eurasian" in ENF. Then it's closest to ANE.

ANE is really a well defined and unique component. Surprisingly so.

Varun R
09-08-2014, 12:15 PM
The E Eurasian components seem in line with what I get from other calculators. It's nice to have the ANE quantified.

29.22% ANE
20.05% ASE
2.46% WHG-UHG
5.13% East_Eurasian
1.34% West_African
2.87% East_African
38.93% ENF

MfA
09-08-2014, 12:42 PM
I'll have to wait for this to be available on Gedmatch as I can't download R as I now use a Chromebook rather than a laptop.

try this http://www.r-ohjelmointi.org/?p=1434

Humanist
09-08-2014, 01:03 PM
My results:


14.01 ANE
2.17 ASE
7.68 WHG-UHG
0.02 East_Eurasian
0.06 West_African
0.51 East_African
75.55 ENF

Jusarius
09-08-2014, 01:09 PM
My (East Finnish) result:

20.85% ANE
2.60% ASE
67.65% WHG-UHG
8.18% East_Eurasian
0.02% West_African
0.61% East_African
0.08% ENF

Dr_McNinja
09-08-2014, 01:11 PM
HRP0143 Hanna (Turkish)

15.76% ANE
2.84% ASE
13.01% WHG-UHG
2.99% East_Eurasian
0.20% West_African
0.10% East_African
65.10% ENF

---

Turkish Central Anatolian

14.59% ANE
2.62% ASE
16.50% WHG-UHG
6.69% East_Eurasian
0.76% West_African
1.22% East_African
57.62% ENF

---

North African Berber (Ignis I believe)

0.85% ANE
0.13% ASE
20.80% WHG-UHG
1.72% East_Eurasian
1.10% West_African
23.55% East_African
51.86% ENF

Dr_McNinja
09-08-2014, 01:17 PM
For South Asians, ENF + East African seems to form the best east-west cline, regardless of North or South (soulblighter was in the middle of the pack of otherwise North/Northwest subcontinent people). WHG doesn't correlate to the locations. But including WHG with those two does seem to cluster people by who should be their closest ethnic relatives.

HRP0393, HRP0341, and HRP0402 (Haryanvi Jatt, Punjabi Jatt Sikh, and (Western) Punjabi Jatt Muslim respectively) wind up within 1% of each other in the 51-52% range which has an example of each kind of Jatt. HRP0282 (Sein) is just ahead of them at 52.5%.

HRP0350 and HRP0353, who are otherwise from HRP0402's neck of the woods, are close to one another at around 50%. My family are close to each other and right behind HRP0350/HRP0353 at 48-49%. This seems like the general range for Punjabis.

Interestingly, dividing that (ENF+E-African+WHG) by ASE in percent form (i.e, 9.32% ASE, 61.04% ENF+E-African+WHG ... .0932/.6104 = .1527 = 15.27%) gives a value which seems like a rough estimate of where Harappa S-Indian should or could be. It puts soulblighter at 47% and Everest at 39% (in Harappa Everest, a Nepal Brahmin, had 33% S-Indian but Parasar, a Bihari Brahmin, was at 39%). It puts me at 35%.

So WHG-UHG in subcontinental South Asians seems to correlate along ancestral ethnic (perhaps even caste insofar as endogamy and genetic drift distinguished castes over time), not local geographic lines. I think genetic drift and endogamy caused the spike in WHG-UHG for Haryanvi Jatts since their overall ENF+E-African+WHG is equivalent to other Punjabi Jatts. It hits 7.8% in Everest otherwise, 6.24% in Sapporo, and 5.76% in my dad. North/Western Punjabi Jatts, who probably weren't as closely associated with the caste system (I think most of this area was Buddhist until the advent of Islam anyway, even the Jatts in Sindh were Buddhist when the Arabs attacked) and who probably had a bigger gene pool, are around 4-5%. It's possible WHG-UHG deteriorated into a generic basal/ENF signal in other South Asians who didn't have these kinds of endogamy habits.

There's probably an error % or a standard deviation of a few percent distinguishing between these components though (all the non-ANE ones) and perhaps a tiny bit also for ANE but that seems stable.

Dr_McNinja
09-08-2014, 01:33 PM
Afghan Pashtun (Sein gave me this one, I think this is HRP0281?)

27.39% ANE
13.57% ASE
6.75% WHG-UHG
7.97% East_Eurasian
1.08% West_African
1.37% East_African
41.87% ENF

---

Sein's other friend (I think some combination of Afghan Tajik and Pashtun, name starts with W)

28.04% ANE
9.59% ASE
8.32% WHG-UHG
8.32% East_Eurasian
0.69% West_African
1.02% East_African
44.01% ENF

DMXX
09-08-2014, 02:40 PM
Well that's interesting, your WHG-UHG is higher than some of the South Asians I've posted, but whereas WHG-UHG correlates to relative NE-Euro amounts in those South Asians, yours doesn't manifest as higher European admixture in the other calculators (you had 6.51% in K36 compared to 10-11% European for those South Asians with 5-6% WHG-UHG and 19% European for those with 13% WHG-UHG). DMXX had 10.58% WHG-UHG going with 10.20% European in K36.

Instead it looks like the ANE came into your K36 North Caucasian component.

I suppose it could mean ANE proportion is filtered through WHG-UHG-like proportion for what ADMIXTURE translates it into...

A proportion (not all) of ANE seems to coincide with North Euro frequencies across Asia. Please note f.ex. Assyrians and Armenians have less of it than Iranians and Kurds, who in turn have less of it than South-Central Asians.

Another thought which might give us another perspective on these scores compared with preceding runs; from where I'm sitting, the other proportion of ANE and some of the ENF equate to "Gedrosian". The rest of ENF is broken down into all of what would then become "Caucasian", most of "SW Asian" and some of "Mediterranean". I suspect the elevated WHG-UHG score in Iranians and Kurds is actually the other portion of "Mediterranean" presenting in a more Euro-specific distinct manner. The various African scores are presumably some sort of archaic post-OoA that's ordinarily present in "SW Asian".

I therefore suspect the WHG-UHG in Iranians and Kurds is of a different origin from the South-Central Asians and the similarity with North Euro scores is something of a red herring. Could be wrong though, but the proportions described above make some sense and are in line with various ADMIXTURE experiments which have broken them up relative to one another (e.g. SW Asian being "Caucasian" with something African thrown in).

parasar
09-08-2014, 03:26 PM
Another thought which might give us another perspective on these scores compared with preceding runs; from where I'm sitting, the other proportion of ANE and some of the ENF equate to "Gedrosian" ...

No doubt, that is the case. The Gedrosian cline does not match the ANE cline along the Indus for that reason - Gedrosia goes down going northwards from Baloch to Burusho while ANE goes up.

In the Burusho (Burusho HGDP00444 0.372507) ANE almost gets to Karitiana type numbers (Karitiana_HGDP00998 ANE 41.56%).

Within South Asia (cf Piramalai Kallar PK6415 0.310205) ANE stays consistently high:

Courtesy Dr_McNinja:
Punjabi Jatt Muslim, Central/Eastern Punjab
HRP0349, Me (23andMe V3):
29.23% ANE
HRP0349, Me (FTDNA):
29.25% ANE

HRP0374 (My mother)
29.89% ANE

My father (FTDNA, 97.8% genotyping rate)
29.02% ANE

Grandmother (mother's mother) (FTDNA, genotyping rate 97.5%) (Might be half or 1/4 Western Punjab)
30.09% ANE

Other South Asian members:

HRP0393 (Haryana Jatt) (23andMe V4)
30.16% ANE

HRP0341 (Sapporo) (Punjabi Jatt Sikh)
30.70% ANE

HRP0402 (Punjabi Jatt Muslim, Gondal) (Western Punjab) (23andMe V4)
30.08% ANE

soulblighter (99.4% genotyping rate) (South Indian[Brahman])
28.36% ANE

HRP0315 (Everest) (Nepal Brahmin)
28.59% ANE

HRP0353 (Punjabi Gujjar Muslim) (Western Punjab)
30.94% ANE

HRP0350 (Punjabi Jatt Muslim) (Western Punjab)
29.10% ANE


HRP0282 (Sein) (Pashtun Pakistan)
29.26% ANE

Half Afghan Panjsheri Tajik, Half Afghan Pashtun (23andMe V4)
29.38% ANE

HRP0370 (Afghan Pashtun Popalzai from Kandahar)
26.42% ANE

So ANE hardly looks Indo Aryan spread unless the ANE numbers in Indo-Aryans got depressed in a later period due to an ENF type influx from the northwest.
Even this latter scenario does not explain why a Vadama Brahman would have the same ANE proportion as a Piramalai Kallar.

Who knows, ANE may indeed be the older, thus stable, South Asian component into which others have put a dent.

Hanna
09-08-2014, 03:26 PM
HRP0143 Hanna (Turkish)

15.76% ANE
2.84% ASE
13.01% WHG-UHG
2.99% East_Eurasian
0.20% West_African
0.10% East_African
65.10% ENF

Thanks, McNinja.

Thanks to Sangarius he ran the calculator for my aunt and grandfather as well:

https://fbcdn-sphotos-g-a.akamaihd.net/hphotos-ak-xpa1/v/t1.0-9/10406927_357327351109610_2718107469416654583_n.jpg ?oh=821333fde6614d43b646cc25e7123068&oe=54A3EC94&__gda__=1419977516_4ff89c1decbbdc4345e4f4209cf53f1 3

Dr_McNinja
09-08-2014, 03:59 PM
Parasar, what do you think about this sort of the South Asian list:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=18

I did that by hand. Do you think it's a relative pattern for WHG-UHG? Have you run your data yet? I wonder what you'd get

Ignis90
09-08-2014, 04:01 PM
This is what David said about African scores:


East African peaks in Kenya, as far as I can remember, but it's all over Africa, and it's really mixed with West Eurasian influence. I don't know the proportions, but they're quite high, which means even groups like the Yoruba are probably mixed heavily.

West African peaks among the Pygmies.


For some people, ASE (and African) scores are partially a sign of Basal Eurasian not captured by ENF. About Jusarius (still by David):

Very interesting. This might mean that Finns, especially East Finns, have very little Basal Eurasian ancestry. But you do have some, because it's probably being expressed via the African and ASE scores.

Thus, another sign no population in Western Eurasia is free from Basal Eurasian influence although some get very little of it.

evon
09-08-2014, 04:03 PM
Ok, here are all of us, i assume anything below 1% is noise right?

Me:
15.64% ANE
2.72% ASE
68.31% WHG-UHG
0.44% East_Eurasian
0.29% West_African
0.64% East_African
11.95% ENF

Mother(ftdna):
16.58% ANE
2.39% ASE
67.54% WHG-UHG
0.97% East_Eurasian
0.00% West_African
0.79% East_African
11.74% ENF

Maternal Uncle:
16.38% ANE
3.15% ASE
65.88% WHG-UHG
0.87% East_Eurasian
0.16% West_African
0.73% East_African
12.85% ENF

Maternal Grandmother:
16.68% ANE
3.35% ASE
65.50% WHG-UHG
0.03% East_Eurasian
0.30% West_African
0.57% East_African
13.58% ENF

Paternal Aunt:
16.32% ANE
1.75% ASE
66.89% WHG-UHG
1.59% East_Eurasian
0.57% West_African
0.40% East_African
12.49% ENF

This seems legit, my maternal ancestors do pull south east as i expected due to Romani and Tatar connections along that line, and my paternal aunt pulls both south west, likely Iberian ancestry along that linage and north east as i expected, South Saami connection perhaps, and lastly as usual i pull westward, but it seems so does my mother, i think she has some link to UK dating back to the 1500's..Will need to investigate her DNA further..

Varun R
09-08-2014, 04:10 PM
@ Parasar,

What if the proto-Brahmin groups initially had higher ENF admixture, depressing ANE levels relative to other groups. These were then replenished by the Indo-Aryans following the decline of the IVC, leading to stable ANE values across the various castes. (Just a thought.)

Dr_McNinja
09-08-2014, 04:23 PM
So ANE hardly looks Indo Aryan spread unless the ANE numbers in Indo-Aryans got depressed in a later period due to an ENF type influx from the northwest.
Even this latter scenario does not explain why a Vadama Brahman would have the same ANE proportion as a Piramalai Kallar.

Who knows, ANE may indeed be the older, thus stable, South Asian component into which others have put a dent.ANE could be many thousands of years older in India than Indo-Aryans or even the IVC.

Since this calculator does a decent job of isolating ANE, ASE also consequently gets a little isolated here. Makes me wish we had an ancient genome to use for ASI, that could really clear things up for South Asians with those two separated and leaving, presumably, all that basal/generic Eurasian ENF stuff and East Eurasian as the remaining components.

Shaikorth
09-08-2014, 04:23 PM
This is what David said about African scores:




For some people, ASE (and African) scores are partially a sign of Basal Eurasian not captured by ENF. About Jusarius (still by David):


Thus, another sign no population in Western Eurasia is free from Basal Eurasian influence although some get very little of it.

The ASE in Europe is not necessarily Basal Eurasian noise. Here's lots of results from (North)Eastern Europe, the East Finnish result there has 2% lower ASE than Jusarius but seems to have dumped it more to WHG-UHG than to ENF. We could say that BAE is at noise level there, or at least much lower than in the neighbouring populations in any direction.


ANE
ASE
WHG-UHG
East_Eurasian
West_African
East_African
ENF

Ashkenazi 9,2 0,44 29,79 1,71 0 1,9 56,96
Balt 19,74 3,47 72,34 0,29 0,59 0,23 3,34
Bashkir 22,10 1,41 37,80 27,39 0,38 0,69 10,23
Belarusian-East 18,98 2,62 65,21 1,6 0,47 0,34 10,78
Chuvash 23,43 1,33 45,81 21,96 0,46 0,24 6,77
Erzya 20,63 1,91 60,65 7,45 0,47 0,69 8,2
Finnish-East 19,52 0,75 69 8,72 0,18 0,96 0,87
German-Volga 15,00 1,28 60,02 0,94 0,43 0,41 21,92
Kazah 17,41 2,78 14,24 50,76 0,43 0,74 13,64
Komi 23,06 1,79 53,21 14,63 0,2 0,39 6,72
Mari01 24,4 0,15 45,29 25,78 0 1,55 2,83
Moksha 20,85 2,37 58,31 6,49 0,65 0,13 11,20
Russian_Meshtchyora 20,01 2,7 63,14 3,63 0,57 0,43 9,52
Russian-North 19,74 2,13 62,96 7,13 0,61 0,61 6,82
Russian-North-West 19,12 2,44 67,91 1,99 0,36 0,54 7,64
Russian-South 18,46 2,98 63,69 2,32 0,45 0,05 12,05
Russian-Upper-Volga 19,98 2,16 66,20 4,62 0,25 0,84 5,95
Russian-Ural 21,24 1,78 59,05 11,11 0,38 0,47 5,97
Russian-West 19,23 2,44 66,31 2,72 0,32 0,8 8,18
Selkup 26,12 0 27,24 45,63 0 1 0,01
Tatar-Kazan 19,42 2,08 46,94 16,26 0,27 0,8 14,23
Tatar-Lithuanian 17,59 4,17 45,39 17,51 0,52 0,16 14,66
Tatar-Mishar 19,12 2,32 49,12 14,8 0,09 1,35 13,2
Tatar-Siberian 21,27 2,21 23,48 39,35 0,45 0,6 12,64
Udmurt 23,05 1,68 50,63 16,42 0 2,13 6,09
Ukrainian-Carpathy 15,14 0,83 57,14 2,19 0,04 1,3 23,36
Ukrainian-Center 18,11 1,98 65,44 1,55 0,31 0,3 12,31
Ukrainian-East 17,56 2,49 64,40 0,92 0,43 0,52 13,68
Ukrainian-West 17,29 2,73 63,62 0,92 0,27 0,35 14,82

ZephyrousMandaru
09-08-2014, 04:27 PM
Eurogenes ANE K=7

13.12% ANE
2.46% ASE
8.52% WHG-UHG
0.00% East_Eurasian
0.04% West_African
0.28% East_African
75.58% ENF

ZephyrousMandaru
09-08-2014, 04:31 PM
It's the same component.

Minor African signals can be put down to Basal Eurasian ancestry.

I agree, David mentioned that the ENF cluster is very close related to the 44% Basal Eurasian found in the Stuttgart farmer. So any ancient African admixture is probably being masked inside of ENF.

ZephyrousMandaru
09-08-2014, 04:57 PM
PCA of various participants by EastPole.

http://s28.postimg.org/9rmz1qh99/pca12.png

ZephyrousMandaru
09-08-2014, 05:27 PM
PCA of various participants by EastPole.

http://s28.postimg.org/9rmz1qh99/pca12.png

Updated PCA including myself, Humanist and I are very ENF.

http://s28.postimg.org/w64nhyk0t/pca12.png

ZephyrousMandaru
09-08-2014, 05:38 PM
Once the GED Match version is available, I'll post the results for the two Iraqi Mandaeans. Unless Humanist wants to compute their ADMIXTURE Proportions, I would be interested in seeing them.

Hok
09-08-2014, 05:38 PM
Oh crap i forgot to post my results.



16.30% ANE
1.41% ASE
63.57% WHG-UHG
0.50% East_Eurasian
0.00% West_African
1.26% East_African
16.95% ENF

i have the exact same amount of ANE compared to an average Czech.:)

Sein
09-08-2014, 05:43 PM
Afghan Pashtun (Sein gave me this one, I think this is HRP0281?)

27.39% ANE
13.57% ASE
6.75% WHG-UHG
7.97% East_Eurasian
1.08% West_African
1.37% East_African
41.87% ENF

---

Sein's other friend (I think some combination of Afghan Tajik and Pashtun, name starts with W)

28.04% ANE
9.59% ASE
8.32% WHG-UHG
8.32% East_Eurasian
0.69% West_African
1.02% East_African
44.01% ENF

No, this isn't HRP0281 (I wish I had his raw-data, that guy is a genetic carbon copy of myself on most analyses).

My other friend is a mix of Panjsheri Tajik and Pashtun.

Just a side note, although I think the general ballpark is solid for South Asians, the results are obviously different from what one would have gotten in the original ADMIXTURE run.

For example, in David's actual run, I am 31.31% ANE (the run upon which this calculator is based). And this result seems to be partly due to the fact that my 23andMe data is missing a lot of SNPs, since other Pashtuns (HGDP) are around 33%-35% for this same calculator, and even the Afghan Pashtuns are 32% (and I know that I have more ANE admixture in comparison to them). If your'e South Asian, this calculator is really putting you at 30%-37% ANE, since that is what the South Asian reference populations get (interestingly, the Burusho represent the maximum. Maximum Eurasian ANE admixture occurs in Northern Pakistan, which actually sounds very reasonable. But it is definitely a pan-South Asian phenomenon, even South Indians are around 29%-30% ANE). Lower ANE scores for the personal calculator seem to reflect genotyping rate+the "calculator effect" (especially the calculator effect). But, at least now we can compare each other.

I have a lot of files, so I'll try to run them tonight, and I'll make sure to post the results.

Alanson
09-08-2014, 05:55 PM
My results:
13.17% ANE
3.09% ASE
8.60% WHG-UHG
0.35% East_Eurasian
0.43% West_African
4.04% East_African
70.33% ENF

Salkin
09-08-2014, 06:21 PM
Here are my results:

16.20% ANE
2.69% ASE
67.57% WHG-UHG
0.04% East_Eurasian
0.24% West_African
0.67% East_African
12.59% ENF

Dr_McNinja
09-08-2014, 07:36 PM
No, this isn't HRP0281 (I wish I had his raw-data, that guy is a genetic carbon copy of myself on most analyses).

My other friend is a mix of Panjsheri Tajik and Pashtun.

Just a side note, although I think the general ballpark is solid for South Asians, the results are obviously different from what one would have gotten in the original ADMIXTURE run.

For example, in David's actual run, I am 31.31% ANE (the run upon which this calculator is based). And this result seems to be partly due to the fact that my 23andMe data is missing a lot of SNPs, since other Pashtuns (HGDP) are around 33%-35% for this same calculator, and even the Afghan Pashtuns are 32% (and I know that I have more ANE admixture in comparison to them). If your'e South Asian, this calculator is really putting you at 30%-37% ANE, since that is what the South Asian reference populations get (interestingly, the Burusho represent the maximum. Maximum Eurasian ANE admixture occurs in Northern Pakistan, which actually sounds very reasonable. But it is definitely a pan-South Asian phenomenon, even South Indians are around 29%-30% ANE). Lower ANE scores for the personal calculator seem to reflect genotyping rate+the "calculator effect" (especially the calculator effect). But, at least now we can compare each other.

I have a lot of files, so I'll try to run them tonight, and I'll make sure to post the results.Do you know what your other values (ENF, WHG-UHG, East Eurasian, ASE) were in Davidski's run? In other words, where did that excess admixture go? East Eurasian? ASE? WHG-UHG?

Maybe it's due to a calculator effect but I still think the East Eurasian we're getting here is from ANE as much as ASE. Relative to the rest of my family, my grandmother has the lowest East Eurasian and a nearly equivalent rise in both ANE and ASE. HRP0353 is the same way.

Sein
09-08-2014, 08:13 PM
Do you know what your other values (ENF, WHG-UHG, East Eurasian, ASE) were in Davidski's run? In other words, where did that excess admixture go? East Eurasian? ASE? WHG-UHG?

Maybe it's due to a calculator effect but I still think the East Eurasian we're getting here is from ANE as much as ASE. Relative to the rest of my family, my grandmother has the lowest East Eurasian and a nearly equivalent rise in both ANE and ASE. HRP0353 is the same way.

So far, the only thing I know is that my ANE and ASE scores for the actual run were 31.31% and 14.77%. If possible, it would be great if David could tell me the whole results for the original run. That way, we can see how things change between being a part of the calculator, and being "outside" of it.

Pashtun10_17Af is around 15%-16% "South Indian" for HarappaWorld. In David's ADMIXTURE run, this sample is 32.92% ANE and 12.17% ASE (so 1% more ANE, and 2.5% less ASE). I wouldn't compare myself to any sample, since my results are affected by the quality of my data (damn you 23andMe! :biggrin1:). Rather, I would compare Pashtun10_17Af to HGDP00244 and HGDP00226. These two HGDP samples are the closest ones to myself (according to most PCA-based analyses). They are, respectively, 33.41% and 32.42% ANE. That's probably where I should be. It would be nice if we could find out their ASE percentages as well, to compare with Pashtun10_17Af.

Also, David has been quite clear that this test is really meant to isolate solid ANE percentages. In David's own words:

"... six other ancestral components in this test (hence the K=7 in the name). They're basically byproducts of me trying to isolate ANE, and don't necessarily mean anything."
http://bga101.blogspot.com/2014/09/eurogenes-ane-k7.html

And I think it does that quite well. The other clusters are nice, but they are really there to pinpoint ANE. In that respect, the test is a success. For example, I can say with confidence that despite the calculator effect, this calculator will consistently give South Asians the highest ANE admixture out of all individuals who try this, which is accurate (unless an individual whose genetic heritage is predominantly Native American tries this calculator. In that case, their ANE admixture would be higher than South Asians).

The ASE cluster is very interesting though. It is quite high across much of Eurasia, and some Northeastern Europeans are around 4%.

Edit: I have some free time right now, so I'll post results for a few folks.

Safi Pashtun+Kabuli Tajik (V4 raw-data)
ENF=44.25%
ANE=28.34%
ASE=10.93%
WHG-UHG=8.59%
East_Eurasian=4.98%
East_African=2.45%
West_African=0.46%

Kandahari Pashtun (V4 raw-data. On most calculators, this individual is the most western-shifted Pashtun I've ever seen. Their South Indian score is lower than HRP0370, and their Caucasian score for HarappaWorld is 26%)
ENF=50.04%
ANE=27.99%
ASE=10.48%
WHG-UHG=6.64%
East_Eurasian=3.26%
East_African=1.42%
West_African=0.18%

Ghilzai Pashtun (FF raw-data. Has the highest "Baloch" I've ever seen for a Pashtun on HarappaWorld's DIY calculator, my mother excepted)
ENF=48.27%
ANE=29.09%
ASE=11.35%
WHG-UHG=7.57%
East_Eurasian=2.88%
East_African=0.83%
West_African=0.00%

My mother (V3 raw-data. 75% Afghan Pashtun from Nangarhar+Paktia, 25% Pakistani Pashtun from Khyber+Peshawar)
ENF=46.16%
ANE=30.41%
ASE=14.37%
WHG-UHG=5.44%
East_Eurasian=2.02%
East_African=1.55%
West_African=0.03%

bored
09-08-2014, 08:25 PM
I'm guessing the ASE is not that good of a proxy for ASI.

Humanist
09-08-2014, 09:37 PM
Updated PCA including myself, Humanist and I are very ENF.

Any idea what the Bedouin ENF average is? In the 80s?

ZephyrousMandaru
09-08-2014, 09:59 PM
Any idea what the Bedouin ENF average is? In the 80s?

An example was of a Bedouin was posted on David's blog. It's from the HGDP panel.


Bedouin_HGDP00651
ANE 0%
ASE 0%
WHG-UHG 0.05%
East_Eurasian 1.49%
West_African 0%
East_African 8.19%
ENF 90.27%

I think the general range is between 80-90%. My Egyptian Copt friend who clusters quite closely to Arabian populations is about 78% ENF, but only 12% East African. I think she typically scores more in the 15-16% range for her total African percentage, it may be being absorbed by ENF or possibly even ASE.

ZephyrousMandaru
09-08-2014, 10:03 PM
I wonder, does ASE represent a portion of ASI? Could it be possible that ASI is mostly ANE with this ASE mixed into it?

Dr_McNinja
09-08-2014, 11:19 PM
I'm guessing the ASE is not that good of a proxy for ASI.If you divide the ASE by ENF+East-African+WHG-UHG, it does a decent job predicting a Harappa S-Indian-like component (better than Harappa S-Indian, imo).

I have values for it for some South Asians here:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=18

Compare with the DIY Harappa sheet.

(Keeping in mind Harappa S-Indian is a fundamentally flawed component because half of it is basically Caucasian and significantly EEF/Mediterranean-like, so that's why it would be artificially higher as you get out of South Asia... also because ASE is less linked to ASI as you get out of South Asia)

Dr_McNinja
09-08-2014, 11:22 PM
So far, the only thing I know is that my ANE and ASE scores for the actual run were 31.31% and 14.77%. If possible, it would be great if David could tell me the whole results for the original run. That way, we can see how things change between being a part of the calculator, and being "outside" of it.Then it sounds like it is the East Eurasian which is picking up both ANE and ASE admixture since both of those go up as it goes down. I wonder if there's any such connection between the other components. ANE should be close to WHG-UHG too, but it looks like just the East Eurasian is enough to get everyone to their values in David's admixture run.

Dr_McNinja
09-08-2014, 11:29 PM
Some of the people on David's blog are suggesting ENF is too low and WHG-UHG too high for Europeans. That might also be the case with South Asians I suppose. But whatever that WHG-UHG signal is, I do think it is distinct from ENF since it screws up the geographical gradient that ENF+East-African otherwise neatly fall into and starts clustering people in more ancestral ethnic terms (all Jatts together for instance).

ZephyrousMandaru
09-08-2014, 11:36 PM
EastPole's PCA of the Eurogenes K=7 components.

http://s12.postimg.org/mwib6z3do/pca12_L.jpg

bored
09-08-2014, 11:38 PM
If you divide the ASE by ENF+East-African+WHG-UHG, it does a decent job predicting a Harappa S-Indian-like component (better than Harappa S-Indian, imo).

I have values for it for some South Asians here:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=18

Compare with the DIY Harappa sheet.

(Keeping in mind Harappa S-Indian is a fundamentally flawed component because half of it is basically Caucasian and significantly EEF/Mediterranean-like, so that's why it would be artificially higher as you get out of South Asia... also because ASE is less linked to ASI as you get out of South Asia)

Looks like most of the South Indian scores are bumped up using your method. everest went up quite a bit compared to Harappa DIY. Sapporo's went down a little bit!

Dr_McNinja
09-09-2014, 12:57 AM
Looks like most of the South Indian scores are bumped up using your method. everest went up quite a bit compared to Harappa DIY. Sapporo's went down a little bit!Yeah, Everest's went up to be within the vicinity of Parasar's. It's more of a linear gradient because it's based on the ENF portion, rather than trying to measure ASI, which prevents people from pulling to one direction or the other. Having ANE cleared out also helps stabilize it. From my own admixture runs I know Sapporo's is supposed to be a little lower.

ZephyrousMandaru
09-09-2014, 03:07 AM
An East Assyrian's Eurogenes ANE K=7 results.


13.37% ANE
2.84% ASE
10.46% WHG-UHG
0.00% East_Eurasian
0.01% West_African
0.17% East_African
73.14% ENF

Generalissimo
09-09-2014, 03:50 AM
It's useful to keep in mind that these are point estimates only, and there are standard errors in all of these tests, like below for the Karitiana.

https://lh4.googleusercontent.com/-zaXRlS2HyHk/U0pyWIVkieI/AAAAAAAAATA/1YX872RZ2VA/w365-h576-no/MA-1_Fig_2.png

ZephyrousMandaru
09-09-2014, 04:09 AM
It's useful to keep in mind that these are point estimates only, and there are standard errors in all of these tests, like below for the Karitiana.

https://lh4.googleusercontent.com/-zaXRlS2HyHk/U0pyWIVkieI/AAAAAAAAATA/1YX872RZ2VA/w365-h576-no/MA-1_Fig_2.png

Do you think that ENF could be masking some archaic East African admixture?

Generalissimo
09-09-2014, 05:27 AM
Do you think that ENF could be masking some archaic East African admixture?

Probably not, but I think the East African is definitely masking some archaic West Eurasian ancestry, or something that is shared between West Eurasians and most Africans.

soulblighter
09-09-2014, 01:18 PM
So ANE hardly looks Indo Aryan spread unless the ANE numbers in Indo-Aryans got depressed in a later period due to an ENF type influx from the northwest.
Even this latter scenario does not explain why a Vadama Brahman would have the same ANE proportion as a Piramalai Kallar.

Who knows, ANE may indeed be the older, thus stable, South Asian component into which others have put a dent.


@ Parasar,

What if the proto-Brahmin groups initially had higher ENF admixture, depressing ANE levels relative to other groups. These were then replenished by the Indo-Aryans following the decline of the IVC, leading to stable ANE values across the various castes. (Just a thought.)


This has always been my intuition, that ANE like admixture in south Asia is archaic (one possibility among many is that proto-ANE moved north into central Asia from south Asia).
Regarding Indo-Aryans, I feel it would be very difficult to associate one or another allele frequency to them based on testing populations today, possibly because of the following reasons:

1) The Indo-Aryans were more or less south-central Asian populations and hence already had significant ANE like (and possibly even ASI like) admixture when they were located in far northern south Asia (when the first verses of the Rig Veda were composed).
2) The caste system was extremely fluid in the past (probably as social organization based on division of labor in a huge society to keep the wheels turning), and only solidified in the early MahaJanapada/Northern black polished ware era (This idea seems to be corroborated though various publications), and hence the variation we see between extant populations of India is just genetic drift over the centuries in a highly endogamous culture emanating from a common mixed source population.

NK19191
09-09-2014, 03:28 PM
This is user Jesus's results, he is an Iranian-Kuwaiti.


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

17.82% ANE
7.04% ASE
5.73% WHG-UHG
0.08% East_Eurasian
1.31% West_African
6.08% East_African
61.94% ENF

Dr_McNinja
09-09-2014, 08:21 PM
It's uploaded to GEDmatch but it's set to 1e-6 so the results are coming out quite different (in every component except ANE which is still very close) than what we've posted in this thread so far.

MfA
09-09-2014, 08:38 PM
.....

Humanist
09-09-2014, 08:39 PM
It's uploaded to GEDmatch but it's set to 1e-6 so the results are coming out quite different (in every component except ANE which is still very close) than what we've posted in this thread so far.

Yes. This looks very off. Here is the Iraqi Mandaean who ordinarily has values roughly approximate to my own:



ANE 18.66%
ASE 6.45%
WHG-UHG 22.12%
East_Eurasian 3.28%
West_African 2.02%
East_African -
ENF 47.47%

jesus
09-09-2014, 08:40 PM
This is user Jesus's results, he is an Iranian-Kuwaiti.


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

17.82% ANE
7.04% ASE
5.73% WHG-UHG
0.08% East_Eurasian
1.31% West_African
6.08% East_African
61.94% ENF


My results from GEDmatch.

Population
ANE 19.14%
ASE 7.25%
WHG-UHG 12.45%
East_Eurasian 1.99%
West_African 3.40%
East_African 3.75%
ENF 52.01%

Dr_McNinja
09-09-2014, 08:42 PM
Well even setting DIYDodecad to 1d-6 doesn't duplicate GEDmatch's result so I have no idea why it's doing that. It does say my FTDNA kit has 224k SNPs in the test and 23andMe V3 has 219k but in DIYDodecad they were both around 219k (out of 229k total in the test).

parasar
09-09-2014, 09:16 PM
Parasar, what do you think about this sort of the South Asian list:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=18

I did that by hand. Do you think it's a relative pattern for WHG-UHG? Have you run your data yet? I wonder what you'd get

I have lower West Eurasian affinity.
My WHG-UHG is like the Haryana Jats.

ANE 29.49%
ASE 16.62%
WHG-UHG 13.24%
East_Eurasian 9.97%
West_African 2.45%
East_African 0.96%
ENF 27.27%

Dr_McNinja
09-09-2014, 09:20 PM
I have lower West Eurasian affinity.
My WHG-UHG is like the Haryana Jats.

ANE 29.49%
ASE 16.62%
WHG-UHG 13.24%
East_Eurasian 9.97%
West_African 2.45%
East_African 0.96%
ENF 27.27%Everyone's WHG-UHG jumped in the new GEDmatch: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

Not sure why

Shaikorth
09-09-2014, 09:27 PM
It went down for all North Europeans. ANE shot up, and ASE is over 4% everywhere. Here's a west european result to demonstrate:


ANE 20.18%
ASE 4.39%
WHG-UHG 49.09%
East_Eurasian 1.38%
West_African -
East_African -
ENF 24.96%

This individual is German + tiny amount of some South European in the K13 oracle. The ANE is way up there now.

parasar
09-09-2014, 09:30 PM
Everyone's WHG-UHG jumped in the new GEDmatch: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

Not sure why

Wow! Those are huge numbers - looks to have come at the expense of ENF.

Rukha
09-09-2014, 09:46 PM
Everyone's WHG-UHG jumped in the new GEDmatch: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

Not sure why

yep here's mine:

ANE 29.52%
ASE 10.09%
WHG-UHG 14.76%
East_Eurasian 4.75%
West_African 1.59%
East_African -
ENF 39.26%

leonardo
09-09-2014, 10:54 PM
Mine:
ANE 21.00%
ASE 2.87%
WHG-UHG 43.35%
East_Eurasian 4.01%
West_African -
East_African -
ENF 28.77%

Mother's:
ANE 21.11%
ASE 2.42%
WHG-UHG 45.45%
East_Eurasian 4.10%
West_African -
East_African -
ENF 26.91%

Not much difference. I imagine the 1/8 Italian from my father is why my ENF is larger than my mother's. Other than this, it appears that my mother and my father have similar ancestry, although I know my mother has more British Isles ancestry.

icebreaker
09-09-2014, 11:06 PM
R:
16.51% ANE
3.50% ASE
14.14% WHG-UHG
5.60% East_Eurasian
0.51% West_African
0.64% East_African
59.09% ENF

Gedmatch:
ANE 20.57%
ASE 5.15%
WHG-UHG 22.15%
East_Eurasian 8.42%
West_African 1.92%
East_African -
ENF 41.79%

ZephyrousMandaru
09-09-2014, 11:21 PM
Population
ANE 18.73%
ASE 5.76%
WHG-UHG 22.05%
East_Eurasian 1.89%
West_African 1.53%
East_African -
ENF 50.04%

R.Rocca
09-09-2014, 11:32 PM
My results (5/8 Italy + 3/8 Spain)

Population
ANE 15.34%
ASE 4.07%
WHG-UHG 39.71%
East_Eurasian 1.94%
West_African 1.27%
East_African -
ENF 37.66%

Wife's Results (100% Polish Ancestry)

Population
ANE 23.31%
ASE 4.56%
WHG-UHG 48.36%
East_Eurasian 2.24%
West_African -
East_African -
ENF 21.54%

Dr_McNinja
09-09-2014, 11:53 PM
Wow! Those are huge numbers - looks to have come at the expense of ENF.Yeah. If you want, you can e-mail me your data and I'll run it with the original calculator.

Ignis90
09-10-2014, 12:02 AM
With gedmatch:



ANE 9.67%
ASE 3.68%
WHG-UHG 25.31%
East_Eurasian 4.91%
West_African 7.48%
East_African 13.60%
ENF 35.34%


From no ANE to almost 10%!

bored
09-10-2014, 12:03 AM
Looks like everyone's ANE increased. How much ANE are South Asians getting now?

MitchellSince1893
09-10-2014, 12:11 AM
Mine
WHG-UHG 47.90%
ENF 24.27%
ANE 21.45%
ASE 3.86%
East_Eurasian 2.51%
West_African -
East_African -


Mother's
WHG-UHG 47.68%
ENF 24.79%
ANE 21.04%
ASE 4.39%
East_Eurasian 1.50%
East_African 0.52%
West_African 0.08%


Father's
WHG-UHG 47.31%
ENF 25.40%
ANE 20.85%
ASE 3.98%
East_Eurasian 2.46%
West_African -
East_African -


Father's maternal half brother
WHG-UHG 45.81%
ENF 25.86%
ANE 20.72%
ASE 4.42%
East_Eurasian 3.19%
West_African -
East_African -


In my amateur opinion, I'm suspicious whenever I see admixture results where my percentages are not in between my parents results. In Eurogenes ANE_K7, each of my parents is within 1.2% of my results, so it could just be within the margin of error.

Eurogenes K13 is one of the few admixture tools where my results are in between my parents. I therefore put more stock in it...again just my novice observations.

Gray Fox
09-10-2014, 12:18 AM
Here's mine


ANE 20.85%
ASE 4.92%
WHG-UHG 45.62%
East_Eurasian 2.04%
West_African 0.59%
East_African -
ENF 25.97%

Stephen1986
09-10-2014, 12:36 AM
My results using Gedmatch -

WHG-UHG 46.17%
ENF 24.83%
ANE 21.40%
East_Eurasian 4.02%
ASE 3.57%

My brother's results using Gedmatch -

WHG-UHG 46.61%
ENF 25.57%
ANE 20.54%
ASE 4.55%
East_Eurasian 2.66%
West_African 0.07%

Dr_McNinja
09-10-2014, 01:01 AM
Looks like everyone's ANE increased. How much ANE are South Asians getting now?The same actually, less than 1% increase.

DMXX
09-10-2014, 01:04 AM
My results on GEDmatch are completely off aside from the ANE, although that's been bumped up by 4% compared to the ANE K=7 DIY calculator version:



ANE 22.07%
ASE 7.36%
WHG-UHG 22.20%
East_Eurasian 5.99%
West_African 2.45%
East_African -
ENF 39.93%

bored
09-10-2014, 01:11 AM
The same actually, less than 1% increase.

So it seems that everyone got a significant increase in ANE except South Asians. I wonder what that means? Too much discrepancy between GEDmatch and DIY overall it seems.

Sapporo
09-10-2014, 01:42 AM
My Gedmatch results:

Population
ANE 30.90%
ASE 14.20%
WHG-UHG 14.03%
East_Eurasian 6.83%
West_African 1.61%
East_African 1.00%
ENF 31.44%

Stephen1986
09-10-2014, 02:15 AM
The Gedmatch version of this test isn't working - http://www.forumbiodiversity.com/showthread.php/41453-Eurogenes-v3?p=1157937&viewfull=1#post1157937.

Táltos
09-10-2014, 02:34 AM
The Gedmatch version of this test isn't working - http://www.forumbiodiversity.com/showthread.php/41453-Eurogenes-v3?p=1157937&viewfull=1#post1157937.
Oh drat! Well that must explain why I'm getting really different results for myself between my FTDNA kit and 23andme kit.

My FTDNA:
ANE 20.34%
ASE 4.19%
WHG-UHG 43.27%
East_Eurasian 3.71%
West_African 0.21%
East_African -
ENF 28.28%

My 23andme:
ANE 17.08%
ASE 3.49%
WHG-UHG 52.41%
East_Eurasian 0.36%
West_African 0.39%
East_African -
ENF 26.27%

Solothurn
09-10-2014, 03:09 AM
I could only see Eurogenes_ANE K7 Admixture Proportions at Gedmatch. Here are mine



ANE
21.52%


ASE
2.75%


WHG-UHG
47.01%


East_Eurasian
3.79%


West_African
-


East_African
-


ENF
24.92%

Hanna
09-10-2014, 03:35 AM
My grandfather and aunt added in the PCA:

https://www.dropbox.com/s/ehy955lkkn7vdnw/PCAnames6.pdf?dl=0

Dr_McNinja
09-10-2014, 03:58 AM
It looks like the GEDmatch version of this test is way off for many people.

There's nothing I can do about that. You guys need to talk to John.Anyone know John from GEDmatch?

Salkin
09-10-2014, 07:27 AM
Quoted, my DIY result:


16.20% ANE
2.69% ASE
67.57% WHG-UHG
0.04% East_Eurasian
0.24% West_African
0.67% East_African
12.59% ENF

And GEDmatch's dodgy one:

Population
ANE 18.38%
ASE 3.51%
WHG-UHG 59.50%
East_Eurasian 0.51%
West_African -
East_African 0.31%
ENF 17.79%

EDIT: clarified which results are which.

Dr_McNinja
09-10-2014, 11:01 AM
In spite of GEDmatch's screwed results, the ANE results for South Asians are remarkable consistent/close:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

J Man
09-10-2014, 01:37 PM
Use the DIY version for accurate results. GEDmatch is clearly not accurate when it comes to the ANE K=7 calculator.

Dr_McNinja
09-10-2014, 02:50 PM
GEDmatch took it down

ZephyrousMandaru
09-10-2014, 04:40 PM
Updated PCA by EastPole.

https://www.dropbox.com/s/1la07yjpu9wnm28/PCAnames7.pdf?dl=0

Humanist
09-10-2014, 04:46 PM
Updated PCA by EastPole.

https://www.dropbox.com/s/1la07yjpu9wnm28/PCAnames7.pdf?dl=0

Who is the Iraqi Arab?

http://i1096.photobucket.com/albums/g326/dok101/DNA_Tribes_SNP_Analysis/east_pole.jpg

ZephyrousMandaru
09-10-2014, 05:23 PM
Who is the Iraqi Arab?

http://i1096.photobucket.com/albums/g326/dok101/DNA_Tribes_SNP_Analysis/east_pole.jpg

It's my friend Peradam from 23andMe, he's from Baghdad.

Alanson
09-10-2014, 05:47 PM
I wonder where I would plot on that plot. Since I have both Anatolian/Kavkazian like ancestry and an Arabian one.

Humanist
09-10-2014, 06:34 PM
It's my friend Peradam from 23andMe, he's from Baghdad.

Is that the same Iraqi Arab (from Baghdad) as from this past Eurogenes plot?

http://i1096.photobucket.com/albums/g326/dok101/dok101023/Eurogenes_v2_MDS_ME__.jpg

Alanson
09-10-2014, 06:47 PM
I think it depends on where the Iraqi Arab is from, those from the South especially near Saudi Arabia will be Bedouin like and are Bedouins regardless of sectarian affiliation. Well those in the North and center seem to be Arabized with a hint of Iranian ancestry, and those in the West are largely of Bedouin stock hence why they are genetically close to Jordanians and than to Saudis, but are also genetically close to Saudis. Jordanians and Saudis are not that much genetically different in the first place.

Humanist
09-10-2014, 06:54 PM
Well those in the North and center seem to be Arabized with a hint of Iranian ancestry...

Iranian and/or Kurdish ancestry perhaps, given the large Kurdish population that is now found in N Iraq.

vettor
09-10-2014, 07:02 PM
John Olsen and his working partner are not silly people..............maybe they corrected the eurogenes program

ZephyrousMandaru
09-10-2014, 08:17 PM
Is that the same Iraqi Arab (from Baghdad) as from this past Eurogenes plot?

http://i1096.photobucket.com/albums/g326/dok101/dok101023/Eurogenes_v2_MDS_ME__.jpg

Honestly, I'm not sure. Do you know when this plot was done?

Humanist
09-10-2014, 08:27 PM
Honestly, I'm not sure. Do you know when this plot was done?

I believe in 2012, but I am not certain.

ZephyrousMandaru
09-10-2014, 11:08 PM
PCA Updated.

https://www.dropbox.com/s/ywd5cxxghbt9zrn/pcaNames8.pdf?dl=0

jeanL
09-10-2014, 11:34 PM
Here are the results of my paternal grandmother from DIY:

9.56% ANE
0.05% ASE
47.78% WHG-UHG
4.27% East_Eurasian
0.38% West_African
5.63% East_African
32.31% ENF

What gives with the high East African? Also I've noticed that Northern Europeans keep getting ASE in the order of 1-2%, what's up with that?

jeanL
09-10-2014, 11:35 PM
Here is my mother's results from DIY:

8.39% ANE
0.69% ASE
50.60% WHG-UHG
2.06% East_Eurasian
1.15% West_African
4.58% East_African
32.52% ENF

Again what's with the high East_African?

Generalissimo
09-11-2014, 12:00 AM
What gives with the high East African? Also I've noticed that Northern Europeans keep getting ASE in the order of 1-2%, what's up with that?

Again what's with the high East_African?

Do you honestly think these are unusual results considering what we've learned from ancient genomes lately?

Are you really unable to put them into a context?

jeanL
09-11-2014, 12:15 AM
Do you honestly think these are unusual results considering what we've learned from ancient genomes lately?

Are you really unable to put them into a context?

Hey man chill out! I'm not implying anything, I was simply curious by that fact. So do you think the ASE is acting as some sort of proxy for some other component that the 7 components did not capture the variation of it?

ZephyrousMandaru
09-11-2014, 12:26 AM
I don't think the presence of African admixture is unreasonable at all. While I don't register African admixture on all tests, I have scored African admixture on multiple occasions. At this point in time, I think that the African admixture, especially in Middle Easterners is archaic and obscured within other components.

jeanL
09-11-2014, 12:36 AM
I don't think the presence of African admixture is unreasonable at all. While I don't register African admixture on all tests, I have scored African admixture on multiple occasions. At this point in time, I think that the African admixture, especially in Middle Easterners is archaic and obscured within other components.

Well my grandmother scores 1.6% SSA in 23andme, according to 23andme 1.5% is West African, but according to different calculators she scores differently, i.e. when there is a North African reference her SSA is very small or 0, sometimes she gets East African instead. She is of Canary Islander ancestry, so it is very likely she carries a significant(McDonald gave her ~15% North African, 80% Western European, 5% Native American) portion of native Guanche ancestry.

My mom on the other hand scores 2.0% SSA in 23andme, out of which 1.5% is West African, she consistently scores West African in other test, from as low as 2% in Eurogenes K36 to as high as 4% in Globe13. Her North African score is very small though, she has between 1-4% according to different calculators, yet in here is scoring a whole lot more East African than West African.

ZephyrousMandaru
09-11-2014, 12:43 AM
Is that the same Iraqi Arab (from Baghdad) as from this past Eurogenes plot?

http://i1096.photobucket.com/albums/g326/dok101/dok101023/Eurogenes_v2_MDS_ME__.jpg

Yeah, he isn't, he received his results in late February of 2013.

R.Rocca
09-12-2014, 12:33 PM
It looks like the Eurogenes K=7 calculator has been corrected in GEDmatch...

My Results (5/8 Italy + 3/8 Spain)



ANE:
8.86%


ASE:
1.30%


WHG-UHG:
45.51%


East_Eurasian:
-


West_African:
1.24%


East_African:
0.67%


ENF:
42.42%



Wife's Results (100% Polish Ancestry)



ANE:
18.90%


ASE:
3.13%


WHG-UHG:
67.24%


East_Eurasian:
-


West_African:
0.83%


East_African:
0.23%


ENF:
9.67%

Stellaritic
09-12-2014, 12:44 PM
My results (revised version):


ANE -
ASE -
WHG-UHG 18.10%
East_Eurasian 1.76%
West_African 0.82%
East_African 19.10%
ENF 60.22%

Salkin
09-12-2014, 12:45 PM
It looks like the Eurogenes K=7 calculator has been corrected in GEDmatch...

Yes indeed, John Olson announced in the 23andMe K=7 thread a couple of hours ago that it's back up and running and looks right now.

Dr_McNinja
09-12-2014, 01:08 PM
Results are closer to the DIY but it's using significantly less SNPs for some reason. 180k for both 23andMe V3 and FTDNA FF whereas it was 219k out of 229k on DIY.

Here are some South Asians: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

soulblighter
09-12-2014, 01:46 PM
GedMatch (corrected version):

My Results

ANE 28.47%
ASE 20.63%
WHG-UHG 0.70%
East_Eurasian 6.46%
West_African -
East_African 4.72%
ENF 39.02%


My Mom's results:
ANE 28.50%
ASE 20.19%
WHG-UHG 1.32%
East_Eurasian 6.08%
West_African 0.17%
East_African 4.74%
ENF 38.99%

Humanist
09-12-2014, 01:51 PM
My results:


14.01 ANE
2.17 ASE
7.68 WHG-UHG
0.02 East_Eurasian
0.06 West_African
0.51 East_African
75.55 ENF


Yes. This looks very off. Here is the Iraqi Mandaean who ordinarily has values roughly approximate to my own:



ANE 18.66%
ASE 6.45%
WHG-UHG 22.12%
East_Eurasian 3.28%
West_African 2.02%
East_African -
ENF 47.47%

IQ Mandaean 1

ANE 13.16%
ASE 2.88%
WHG-UHG 7.46%
East_Eurasian -
West_African 0.08%
East_African 0.43%
ENF 76.00%


IQ Mandaean 2

ANE 13.20%
ASE 2.37%
WHG-UHG 5.07%
East_Eurasian -
West_African -
East_African 1.47%
ENF 77.89%

Rukha
09-12-2014, 02:33 PM
corrected gedmatch results:

ANE 29.32%
ASE 10.54%
WHG-UHG 9.84%
East_Eurasian 3.02%
West_African 0.17%
East_African 1.71%
ENF 45.41%

Stephen1986
09-12-2014, 03:26 PM
My results using Gedmatch -

WHG-UHG 46.17%
ENF 24.83%
ANE 21.40%
East_Eurasian 4.02%
ASE 3.57%

My brother's results using Gedmatch -

WHG-UHG 46.61%
ENF 25.57%
ANE 20.54%
ASE 4.55%
East_Eurasian 2.66%
West_African 0.07%

Here are our corrected Gedmatch results -

Myself

WHG-UHG 65.34%
ENF 15.92%
ANE 15.57%
ASE 1.52%
East_African 1.47%
West_African 0.19%

My brother

WHG-UHG 63.00%
ENF 17.86%
ANE 15.44%
ASE 2.20
East_African 1.19%
West_African 0.32%

parasar
09-12-2014, 04:31 PM
GedMatch (corrected version):

My Results

ANE 28.47%
ASE 20.63%
WHG-UHG 0.70%
East_Eurasian 6.46%
West_African -
East_African 4.72%
ENF 39.02%


My Mom's results:
ANE 28.50%
ASE 20.19%
WHG-UHG 1.32%
East_Eurasian 6.08%
West_African 0.17%
East_African 4.74%
ENF 38.99%

Remarkably the main distinction between the above results and mine (and perhaps between north and south India in general) is in the WHG-UHG %age.

Population
ANE 29.12%
ASE 18.23%
WHG-UHG 6.16%
East_Eurasian 5.33%
West_African -
East_African 3.11%
ENF 38.05%

Dr_McNinja
09-12-2014, 04:38 PM
Remarkably the main distinction between the above results and mine (and perhaps between north and south India in general) is in the WHG-UHG %age.

Population
ANE 29.12%
ASE 18.23%
WHG-UHG 6.16%
East_Eurasian 5.33%
West_African -
East_African 3.11%
ENF 38.05%
I sorted the list here: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

ViktorL1
09-12-2014, 04:41 PM
My results:

Population
ANE 21.81%
ASE 4.40%
WHG-UHG 14.82%
East_Eurasian 31.14%
West_African 0.27%
East_African 0.80%
ENF 26.75%

I thought that ANE was generally high in Central Asians. My East Eurasian is higher though.

ZephyrousMandaru
09-12-2014, 04:49 PM
IQ Mandaean 1

ANE 13.16%
ASE 2.88%
WHG-UHG 7.46%
East_Eurasian -
West_African 0.08%
East_African 0.43%
ENF 76.00%


IQ Mandaean 2

ANE 13.20%
ASE 2.37%
WHG-UHG 5.07%
East_Eurasian -
West_African -
East_African 1.47%
ENF 77.89%

I sent John the modified Eurogenes ANE K=7 file with the Eurogenes EUtest V2 K15 SNPs, it seems to has resolved the issue.



Population
ANE 13.28%
ASE 2.53%
WHG-UHG 8.83%
East_Eurasian -
West_African -
East_African 0.36%
ENF 75.00%

Sapporo
09-12-2014, 05:04 PM
Corrected Gedmatch:

Population
ANE 30.43%
ASE 15.14%
WHG-UHG 6.33%
East_Eurasian 3.02%
West_African -
East_African 2.47%
ENF 42.61%

David Mc
09-12-2014, 05:45 PM
My results from the updated K7 Test are below.I tend to identify myself as simply "British." My father is Northern Irish (the descendant of Lowland Scots and English settlers, although the paternal line seems to have come to Ireland from the Isle of Man sometime around 1700 AD). My mother is English, with lines coming in from East Anglia, Middlesex (London), Wiltshire and Devon, (but her maternal great grandfather was "Irish-Irish," so...). Anyway, the results:

Population
ANE 15.19%
ASE 1.88%
WHG-UHG 63.59%
East_Eurasian 0.41%
West_African 0.24%
East_African 0.50%
ENF 18.20%

David Mc
09-12-2014, 05:54 PM
Can I ask anyone who knows about such things, are my results fairly consistent with a broadly British Isles origin? I am unaware of any non-British lines, but I am seeing a higher level of ENF than I would have expected. I also tend to show higher Mediterranean than I would expect in other tests. For example, in my Eurogenes K13 Oracle results I have:

Admix Results
1 North_Atlantic 48.43
2 Baltic 24.44
3 West_Med 14.74
4 East_Med 4.2
5 West_Asian 4.16
6 South_Asian 1.91
7 Red_Sea 0.88
8 Oceanian 0.71
9 Amerindian 0.32
10 Northeast_African 0.14
11 Sub-Saharan 0.08

The South Asian is also higher than I would have expected, but then I'm on a fairly steep learning curve as far as how these admixtures work.

I suppose I have two questions, really:

Are my results within the spectrum one expects from the Isles?

If not, is it likely just a case of random admixture in my own DNA or is it more likely that one of my more distant ancestors was French, Spanish, or perhaps Jewish (or Indian for that matter)?

I should say that I'm hoping to get both of my parents tested, as well. Thanks in advance.

dp
09-12-2014, 06:14 PM
GEDmatch for FN87877: 2574
ANE 14.39%
ASE 1.02%
WHG-UHG 63.14%
East_Eurasian 0.94%
West_African 0.22%
East_African 1.22%
ENF 19.07%

GEDmatch for F345328: 2575
ANE 14.70%
ASE 1.96%
WHG-UHG 64.18%
East_Eurasian 0.19%
West_African 0.62%
East_African 1.32%
ENF 17.03%

vettor
09-12-2014, 06:26 PM
my wife's numbers

North-East italian - Austrian lands

Population
ANE 11.88%
ASE 1.20%
WHG-UHG 52.35%
East_Eurasian 0.72%
West_African 0.65%
East_African 0.17%
ENF 33.02%

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

my numbers below

North-East italian and tyrol areas

Population
ANE 11.49%
ASE 1.09%
WHG-UHG 49.96%
East_Eurasian 0.53%
West_African 0.37%
East_African 0.64%
ENF 35.92%


the naming I used is to distinguish that I sit on the western side of the eastern alps while my wife sits east of me

dp
09-12-2014, 06:33 PM
from David Mc;51638:


Eurogenes K13
Admix Results Irish French SE_English SW_English
1 North_Atlantic 48.43 52.23 42.60 50.52 51.98
2 Baltic 24.44 24.02 17.49 23.27 22.1
3 West_Med 14.74 12.39 19.50 13.98 15.1
4 East_Med 4.2 1.41 10.33 5.09 2.7
5 West_Asian 4.16 6.32 4.96 4.46 4.92
6 South_Asian 1.91 1.09 1.08 1.08 .88
7 Red_Sea 0.88 .82 2.82 .09 .95
8 Oceanian 0.71 .33 .23 .62 .59
9 Amerindian 0.32 .99 .38 .62 .15
10 Northeast_African 0.14 .18 .07 .12 .28
11 Sub-Saharan 0.08 .01 .12 .04 .16

Are my results within the spectrum one expects from the Isles?

If not, is it likely just a case of random admixture in my own DNA or is it more likely that one of my more distant ancestors was French, Spanish, or perhaps Jewish (or Indian for that matter)?

>>Hope this helps... Looks more (SE) English to me than Irish. dp :ranger:
PS: the spreadsheet is at https://docs.google.com/spreadsheet/ccc?key=0Ato3EYTdM8lQdEUtZjRwTkQxRzBCeHdTaTdWUUY4Z 0E#gid=0

Solothurn
09-12-2014, 06:35 PM
Corrected version :)

ANE 16.12%
ASE 1.29%
WHG-UHG 63.30%
East_Eurasian 1.09%
West_African -
East_African 0.55%
ENF 17.65%



I could only see Eurogenes_ANE K7 Admixture Proportions at Gedmatch. Here are mine



ANE
21.52%


ASE
2.75%


WHG-UHG
47.01%


East_Eurasian
3.79%


West_African
-


East_African
-


ENF
24.92%

vettor
09-12-2014, 06:37 PM
would it be fair to say, that as of eurogenes ENF samples, .......a ENF in the thirty to forty numbers range would be danubian/central european?

Táltos
09-12-2014, 06:56 PM
Oh drat! Well that must explain why I'm getting really different results for myself between my FTDNA kit and 23andme kit.

My FTDNA:
ANE 20.34%
ASE 4.19%
WHG-UHG 43.27%
East_Eurasian 3.71%
West_African 0.21%
East_African -
ENF 28.28%

My 23andme:
ANE 17.08%
ASE 3.49%
WHG-UHG 52.41%
East_Eurasian 0.36%
West_African 0.39%
East_African -
ENF 26.27%
Here are my updated results. Much better, as there is not so much discrepancy between my FTDNA and 23andme data.

FTDNA:
ANE 14.69%
ASE 2.54%
WHG-UHG 57.87%
East_Eurasian -
West_African 0.57%
East_African 0.51%
ENF 23.82%

23andme:
ANE 14.59%
ASE 2.53%
WHG-UHG 57.94%
East_Eurasian -
West_African 0.52%
East_African 0.40%
ENF 24.02%

David Mc
09-12-2014, 07:18 PM
That is extremely helpful, David. Thank you so much. Apart from a tiny spike in South Asian, and a slightly lower ANE, I seem to mostly be a blend of the Isles regions.


from David Mc;51638:


Eurogenes K13
Admix Results Irish French SE_English SW_English
1 North_Atlantic 48.43 52.23 42.60 50.52 51.98
2 Baltic 24.44 24.02 17.49 23.27 22.1
3 West_Med 14.74 12.39 19.50 13.98 15.1
4 East_Med 4.2 1.41 10.33 5.09 2.7
5 West_Asian 4.16 6.32 4.96 4.46 4.92
6 South_Asian 1.91 1.09 1.08 1.08 .88
7 Red_Sea 0.88 .82 2.82 .09 .95
8 Oceanian 0.71 .33 .23 .62 .59
9 Amerindian 0.32 .99 .38 .62 .15
10 Northeast_African 0.14 .18 .07 .12 .28
11 Sub-Saharan 0.08 .01 .12 .04 .16

Are my results within the spectrum one expects from the Isles?

If not, is it likely just a case of random admixture in my own DNA or is it more likely that one of my more distant ancestors was French, Spanish, or perhaps Jewish (or Indian for that matter)?

>>Tried the best I could do. My spreadsheet of the populations is not coming up well.
dp :-)

Salkin
09-12-2014, 07:31 PM
Updated GEDmatch K=7:

Population
ANE 15.96%
ASE 2.73%
WHG-UHG 67.22%
East_Eurasian -
West_African 0.24%
East_African 0.97%
ENF 12.89%

R.Rocca
09-12-2014, 07:43 PM
would it be fair to say, that as of eurogenes ENF samples, .......a ENF in the thirty to forty numbers range would be danubian/central european?

No. In his own words...


Early Neolithic Farmer (ENF): I'd say that this is the component of the earliest Neolithic farmers from the Fertile Crescent.

vettor
09-12-2014, 08:28 PM
No. In his own words...
thanks

but I was more interested in the trend/pattern of ENF as it seems to be 40 to 50 is more southern europe, while 10 to 20 more northern europe

Gray Fox
09-12-2014, 09:42 PM
Here's my update

ANE 15.21%
ASE 2.45%
WHG-UHG 61.67%
East_Eurasian -
West_African 1.19%
East_African 0.13%
ENF 19.35%

Wulf Warrior
09-12-2014, 09:49 PM
Mine

ANE 15.95%
ASE 2.05%
WHG-UHG 63.60%
East_Eurasian 0.59%
West_African -
East_African 1.09%
ENF 16.72%

leonardo
09-12-2014, 11:35 PM
My updated results:
ANE 15.21%
ASE 1.34%
WHG-UHG 58.35%
East_Eurasian 0.99%
West_African 0.62%
East_African -
ENF 23.49%

My mother's:
ANE 15.02%
ASE 1.12%
WHG-UHG 61.92%
East_Eurasian 0.81%
West_African -
East_African 1.11%
ENF 20.02%

kenji.aryan
09-13-2014, 12:56 AM
EUROGENES_ANE K7

Population
ANE 29.08%
ASE 17.96%
WHG-UHG 2.52%
East_Eurasian 6.67%
West_African 0.63%
East_African 2.33%
ENF 40.81%

soulblighter
09-13-2014, 01:01 AM
EUROGENES_ANE K7

Population
ANE29.08%
ASE17.96%
WHG-UHG2.52%
East_Eurasian6.67%
West_African0.63%
East_African2.33%
ENF40.81%

Are you from South India or Maharashtra?

Dr_McNinja
09-13-2014, 01:15 AM
Apparently he's a Brahmin from Himachal Pradesh.

soulblighter
09-13-2014, 01:19 AM
Apparently he's a Brahmin from Himachal Pradesh.

Interesting....his breakdown looks nothing like yours or Parasar's! He seems much more like someone from the south. I wish BMG would post his results as well.

bol_nat
09-13-2014, 01:21 AM
EUROGENES_ANE K7

Population
ANE29.08%
ASE17.96%
WHG-UHG2.52%
East_Eurasian6.67%
West_African0.63%
East_African2.33%
ENF40.81%

Can you post MDLP K23b results?

Dr_McNinja
09-13-2014, 01:22 AM
Interesting....his breakdown looks nothing like yours or Parasar's! He seems much more like someone from the south. I wish BMG would post his results as well.He has the ENF and ASE for a North Indian, but his WHG is a little low. But the high East Eurasian makes sense for Himachal Pradesh. I remember seeing a Brahmin from Uttarakhand or Himachal Pradesh from one of the datasets Everest and I were using (from that Estonian lab) and they had a lot of East Eurasian but were otherwise similar to other Brahmins.

bol_nat
09-13-2014, 01:24 AM
Interesting....his breakdown looks nothing like yours or Parasar's! He seems much more like someone from the south. I wish BMG would post his results as well.

I don't think so, MDLP K23b is better calc to determine that.

MitchellSince1893
09-13-2014, 03:42 AM
My Old/New Eurogenes ANE-K7 results
WHG-UHG 47.90% / 63.87%
ENF 24.27% / 16.06%
ANE 21.45% / 16.89%
ASE 3.86% / 2.44%
East_Eurasian 2.51% / -
West_African - / -
East_African - / .74%


Mother's Old/New
WHG-UHG 47.68% / 63.66%
ENF 24.79% / 15.74%
ANE 21.04% / 16.46%
ASE 4.39% / 2.54%
East_Eurasian 1.50% / -
East_African 0.52% / 1.46%
West_African 0.08% / .14%

Father's Old/New
WHG-UHG 47.31% / 62.61%
ENF 25.40% / 17.86%
ANE 20.85% / 16.23%
ASE 3.98% / 2.32%
East_Eurasian 2.46% / 0.11%
West_African - / -
East_African - / 0.87%

Father's maternal half brother
WHG-UHG 45.81% / 62.40%
ENF 25.86% / 19.68%
ANE 20.72% / 14.90%
ASE 4.42% / 2.61%
East_Eurasian 3.19% / -
West_African - / 0.41%
East_African - / -

kenji.aryan
09-13-2014, 11:24 AM
Can you post MDLP K23b results?

# Population Percent
1 South_Indian 38.02
2 South_Central_Asian 31.7
3 Caucasian 9.47
4 European_Hunters_Gatherers 5.78
5 Ancestral_Altaic 4.5
6 South_East_Asian 3
7 Near_East 2.06
8 Arctic 1.68
9 East_Siberian 1.02
10 Amerindian 0.94
11 Melano_Polynesian 0.69
12 North_African 0.54
13 Tungus-Altaic 0.21
14 Archaic_Human 0.18
15 Australoid 0.11
16 Khoisan 0.06
17 Archaic_African 0.03

DON'T QUOTE

and can you guys post yours so that i can compare.

kenji.aryan
09-13-2014, 11:33 AM
Are you from South India or Maharashtra?

No

and if you said it based on my results then my ASE which is similar to ASI is way less then your and even prashar's result.

soulblighter
09-13-2014, 11:44 AM
No

and if you said it based on my results then my ASE which is similar to ASI is way less then your and even prashar's result.

Your ASE is not that different from all of us... it is the east Eurasian that threw me off... I guess the source of east Eurasian is different though.

kenji.aryan
09-13-2014, 11:59 AM
@Soulblighter

How to know your paternal Y and mt dna from your raw data.
your father mtdna is similar to my mtdna U7.

kenji.aryan
09-13-2014, 12:08 PM
Your ASE is not that different from all of us... it is the east Eurasian that threw me off... I guess the source of east Eurasian is different though.

I got my ANE_K7 result based on 162594 SNPs on V4.

NK19191
09-13-2014, 12:12 PM
Population
ANE 18.95%
ASE 6.12%
WHG-UHG 9.32%
East_Eurasian 0.76%
West_African 0.50%
East_African 1.07%
ENF 63.29%

Hok
09-13-2014, 12:13 PM
PCA Updated.

https://www.dropbox.com/s/ywd5cxxghbt9zrn/pcaNames8.pdf?dl=0

Who is SEJJ? I am close to his grandfather.

soulblighter
09-13-2014, 12:31 PM
@Soulblighter

How to know your paternal Y and mt dna from your raw data.
your father mtdna is similar to my mtdna U7.

23andme gives your paternal Y-DNA and maternal mtDNA. You cannot find the others using your kit.
For your maternal Y-DNA and paternal mtDNA, you need to have your maternal uncle/grandpa and paternal uncle/dad/aunt tested respectively.

bol_nat
09-13-2014, 12:52 PM
ˇ
ˇ

DON'T QUOTE

and can you guys post yours so that i can compare.

I posted in MDLP thread, anyway here it is.

Pakistan punjabi jatt.

# Population Percent
1 South_Central_Asian 35.98
2 South_Indian 33.22
3 Caucasian 13.87
4 Ancestral_Altaic 5.03
5 European_Hunters_Gatherers 4.89
6 Australoid 1.39
7 Paleo_Siberian 1.31
8 Near_East 1.2
9 Arctic 1.12
10 East_Siberian 0.85
11 Melano_Polynesian 0.74
12 South_East_Asian 0.26
13 Archaic_Human 0.11
14 European_Early_Farmers 0.04

Gray Fox
09-13-2014, 01:29 PM
What group/groups represent the West African? Mine shot up a bit this last revision go-round. I'm assuming this is noise as the percentage for this group has never exceeded one percent.

parasar
09-13-2014, 02:30 PM
He has the ENF and ASE for a North Indian, but his WHG is a little low. But the high East Eurasian makes sense for Himachal Pradesh. I remember seeing a Brahmin from Uttarakhand or Himachal Pradesh from one of the datasets Everest and I were using (from that Estonian lab) and they had a lot of East Eurasian but were otherwise similar to other Brahmins.

In Uttarkhand there is a lot of East Asiatic type influence, more that regions much farther to the east, say eastern Bengal/Bangladesh.

everest59
09-13-2014, 02:58 PM
In Uttarkhand there is a lot of East Asiatic type influence, more that regions much farther to the east, say eastern Bengal/Bangladesh.

I wonder if the Uttarkhand Brahmin is an outlier or just typical. Now if that guy is typical then Nepalese Brahmins like myself are real anomalies. Some Nepalese Brahmins are similar to that Uttaranchal sample.
Where did hill Brahmins like myself come from? The more I think about it the more I get confused. My total ENA is similar to yours, but how should I divide up that East Asian portion, which I would say is an excess of 5%. A couple percent seems natural every where.
Nepalese Brahmins are divided into 3 subcastes. You have the Upadhyay Brahmin, who are on top. We have the Kumain Brahmins who hold from Kumaon in Uttaranchal. Then we have the Jaishis. I think it's possible the three groups have different levels of East Asian.
Now, an uncle of mine wrote a book that mentions that we came from Kannauj. He even has names of our ancestors. Now the question is, how much should I trust that book? It's possible only if Brahmins from Kannauj are similar to that one Tyagi sample of Harappa.
I am thinking we are mixes of various Brahmin types from all around.

everest59
09-13-2014, 03:13 PM
He has the ENF and ASE for a North Indian, but his WHG is a little low. But the high East Eurasian makes sense for Himachal Pradesh. I remember seeing a Brahmin from Uttarakhand or Himachal Pradesh from one of the datasets Everest and I were using (from that Estonian lab) and they had a lot of East Eurasian but were otherwise similar to other Brahmins.

Yeah, I remember that sample. He was way different. A whopping 30 percent east Asian score, which is not similar to me or Kenji. The question is if he is typical or an outlier. We really need samples from that
area.

bored
09-13-2014, 03:41 PM
Yeah, I remember that sample. He was way different. A whopping 30 percent east Asian score, which is not similar to me or Kenji. The question is if he is typical or an outlier. We really need samples from that
area.

I suspect that there is an East to West gradient here wrt East Asian admix. I would expect Nepali Brahmins to have more East Asian than Uttarakhand Brahmins. Though it's possible that Nepalis have less if you guys are relatively recent migrants from the plains and were fairly endogamous. Looks like Uttarakhand Brahmins were not.
Himachal Brahmins like Kenji would have lower East Asian than Uttarakhand and Nepali Brahmins. Finally Jammu Brahmins and Kashmiri Pandits will have the least amount of East Asian admixture. Intuitively this makes sense.

everest59
09-13-2014, 03:54 PM
I suspect that there is an East to West gradient here wrt East Asian admix. I would expect Nepali Brahmins to have more East Asian than Uttarakhand Brahmins. Though it's possible that Nepalis have less if you guys are relatively recent migrants from the plains and were fairly endogamy. Looks like Uttarakhand Brahmins were not.
Himachal Brahmins like Kenji would have lower East Asian than Uttarakhand and Nepali Brahmins. Finally Jammu Brahmins and Kashmiri Pandits will have the least amount of East Asian admixture. Intuitively this makes sense.

This is my expectation as well. You should have similar levels to Pandits in my opinion.
Kenji and I seem to have similar levels. The difference may be 1% higher in me. On 23andme PCA we are right next to each other, although that PCA is in all honesty absolute crap.

Anyways, my K7

ANE 28.44%
ASE 17.79%
WHG-UHG 8.18%
East_Eurasian 7.85%
West_African 0.36%
East_African 2.30%
ENF 35.07%

Dr_McNinja
09-13-2014, 05:34 PM
This is my expectation as well. You should have similar levels to Pandits in my opinion.
Kenji and I seem to have similar levels. The difference may be 1% higher in me. On 23andme PCA we are right next to each other, although that PCA is in all honesty absolute crap.

Anyways, my K7

ANE 28.44%
ASE 17.79%
WHG-UHG 8.18%
East_Eurasian 7.85%
West_African 0.36%
East_African 2.30%
ENF 35.07%Yours seems close to Haryanvi Jatts mixed with some East Asian:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

Your ANE seems low by half a percent like me and my dad, so I wonder if some of your East Eurasian isn't actually from it or displacing it.

everest59
09-13-2014, 05:43 PM
Yours seems close to Haryanvi Jatts mixed with some East Asian:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

Your ANE seems low by half a percent like me and my dad, so I wonder if some of your East Eurasian isn't actually from it or displacing it.

It's possible. What does MA-1 score in this calculator? We can get an idea from that.
Edit. I found the DIY calculator. This is what I got:

52.93% ANE
15.80% ASE
19.24% WHG-UHG
7.37% East_Eurasian
0.01% West_African
4.63% East_African
0.01% ENF

From that, yes, it's possible that some East Eurasian is ANE.

Dr_McNinja
09-13-2014, 06:04 PM
I added Kenji to my Harappa sheet here:

https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=6

In the sort, he comes out between Haryanvi Jatts and Northern Brahmins (Everest and Parasar), but closer to the Haryanvis (expected due to Punjabi levels of Caucasian).

That method still seems to work best with Harappa.

I put him in the cluster continuum starting with my mother and ending with him, including two Haryanvi Jatts and me and my father.

Dr_McNinja
09-13-2014, 06:11 PM
It's possible. What does MA-1 score in this calculator? We can get an idea from that.
Edit. I found the DIY calculator. This is what I got:

52.93% ANE
15.80% ASE
19.24% WHG-UHG
7.37% East_Eurasian
0.01% West_African
4.63% East_African
0.01% ENF

From that, yes, it's possible that some East Eurasian is ANE.If you mixed that with a population that is 7% ANE, 14% ASE, 80% ENF, you'd get a modern South Asian.

So that would correspond to a pre-ANE Ancestral North Indian group (maybe not even located in North India), which is what I meant about modern South Indian being a poor substitute for such a hypothetical group. Modern South Indians are 28-30% ANE, probably as much as if not more ANE than Baloch, probably not more than 25-30% ASE, and maybe as much East Eurasian as ENF.

I wonder if it's possible to phase a "parent" population from Malta and modern South Asians to get this kind of hypothetical ANI (use modern South Asians as a child, and Malta as one parent, phase for missing parent, kind of like GEDmatch's tool).

ViktorL1
09-14-2014, 02:24 AM
yep here's mine:

ANE 29.52%
ASE 10.09%
WHG-UHG 14.76%
East_Eurasian 4.75%
West_African 1.59%
East_African -
ENF 39.26%

Rukha, it's interesting how discrepant our results are despite the fact that we're both Tajik. I guess Afghan Tajiks and Tajikistani Tajiks are clearly seperate? I'm Sughdi Tajik, which is even more different, I think :p

Sapporo
09-14-2014, 02:41 AM
Rukha, it's interesting how discrepant our results are despite the fact that we're both Tajik. I guess Afghan Tajiks and Tajikistani Tajiks are clearly seperate? I'm Sughdi Tajik, which is even more different, I think :p

Viktor, it's not necessarily just differences between Afghan and Tajikistani Tajiks. Tajiks seem much more heterogeneous than Pashtuns and likely vary on a provincial level. Even then, there are likely difference between individuals of the same village. Some are heavily East Eurasian influenced while others are slightly more South Asian influenced and closer to Pashtuns. The rest are sort of in between those who are heavily East Asian influenced and those closer to Pashtuns. Then, you have isolated groups like the Pamiris from Rushan, Ishkashim and Shugnan who retain little contact with other Tajiks and Pashtuns or the Yagnobi who I presume are from Sughd yet likely quite different on average from mainstream Tajiks in Sughd such as yourself. On the other hand, Pashtuns seem to form a gradient between slightly more Turkic/Caucasus influence and those with more South Asian influence. Yet, their West Eurasian ancestry is rather consistent overall other than an outlier here and there such as the clearly Turkic admixed HGDP 00220 Pashtun or heavily South Asian influenced HGDP00239. I believe that Di Cristofano noted they are rather homogenous overall.

http://www.harappadna.org/2013/07/pathan-pashtun-admixture-results/

Although, some Afghan Pashtuns vary as well but I believe those from Kunduz and Baghlan are the exception rather than the norm as they are somewhat isolated from other Pashtuns further south.

http://www.harappadna.org/2014/01/afghan-dataset/

I think this isn't surprising since the Tajik identity is much more fluid than that of Pashtuns who fit the definition of a modern day ethnic group much more firmly.

Rukha
09-14-2014, 02:59 AM
Rukha, it's interesting how discrepant our results are despite the fact that we're both Tajik. I guess Afghan Tajiks and Tajikistani Tajiks are clearly seperate? I'm Sughdi Tajik, which is even more different, I think :p

I think it varies, Tajiks from northern provinces like Balkh, Faryab, etc. likely display affinity to Tajiks from southern Tajikistan. Tajiks from Herat are probably closer to Iranians. Tajiks from eastern provinces like Parwan, Panjsher (where my father is from), and Takhar seem to be closer to isolated Tajik communities like Pamiris and Yaghnobis along with Pashtuns. Tajik is a heteronym and was even used to designate Persians of Iran in the past, so the discrepancy in our results isn't that surprising :p

There's a Kabuli Tajik on Harappa who is very much East Eurasian influenced as well, so it's not a matter of Afghan Tajik vs. Tajikistani Tajik like Sapporo said.

Here are my corrected results btw, pretty much what Dr_McNinja originally posted:

ANE 29.32%
ASE 10.54%
WHG-UHG 9.84%
East_Eurasian 3.02%
West_African 0.17%
East_African 1.71%
ENF 45.41%

Sein
09-14-2014, 03:05 AM
Historically speaking, Tajik is really a broad term that was used to describe sedentary (by sedentary, I mean village+city dwelling) Farsi/Dari speakers across Central Asia and the Iranian plateau. The term doesn't carry any connotation of shared ancestry/common descent.

By contrast, the Pashtun/Pukhtun title is supposed to emanate from shared ancestry (common descent from a single ancestor, who supposedly lies right at the beginning of all the various tribal genealogies. Very implausible, certainly mythical). Also, it involves a language which is much more homogeneous in comparison to the varieties of Farsi. On top of that, Pashtuns are supposed to share a common tribal law or code of conduct, Pashtunwali/Pukhtunwali (although it seems the Durrani aren't very familiar with things like jirga, wesh, etc).

parasar
09-14-2014, 04:38 AM
Historically speaking, Tajik is really a broad term that was used to describe sedentary (by sedentary, I mean village+city dwelling) Farsi/Dari speakers across Central Asia and the Iranian plateau. The term doesn't carry any connotation of shared ancestry/common descent.
...


Do we know when the transition happened when the term Tajik became associated with "sedentary (by sedentary, I mean village+city dwelling) Farsi/Dari speakers?"
Traditionally in Indic and Chinese material Tajik means Arab, perhaps following Persian usage of the term.

http://books.google.com/books?id=8rLUbuZLiaIC&pg=PA89
http://books.google.com/books?id=lEB11tKmCgcC&pg=PA166
http://books.google.com/books?id=Pl9QnrfoYS8C&pg=PT48

Gray Fox
09-14-2014, 04:38 AM
What group/groups represent the West African? Mine shot up a bit this last revision go-round. I'm assuming this is noise as the percentage for this group has never exceeded one percent.

Am I talking to myself here??

Generalissimo
09-14-2014, 04:48 AM
Am I talking to myself here??

Pygmies.

It's not important though, because less than 1% just indicates noise from something really basal that has nothing to do with Pygmies.

tamilgangster
09-14-2014, 05:02 AM
This ANE component changes all of our fundamental knowledge of genetic anthropology. Since ANE is such an ancient component, due to isolation the ANE found in Amerindian, Europeans and South Asians are all very different from eachother. The Karitiana who are one of the more mongoloid looking amerindians, unlike North american Indians who appear to have caucasoid influences. Also there have been previous studies discussing about a population in south asia who were protocaucasoid/mongoloid before the west east eurasian split. I thinks its very likely that the ANE came out of India.

tamilgangster
09-14-2014, 05:08 AM
The score for ASE is almost Identicle to the Onge score on harappa DNA. The East Eurasian score corresponds to the AUstroasiatic peoples. But there needs to be more sample from Tribals, and middle-low castes from south India, and central Indians also. Im pretty sure the onge would score almost 100% ASE

Generalissimo
09-14-2014, 05:57 AM
This ANE component changes all of our fundamental knowledge of genetic anthropology. Since ANE is such an ancient component, due to isolation the ANE found in Amerindian, Europeans and South Asians are all very different from eachother. The Karitiana who are one of the more mongoloid looking amerindians, unlike North american Indians who appear to have caucasoid influences. Also there have been previous studies discussing about a population in south asia who were protocaucasoid/mongoloid before the west east eurasian split. I thinks its very likely that the ANE came out of India.

Based on a couple of the tweets from the Ust-Ishim talk at SMBE 2014, it seems to me that this genome was an older version of Mal'ta boy. In other words, a more basal Ancient North Eurasian.

If that's true, then ANE was in West Siberia 45,000 years ago. It'll be very tricky to find out where it was before that. The fact that Ust-Ishim lacks any detectable Denisovan ancestry might be a clue though.

Shaikorth
09-14-2014, 06:42 AM
Based on a couple of the tweets from the Ust-Ishim talk at SMBE 2014, it seems to me that this genome was an older version of Mal'ta boy. In other words, a more basal Ancient North Eurasian.

If that's true, then ANE was in West Siberia 45,000 years ago. It'll be very tricky to find out where it was before that. The fact that Ust-Ishim lacks any detectable Denisovan ancestry might be a clue though.

Ust-Ishim based on what I heard about that conference was equally close to West and East Eurasians, maybe a bit closer to east. That wouldn't be really ANE because MA-1 is much more distant from She than from West Eurasians. More Tianyuan-like.

Generalissimo
09-14-2014, 07:00 AM
Ust-Ishim based on what I heard about that conference was equally close to West and East Eurasians, maybe a bit closer to east. That wouldn't be really ANE because MA-1 is much more distant from She than from West Eurasians. More Tianyuan-like.

According to a tweet from Reich's talk, the Ust-Ishim genome agrees with the concept of the Ancient North Eurasians. I don't know what that means exactly, but I assume that it means Ust-Ishim is part of the ANE clade.

ViktorL1
09-14-2014, 07:18 AM
Historically speaking, Tajik is really a broad term that was used to describe sedentary (by sedentary, I mean village+city dwelling) Farsi/Dari speakers across Central Asia and the Iranian plateau. The term doesn't carry any connotation of shared ancestry/common descent.

By contrast, the Pashtun/Pukhtun title is supposed to emanate from shared ancestry (common descent from a single ancestor, who supposedly lies right at the beginning of all the various tribal genealogies. Very implausible, certainly mythical). Also, it involves a language which is much more homogeneous in comparison to the varieties of Farsi. On top of that, Pashtuns are supposed to share a common tribal law or code of conduct, Pashtunwali/Pukhtunwali (although it seems the Durrani aren't very familiar with things like jirga, wesh, etc).

Yes, I'd agree. Even in Tajikistan, ethnicity has been very regional (Sughdi, Pamiri, Kulobi, etc.). I think those identifiers would correspond to genetic clusters. It's clear that I am more genetically similar to the average Uzbek than southern Tajik, for instance. Looking at the history of the region, Sughd (the region where my ancestors are from), has seen countless waves of invaders from Eastern Asia (Turks, Mongols, Huns), and many ethnic Chinese even settled in Soghdiana prior to the arrival of Islam. Even today, if the borders were opened, Kyrgyzstan (lingusistically and racially Turkic) is less than 2 hours from Khujand, while Dushanbe is over 10 hours away, across 3000m mountains. Phenotypically (despite the subjectivity inherent), Sughdi Tajiks tend to look very different from the Southerners.

NB: I should add that Uzbek identity has been fairly fluid as well, although Uzbeks on average have a lot more East Eurasian admixture than Tajiks.

Shaikorth
09-14-2014, 07:28 AM
According to a tweet from Reich's talk, the Ust-Ishim genome agrees with the concept of the Ancient North Eurasians. I don't know what that means exactly, but I assume that it means Ust-Ishim is part of the ANE clade.

This is really the most informative tweet on that matter: "Fu on genome of 45kya old modern human, ust' ishim, clearly ooa, fall in weakly in Asian clade ,but close to Anc of eurasia"

Combined with the abstract (https://mcidublin.conference-services.net/reports/template/onetextabstract.xml?xsl=template/onetextabstract.xsl&conferenceID=3958&abstractID=811773) I take it to mean he was Tianyuan-like, but with a much better sequence. MA-1 is specific enough that they should have said if Ust-Ishim was closest to it, or distant from East Asians and closer to native americans etc.

evon
09-14-2014, 09:08 AM
Yes, I'd agree. Even in Tajikistan, ethnicity has been very regional (Sughdi, Pamiri, Kulobi, etc.). I think those identifiers would correspond to genetic clusters. It's clear that I am more genetically similar to the average Uzbek than southern Tajik, for instance. Looking at the history of the region, Sughd (the region where my ancestors are from), has seen countless waves of invaders from Eastern Asia (Turks, Mongols, Huns), and many ethnic Chinese even settled in Soghdiana prior to the arrival of Islam. Even today, if the borders were opened, Kyrgyzstan (lingusistically and racially Turkic) is less than 2 hours from Khujand, while Dushanbe is over 10 hours away, across 3000m mountains. Phenotypically (despite the subjectivity inherent), Sughdi Tajiks tend to look very different from the Southerners.

NB: I should add that Uzbek identity has been fairly fluid as well, although Uzbeks on average have a lot more East Eurasian admixture than Tajiks.

Do you have any Yaghnobi ancestry? It would be interesting to compare you and the DNA from other Tadjiks to ancient Sogdians, too bad it seems the best proxy we have are Yaghnobis..

Dr_McNinja
09-14-2014, 11:07 AM
The score for ASE is almost Identicle to the Onge score on harappa DNA. The East Eurasian score corresponds to the AUstroasiatic peoples. But there needs to be more sample from Tribals, and middle-low castes from south India, and central Indians also. Im pretty sure the onge would score almost 100% ASEIt's off by a few percent for many, the ASE is probably more more accurate than the old Onge component in Harappa. East Eurasian would correspond to all East Asian, not just Southeast Asian.

Dr_McNinja
09-14-2014, 11:33 AM
I don't suppose there's any way to see the population averages for Baloch/Makrani in this calculator? Or everest, if you have some Gedrosian "zombies" from an unsupervised run in another calculator, could you see what their breakdown would be?

NK19191
09-14-2014, 12:29 PM
Do we know when the transition happened when the term Tajik became associated with "sedentary (by sedentary, I mean village+city dwelling) Farsi/Dari speakers?"
Traditionally in Indic and Chinese material Tajik means Arab, perhaps following Persian usage of the term.

http://books.google.com/books?id=8rLUbuZLiaIC&pg=PA89
http://books.google.com/books?id=lEB11tKmCgcC&pg=PA166
http://books.google.com/books?id=Pl9QnrfoYS8C&pg=PT48

The word Tazi or Tajik came to have two distinct meanings one refers to the Arabs, and one refers to Ethnic Tajiks of today.


it should be understood that the Persian words (a) tāzi ‘Arabian, Arabic, Arab’ and (b) tāzik, tāžik, tājik ‘Persian, Iranian, Tajik,’ though originating as doublets (or cognates) of the same word, are completely separate in form and meaning throughout New Persian (and Islamic Turkic and Indic) literature.



The Ṭayyiʾ bedouin of northern central Najd, famous in Arab and later Islamic lore for the generosity of the fabled Ḥātem, were (on the evidence of Greek and Persian loanwords: Rabin, p. 201) in contact with Byzantium and Sasanian Iran from an early date. They came into prominence on the eve of Islam, when Ḵosrow II Parviz (r. 590-628, with interruptions) abolished the Arab Lakhmid dynasty and appointed a Ṭayyiʾ chieftain as his governor of Ḥira, in effect march-warden of the Euphrates. It is not surprising, therefore, that their name was generalized in Syriac and Aramaic, from perhaps the fourth century, in the form Ṭayyāye (Segal, pp. 100-3) and Middle Persian, Armenian, Georgian (apparently directly from Middle Persian or Parthian: see Andronikashvili, p. 568), and Sanskrit —to denote Arabs as a whole


The most plausible and generally accepted origin of the word is Middle Persian tāzīk ‘Arab’ (cf. New Persian tāzi), or an Iranian (Sogdian or Parthian) cognate word. The Muslim armies that invaded Transoxiana early in the eighth century, conquering the Sogdian principalities and clashing with the Qarluq Turks (see Bregel, Atlas, Maps 8–10) consisted not only of Arabs but also of Persian converts from Fārs and the central Zagros region (Bartol’d , “Tadžiki,” pp. 455-57). [B]Hence the Turks of Central Asia adopted a variant of the Iranian word, täžik, to designate their Muslim adversaries in general. By the eleventh century (Yusof Ḵāṣṣ-ḥājeb, Qutadḡu bilig, lines 280, 282, 3265) the Qarakhanid Turks applied this term more specifically to the Persian Muslims in the Oxus basin and Khorasan, who were variously the Turks’ rivals, models, overlords (under the Samanid Dynasty), and subjects (from Ghaznavid times on).


Persian writers of the Ghaznavid, Seljuq and Atābak periods (ca. 1000–1260) adopted the term and extended its use to cover Persians in the rest of Iran, now under Turkish rule, as early as the poet ʿOnṣori, ca. 1025 (Dabirsiāqi, pp. 3377, 3408). Iranians soon accepted it as an ethnonym, as is shown by a Persian court official’s referring to mā tāzikān “we Tajiks” (Bayhaqi, ed. Fayyāz, p. 594).

everest59
09-14-2014, 01:59 PM
I don't suppose there's any way to see the population averages for Baloch/Makrani in this calculator? Or everest, if you have some Gedrosian "zombies" from an unsupervised run in another calculator, could you see what their breakdown would be?

Okay, I will do this when I get a chance. I have "Baloch" zombies of Harappa with me. Also k12b Gedrosian I think. I need to check it out.

parasar
09-14-2014, 02:49 PM
According to a tweet from Reich's talk, the Ust-Ishim genome agrees with the concept of the Ancient North Eurasians. I don't know what that means exactly, but I assume that it means Ust-Ishim is part of the ANE clade.

Possible, but I think this is doubtful (at least on the Y side). He had pretty large Neanderthal segments which could point to some kind of an east Asiatic (cf Xujiayao) moving west.
http://www.icr.org/article/neandertals-mixed-with-humans-china/


"modern humans had spread across southern Eurasia by 100,000 years ago, and yet archaic humans [who supposedly lived before modern humans] remained across the more northern areas, and even displaced the modern humans in Southwest Asia for an additional 50,000-70,000 years ... modern humans expanded then, after remaining in Africa and southern Asia for 50,000 plus years.

Generalissimo
09-14-2014, 02:56 PM
He had pretty large Neanderthal segments which could point to some kind of an east Asiatic (cf Xujiayao) moving west.

Neanderthal admixture with humans probably took place in the Near East. So it's most likely that Ust-Ishim was the descendent of a direct migration from the Near East to West Siberia.

The reason these segments were relatively larger than they are now among West Eurasians, is because Ust-Ishim lived closer to the date of the admixture.

parasar
09-14-2014, 03:02 PM
The word Tazi or Tajik came to have two distinct meanings one refers to the Arabs, and one refers to Ethnic Tajiks of today...


Persian writers of the Ghaznavid, Seljuq and Atābak periods (ca. 1000–1260) adopted the term and extended its use to cover Persians in the rest of Iran, now under Turkish rule, as early as the poet ʿOnṣori, ca. 1025 (Dabirsiāqi, pp. 3377, 3408). Iranians soon accepted it as an ethnonym, as is shown by a Persian court official’s referring to mā tāzikān “we Tajiks”


So it appears that the transition happened in Ghaznavid period.

Even as late as the early British period there was some remembrance of the original connotation. Bellew notes and opines:

The Tajik, or, as he is frequently called, the Parsiwan, constitute a numerous and widely spread portion of the inhabitants of Afghanistan, from whom they differ in language, internal government, and manners and customs. They are the representatives of the ancient Persian inhabitants of the country, as the Afghans are of its ancient Indian inhabitants ... an Arab settling in the country, and not intermarrying with its people, retained his proper national title through successive generations. But the Arab intermarrying with the people of the country lost his proper nationality, and, in the succeeding generations, was called Tajik by the Persians ... The name is applied nowadays in a very loose way, and is made to include all the Persian-speaking people of the country who are not either Hazarah, Afghan, or Sayyid. Thus the Indian races on the southern slopes of Hindu Kush, who have been converted to Muhammadanism and speak Persian (as well as to some extent their native dialects), are commonly called Tajik.

parasar
09-14-2014, 03:09 PM
Neanderthal admixture with humans probably took place in the Near East. So it's most likely that Ust-Ishim was the descendent of a direct migration from the Near East to West Siberia.

The reason these segments were relatively larger than they are now among West Eurasians, is because Ust-Ishim lived closer to the date of the admixture.

I agree that that has been the thinking, but a more plausible scenario is that it occurred only in East Asia rather than twice as posited by Wall et al.:
http://www.genetics.org/content/early/2013/02/04/genetics.112.148213.short?rss=1

This observation was consistent with a single episode of admixture from Neanderthals into the ancestors of all non-Africans when the two groups coexisted in the Middle East 50-80 Kya. We examined the relationship between Neanderthals and modern humans in greater detail by applying two complementary methods to the published draft Neanderthal genome and an expanded set of high-coverage modern human genome sequences. We find that, consistent with the recent finding of Meyer et al. (2012), Neanderthals contributed more DNA to modern East Asians than to modern Europeans. Furthermore we find that the Maasai of East Africa have a small but significant fraction of Neanderthal DNA. Because our analysis is of several genomic samples from each modern human population considered, we are able to document the extent of variation in Neanderthal ancestry within and among populations. Our results combined with those previously published show that a more complex model of admixture between Neanderthals and modern humans is necessary to account for the different levels of Neanderthal ancestry among human populations. In particular, at least some Neanderthal-modern human admixture must postdate the separation of the ancestors of modern European and modern East Asian populations.



A single admixture event in East Asia would explain why the Neanderthal element is strongest in East Asia and the weakest in EEF-rich folk like the Sardinians.

Generalissimo
09-14-2014, 03:45 PM
A single admixture event in East Asis would explain why the Neanderthal element is strongest in East Asia and the weakest in EEF-rich folk like the Sardinians.

Neanderthals didn't live in East Asia. The Altai is about as east that they ever lived.

Shaikorth
09-14-2014, 04:09 PM
Possible, but I think this is doubtful (at least on the Y side). He had pretty large Neanderthal segments which could point to some kind of an east Asiatic (cf Xujiayao) moving west.
http://www.icr.org/article/neandertals-mixed-with-humans-china/

He had longer chunks of Neanderthal ancestry, but quantitative differences to modern Eurasians are unknown, supposedly very similar.

Anyway, this pretty much is how I expect he'll differ from MA-1 in relation to modern populations.

http://oi59.tinypic.com/9qfnfc.jpg

parasar
09-14-2014, 04:37 PM
Neanderthals didn't live in East Asia. The Altai is about as east that they ever lived.

An admixed type with part Neanderthal features was indeed present further east.

http://www.pnas.org/content/111/29/10509
Wu et. al.

We present the temporal labyrinths of four eastern Eurasian Pleistocene Homo, one each of Early (Lantian 1), Middle (Hexian 1), and Late (Xujiayao 15) Pleistocene archaic humans and one early modern human (Liujiang 1). The labyrinths of the two earlier specimens and the most recent one conform to the proportions seen among western early and recent modern humans, reinforcing the modern human pattern as generally ancestral for the genus Homo. The labyrinth of Xujiayao 15 is in the middle of the Neandertal variation and separate from the other samples. This eastern Eurasian labyrinthine dichotomy occurs in the context of none of the distinctive Neandertal external temporal or other cranial features. As such, it raises questions regarding possible cranial and postcranial morphological correlates of Homo labyrinthine variation, the use of individual “Neandertal” features for documenting population affinities, and the nature of late archaic human variation across Eurasia ...


http://www.nature.com/nature/journal/v507/n7492/full/nature12961.html
http://2.bp.blogspot.com/-iPjWPx5aHF0/UulquRDwSEI/AAAAAAAAJew/XCKk9JDFwsQ/s1600/neandertal.png

Sein
09-14-2014, 05:47 PM
Okay, I will do this when I get a chance. I have "Baloch" zombies of Harappa with me. Also k12b Gedrosian I think. I need to check it out.

Everest,

If possible, could you run all HarappaWorld components? It would be very interesting to see how they breakdown. Thanks in advance.

The ANE cluster here is almost perfect, but I think the ASE component isn't really equivalent to ASI. It's too widespread across Eurasia to be ASI (many Europeans range between 2%-5% ASE, and the cluster is very important in East Asian groups as well). Also, it correlates with ANE, almost following it. For example, Sardinians don't have ANE admixture, and the Sardinian example posted by David has 0% ASE. The Bedouin sample David posted has 0% ANE, and they are also 0% ASE. Yet, the Lezgin example is 26.74% ANE, and 3.88% ASE. The Native American example is 41.56% ANE, and 0.41% ASE. In Europeans, it's ANE-rich Europeans who show the highest ASE. When Everest ran MA1 through the calculator, MA1 basically scored 0% ENF, but did score 16% ASE. So, there is something exceedingly ANE about this ASE cluster. The Malay are Indian-admixed, and show above noise levels of ANE admixture in David's run. ASE peaks in them. I think that adds a subtle ANE bias to the ASE cluster.

Then again, ASI is a poorly substantiated idea, based only on modern DNA, and that too from a proxy which is supposedly quite distinct from it (the Onge, who are pretty diverged from the theoretical construct of ASI). Perhaps, with the presence of true ancient clusters (ANE, and probably ENF and WHG as well), we've found something that is more robust than ASI, a submerged genetic substratum across Eurasia? Being pan-Eurasian could also explain why it seems to follow ANE around in Europe and the Caucasus.

parasar
09-14-2014, 07:37 PM
...
Then again, ASI is a poorly substantiated idea, based only on modern DNA, and that too from a proxy which is supposedly quite distinct from it (the Onge, who are pretty diverged from the theoretical construct of ASI). Perhaps, with the presence of true ancient clusters (ANE, and probably ENF and WHG as well), we've found something that is more robust than ASI, a submerged genetic substratum across Eurasia? Being pan-Eurasian could also explain why it seems to follow ANE around in Europe and the Caucasus.

Plus ASI was a counterpart to ANI, which latter is absent in an analysis where now ANE is fixed as Karitiana minus East Asian.

Shaikorth
09-14-2014, 11:10 PM
An admixed type with part Neanderthal features was indeed present further east.

http://www.pnas.org/content/111/29/10509
Wu et. al.


http://www.nature.com/nature/journal/v507/n7492/full/nature12961.html
http://2.bp.blogspot.com/-iPjWPx5aHF0/UulquRDwSEI/AAAAAAAAJew/XCKk9JDFwsQ/s1600/neandertal.png

IBD reveals North Chinese Neanderthal ancestry peak too.

http://33.media.tumblr.com/97f8fa755451ce5b2d761329d61dda0b/tumblr_n3otgc5iL81r46foao6_1280.png
http://31.media.tumblr.com/4faeef983630d6d528d3b4ec0b542084/tumblr_n3otgc5iL81r46foao7_1280.png

http://33.media.tumblr.com/9558d8b28474b4b7a9eebf8ac29541a9/tumblr_n3otgc5iL81r46foao8_1280.png

Dr_McNinja
09-14-2014, 11:54 PM
Everest,

If possible, could you run all HarappaWorld components? It would be very interesting to see how they breakdown. Thanks in advance.

The ANE cluster here is almost perfect, but I think the ASE component isn't really equivalent to ASI. It's too widespread across Eurasia to be ASI (many Europeans range between 2%-5% ASE, and the cluster is very important in East Asian groups as well). Also, it correlates with ANE, almost following it. For example, Sardinians don't have ANE admixture, and the Sardinian example posted by David has 0% ASE. The Bedouin sample David posted has 0% ANE, and they are also 0% ASE. Yet, the Lezgin example is 26.74% ANE, and 3.88% ASE. The Native American example is 41.56% ANE, and 0.41% ASE. In Europeans, it's ANE-rich Europeans who show the highest ASE. When Everest ran MA1 through the calculator, MA1 basically scored 0% ENF, but did score 16% ASE. So, there is something exceedingly ANE about this ASE cluster. The Malay are Indian-admixed, and show above noise levels of ANE admixture in David's run. ASE peaks in them. I think that adds a subtle ANE bias to the ASE cluster.

Then again, ASI is a poorly substantiated idea, based only on modern DNA, and that too from a proxy which is supposedly quite distinct from it (the Onge, who are pretty diverged from the theoretical construct of ASI). Perhaps, with the presence of true ancient clusters (ANE, and probably ENF and WHG as well), we've found something that is more robust than ASI, a submerged genetic substratum across Eurasia? Being pan-Eurasian could also explain why it seems to follow ANE around in Europe and the Caucasus.That's probably why David named it "Ancestral South Eurasian", it isn't ASI but it's what we were probably referring to as ASI earlier. We might have to change our conceptions of ancient Indians until we get more ancient DNA evidence.

The correlation with ANE is probably because MA-1 himself was like 52% ANE and still had 15-16% ASE, so some of that ASE is leaking over from ANE admixture even now. There is some ASE (and East Eurasian) still within ANE that hasn't been fully isolated. That's probably the difference to our ideal 30-40% range.

Generalissimo
09-15-2014, 12:28 AM
MA-1 is actually a fairly low quality sample.

Its ASE score drops to 14.22% at 3x coverage, and I'd hazard a guess it would fall even more at much higher coverage.

Dr_McNinja
09-15-2014, 01:16 AM
MA-1 is actually a fairly low quality sample.

Its ASE score drops to 14.22% at 3x coverage, and I'd hazard a guess it would fall even more at much higher coverage.How close to 100% ANE in that test should it be, ideally?

I have a feeling a lot of people's results would change with the kind of coverage offered by new next generation autosomal full genome tests (I think FGC just debuted such a test at $1800 or something?). I was playing around with 900k SNP datasets (23andMe V3 users only) and noticed a clear trend in lower ASI/ASE type scores for modern South Asians. I have a feeling results from autosomal tests with much greater coverage would not be what we expect entirely. I think our general conclusions on populations might be borne out (except perhaps this ASI/ANI debate), but individuals will shift around a lot.

Generalissimo
09-15-2014, 01:28 AM
How close to 100% ANE in that test should it be, ideally?

With high coverage and enough markers, close to 100%.

But the coverage is really poor and most markers are missing. At just 7x coverage Oetzi gets results similar to those of Sardinians, which is what we'd expect. Stuttgart is sequenced at 19x, so its results will be even better.

Sein
09-15-2014, 01:53 AM
With high coverage and enough markers, close to 100%.

But the coverage is really poor and most markers are missing. At just 7x coverage Oetzi gets results similar to those of Sardinians, which is what we'd expect. Stuttgart is sequenced at 19x, so its results will be even better.

The impressive thing about your ADMIXTURE test is that your ANE cluster isn't biased towards Siberians, and doesn't show inflated levels among Native Americans, despite constituting the non-ENA portion of Karitiana genomes. And the European results are superior to what we see in Lazaridis et al. Again, pretty impressive.

I guess gauging the correct levels of ANE admixture across populations was always a matter of markers and coverage.

ZephyrousMandaru
09-15-2014, 02:02 AM
MDS Plot of Anthrogenica members for Eurogenes ANE K=7.

http://s1.postimg.org/ta59k1533/Eurogenes_ANE_K_7_MDS.png

ZephyrousMandaru
09-15-2014, 03:17 AM
Updated MDS

http://s4.postimg.org/n4szu4re5/Eurogenes_ANE_K_7_MDS_2.png

Dendrogram

http://s4.postimg.org/th853yugd/Eurogenes_ANE_K_7_Dendrogram.png

Sapporo
09-15-2014, 05:19 AM
ZephyrousMandaru, awesome MDS and dendogram. Could you add Sein and Dr_McNinja to them?

I would thank you but on my phone and no internet access on laptop for a few days.

ZephyrousMandaru
09-15-2014, 05:44 AM
ZephyrousMandaru, awesome MDS and dendogram. Could you add Sein and Dr_McNinja to them?

I would thank you but on my phone and no internet access on laptop for a few days.

Yes, do you know their Eurogenes ANE K=7 values?

Sapporo
09-15-2014, 05:56 AM
Yes, do you know their Eurogenes ANE K=7 values?

They are on the spreadsheet in McNinja's signature. I would add some others too but not too many because some will be on top of one another.

https://docs.google.com/spreadsheet/lv?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoYl E&f=true&noheader=false&gid=18

Dr_McNinja
09-15-2014, 10:50 AM
Here's the GEDmatch K7: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=22

DIY/R K7: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=18

Dr_McNinja
09-15-2014, 03:48 PM
Looking at the results of this and the new MDLP calculator I estimate for North/Northwest Indians (and some upper castes from other parts of India) there seems to be 29-30% ANE, 28-31% ENF-like Caucasian (from Harappa Caucasian and Harappa S-Indian), 14-17% ASE, 2-5% East Eurasian, which leaves... around 22-25% of some kind of SW-Asian-like "basal" West Eurasian which otherwise goes into Gedrosian (and added together to the ~30% ENF-like stuff mentioned earlier gives something similar to Geno 2.0's ~50+% SW-Asian for South Asians or rather, that run David did with Bedouin, ANE, ENA... because Geno 2.0 had the problematic Mediterranean).

Also my Harappa ASI (and Geno 2.0's Southeast Asian) is inflated by 2-3% over Eurogenes ANE K7's ASE component (~19.x% vs. ~16.x%). Others are inflated to varying degrees as well, though not often as high as mine. I think the low SNP count causes this and David somehow overcame that by emphasizing ANE. In Zack's run (not DIY) the Harappa ASI matches the ANE K7's ASE and the excess is spun off as various East Eurasian components (mostly Arctic-like).

That leads me to believe that a lot of Gedrosian for us is just ANE (something like 30-40%), but not sure why or how. Not sure where the WHG/ENF stuff in NE-Euro is coming from, but there seems to be a reason that giant 50% SW-Asian-esque chunk kind of seems to split up in behavior into two ENF-like chunks. One of these is consistent across most of the subcontinent at 28-31%, marginally but consistently increasing slowly as you move into Afghanistan, and not ANE-admixed while the other fluctuates and is ANE-admixed (perhaps more recent origin? ANI relic?).

It's not a direct relationship to Geno 2.0 because there the Mediterranean fluctuated while SW-Asian (which was associated with Gedrosian was consistent), but Geno 2.0 didn't isolate ANE either.

parasar
09-15-2014, 04:25 PM
Looking at the results of this and the new MDLP calculator I estimate for North/Northwest Indians (and some upper castes from other parts of India) there seems to be 29-30% ANE, 28-31% ENF-like Caucasian (from Harappa Caucasian and Harappa S-Indian), 14-17% ASE, 2-5% East Eurasian, which leaves... around 22-25% of some kind of SW-Asian-like "basal" West Eurasian which otherwise goes into Gedrosian (and added together to the ~30% ENF-like stuff mentioned earlier gives something similar to Geno 2.0's ~50+% SW-Asian for South Asians or rather, that run David did with Bedouin, ANE, ENA... because Geno 2.0 had the problematic Mediterranean).

...

That leads me to believe that a lot of Gedrosian for us is just ANE (something like 30-40%), but not sure why or how ...

It is my thinking that ~40% of Gedrosian is ANE and the rest ~60% basal West Eurasian.

We know the following:
1. Basal West Eurasian and ANE have very little affinity to each other so may thought of as nearly independent and if present complementary - ie one comes at the other's expense when summed as a modern component like Gedrosia.
2. Gedrosia goes down as we go north along the Indus (Brahui to Burusho).
3. ANE goes up as we go north along the Indus (Brahui to Burusho).

But the rise in ANE as we go north is not able to compensate for the decrease in basal West Eurasian, indicating to me that the latter is a slightly larger component of Gedrosia than the former.

That south-west Indus region (Makran/Balochistan) actually may be the point where basal Eurasian originates based on its Denisovan minima.

Dr_McNinja
09-15-2014, 04:26 PM
Here's some results of what I'm talking about:

McNinja-HRP0349: 28.88 ANE, 16.90 ASE, 5.50 East Eurasian, 28.01 SW-ENF, 20.71 Unknown ENF (or mine could be 30.86 SW-ENF, 17.86 Unknown-ENF, I don't know because I had 3.5% EEF instead of Caucasian in MDLP K23b which screws it up and both are possible according to my parent's results)

My Father: 28.65 ANE, 16.37 ASE, 5.51 East Eurasian, 30.97 SW-ENF, 18.50 Unknown ENF

My Mother-HRP0374: 29.40 ANE, 16.89 ASE, 3.97 East Eurasian, 29.96 SW-ENF, 19.78 Unknown ENF

Sapporo-HRP0341: 30.43 ANE, 15.14 ASE, 3.02 East Eurasian, 28.14 SW-ENF, 23.27 Unknown ENF

Haryanvi-HRP0170: 29.10 ANE, 13.30 ASE, 5.13 East Eurasian, 28.79 SW-ENF, 23.68 Unknown ENF

HRP0370-Pashtun: 26.58 ANE, 9.48 ASE, 2.83 East Eurasian, 34.35 SW-ENF, 26.76 Unknown ENF

Everest-Nepal: 28.44 ANE, 17.70 ASE, 7.85 East Eurasian, 28.61 SW-ENF, 17.4 Unknown ENF

JattPahari-HRP0350: 29.06 ANE, 16.14 ASE, 4.24 East Eurasian, 31.68 SW-ENF, 18.88 Unknown ENF

Soulblighter: 28.47 ANE, 20.63 ASE, 6.46 East Eurasian, 34.17 SW-ENF, 10.27 Unknown ENF

Gujjar-HRP0353: 30.65 ANE, 17.32 ASE, 1.38 East Eurasian, 28.24 SW-ENF, 22.41 Unknown ENF

Sein-HRP0282: 28.96 ANE, 13.67 ASE, 4.40 East Eurasian, 34.28 SW-ENF, 18.69 Unknown ENF

HRP0286-Pashtun: 26.91 ANE, 9.10 ASE, 3.37 East Eurasian, 36.51 SW-ENF, 24.11 Unknown ENF

HRP0281-Pashtun: 28.05 ANE, 13.49 ASE, 2.74 East Eurasian, 33.49 SW-ENF, 22.23 Unknown ENF

(These are using MDLP K23b... Harappa's Gedrosian seems to capture more of this Unknown ENF whereas in MDLP K23b, everyone's Gedrosian is a little lower... but everyone is in the 28-31% range almost of 'Unknown ENF', with South Indians being at the upper range of 31%, not as high as here, and the Kandahari Pashtun being even higher at 37-38% ... in Harappa I used Caucasian, SW-Asian, S-Indian, Med)

EDIT: Yeah, the reason Harappa won't work well is because NE-Euro captures their WHG-UHG so that's a wrench thrown into the mix. MDLP K23b splits up NE-Euro into its root WHG which can be added into the West Eurasian components whereas NE-Euro in Harappa has some East Eurasian.

ZephyrousMandaru
09-15-2014, 04:53 PM
Which IDs belong to Sein and Dr_McNinja?

Dr_McNinja
09-15-2014, 04:59 PM
Considering how it is low in South Indian and in far northern people (Nepal Brahmin, Pahari Jatt but not Gujjar, Sein, my parents), it could be of recent West Asian origin. So it is inversely correlated to the extra East Eurasian present in mountainous-origin folks. But my family has extra East Eurasian which might've come off from ANE so I wouldn't try to match ours with the others as representative of our origins, my mother's is probably a better fit for where we really are.

Dr_McNinja
09-15-2014, 05:00 PM
Which IDs belong to Sein and Dr_McNinja?

I am HRP0349 and he is HRP0282

Sapporo
09-15-2014, 05:02 PM
Which IDs belong to Sein and Dr_McNinja?

HRP0282 and HRP0349. I would add HRP0281 and HRP0370 as well. They are both Afghan Pashtun.

Edit: Dr_McNinja beat me to it.

Dr_McNinja
09-15-2014, 05:15 PM
It is my thinking that ~40% of Gedrosian is ANE and the rest ~60% basal West Eurasian.

We know the following:
1. Basal West Eurasian and ANE have very little affinity to each other so may thought of as nearly independent and if present complementary - ie one comes at the other's expense when summed as a modern component like Gedrosia.
2. Gedrosia goes down as we go north along the Indus (Brahui to Burusho).
3. ANE goes up as we go north along the Indus (Brahui to Burusho).

But the rise in ANE as we go north is not able to compensate for the decrease in basal West Eurasian, indicating to me that the latter is a slightly larger component of Gedrosia than the former.

That south-west Indus region (Makran/Balochistan) actually may be the point where basal Eurasian originates based on its Denisovan minima.East Eurasian also seems to inversely correlate with this basal West Eurasian, so that limits Gedrosian which starts leaving excess ANE around. In other cases (Haryana Jatts), the basal West Eurasian is almost the same but some of it more resembles WHG-UHG than ENF so there's the extra ANE left over which has not gone into Gedrosian so it goes into NE-Euro. In my family's case, the ANE is lower (turned into some East Eurasian?), the basal West Eurasian higher so we have extra non-WHG-UHG-like basal West Eurasian which is most like the "SW-ENF" present within Harappa S-Indian (complemented by extra East Eurasian) as well as our Mediterranean/Basque affinity but in Pahari Jatts, the ANE is more fully captured in Gedrosian (so high Gedrosian) but the extra non-WHG-UHG-like ENF goes into Harappa Caucasian instead of NE-Euro since their slightly higher S-Indian (by 1-2%) leaves little extra East Eurasian behind.

There's also the question of where that Siberian/Arctic-like admixture in South Asians has come from. Could ANE have brought it? Is it possible that ANE just turned into it with genetic recombination over many thousands of years? I don't think it is real because we don't get IBS/IBD matches with Siberians or such segments on people who do have genuine Arctic-like admixture, plus it sometimes shows up as Southeast Asian. Pashtun on the other hand do get Northeast Asian matches. The Southeast Asian isn't real in either Pashtun or NW-Indians I don't believe. Has anyone seen real matches to say, Vietnamese? I haven't seen a single one yet.

But, that Northeastern Europe does result in significant number of matches. There are routinely matches of ~5 generations back to MRCA with Europeans, so I don't know if that WHG-UHG is real but I'm inclined to think it somehow is connected. The sheer number of matches also makes me wonder if it's possible that just genetic recombination and chance/coincidence created these but that is also possible (I have more matches than both parents but of course, even GEDmatch's database isn't that big to make a definitive judgment).

parasar
09-15-2014, 05:36 PM
Considering how it is low in South Indian and in far northern people (Nepal Brahmin, Pahari Jatt but not Gujjar, Sein, my parents), it could be of recent West Asian origin. So it is inversely correlated to the extra East Eurasian present in mountainous-origin folks. But my family has extra East Eurasian which might've come off from ANE so I wouldn't try to match ours with the others as representative of our origins, my mother's is probably a better fit for where we really are.

It is more likely that the East Eurasian (the non-ANE portion, Y-O) at least in the north is of a more recent (relatively) origin, we really need ancient DNA to confirm.

The south is difficult to figure out. The southern part of India, ironically may be India's most west Eurasian region. On the Y side, the F-M89(xK) has its strongest representation in the south. It is just that the same east Eurasian populations have swamped both northern/NW India (R1a+R2) and Europe (R1b+R1a) making northern India look more west Eurasian in modern terms. Then the eastern and Himalayan parts had a subsequent influx of Y-O, which did not reach Europe making these regions look East Eurasian in comparison.

ZephyrousMandaru
09-15-2014, 05:38 PM
Updated MDS

https://docs.google.com/file/d/0B0cAhtjgcN_yWDJnRVk4Y21sTXM/edit

Dendrogram

http://s21.postimg.org/5hg2oak8n/Eurogenes_ANE_K_7_Dendrogram.png

Dr_McNinja
09-15-2014, 05:59 PM
Here's some results of what I'm talking about:

McNinja-HRP0349: 28.88 ANE, 16.90 ASE, 5.50 East Eurasian, 28.01 SW-ENF, 20.71 Unknown ENF (or mine could be 30.86 SW-ENF, 17.86 Unknown-ENF, I don't know because I had 3.5% EEF instead of Caucasian in MDLP K23b which screws it up and both are possible according to my parent's results)

My Father: 28.65 ANE, 16.37 ASE, 5.51 East Eurasian, 30.97 SW-ENF, 18.50 Unknown ENF

My Mother-HRP0374: 29.40 ANE, 16.89 ASE, 3.97 East Eurasian, 29.96 SW-ENF, 19.78 Unknown ENF

Sapporo-HRP0341: 30.43 ANE, 15.14 ASE, 3.02 East Eurasian, 28.14 SW-ENF, 23.27 Unknown ENF

Haryanvi-HRP0170: 29.10 ANE, 13.30 ASE, 5.13 East Eurasian, 28.79 SW-ENF, 23.68 Unknown ENF

HRP0370-Pashtun: 26.58 ANE, 9.48 ASE, 2.83 East Eurasian, 34.35 SW-ENF, 26.76 Unknown ENF

Everest-Nepal: 28.44 ANE, 17.70 ASE, 7.85 East Eurasian, 28.61 SW-ENF, 17.4 Unknown ENF

JattPahari-HRP0350: 29.06 ANE, 16.14 ASE, 4.24 East Eurasian, 31.68 SW-ENF, 18.88 Unknown ENF

Soulblighter: 28.47 ANE, 20.63 ASE, 6.46 East Eurasian, 34.17 SW-ENF, 10.27 Unknown ENF

Gujjar-HRP0353: 30.65 ANE, 17.32 ASE, 1.38 East Eurasian, 28.24 SW-ENF, 22.41 Unknown ENF

Sein-HRP0282: 28.96 ANE, 13.67 ASE, 4.40 East Eurasian, 34.28 SW-ENF, 18.69 Unknown ENF

HRP0286-Pashtun: 26.91 ANE, 9.10 ASE, 3.37 East Eurasian, 36.51 SW-ENF, 24.11 Unknown ENF

HRP0281-Pashtun: 28.05 ANE, 13.49 ASE, 2.74 East Eurasian, 33.49 SW-ENF, 22.23 Unknown ENF

(These are using MDLP K23b... Harappa's Gedrosian seems to capture more of this Unknown ENF whereas in MDLP K23b, everyone's Gedrosian is a little lower... but everyone is in the 28-31% range almost of 'Unknown ENF', with South Indians being at the upper range of 31%, not as high as here, and the Kandahari Pashtun being even higher at 37-38% ... in Harappa I used Caucasian, SW-Asian, S-Indian, Med)

EDIT: Yeah, the reason Harappa won't work well is because NE-Euro captures their WHG-UHG so that's a wrench thrown into the mix. MDLP K23b splits up NE-Euro into its root WHG which can be added into the West Eurasian components whereas NE-Euro in Harappa has some East Eurasian.
Alternative results if we treat WHG as a legitimate signal having its own distinct origin:

Using Harappa:

Everest-Nepal: 28.44 ANE, 17.7 ASE, 7.85 East Eurasian, 27 SW-ENF, 8.18 WHG-UHG, 10.83 Unknown ENF

Using MDLP K23b:

Everest-Nepal: 28.44 ANE, 17.7 ASE, 7.85 East Eurasian, 28.61 SW-ENF, 8.18 WHG-UHG, 9.22 Unknown ENF

List: https://docs.google.com/spreadsheet/ccc?key=0AuXBmvmgdkfVdFMtRHVlZDBuQ3lMcjhxMDE4V3JoY lE&usp=drive_web#gid=25

The pattern is that this unknown ENF-like component peaks in Punjabi Gujjar, then in Jatt Sikh, Afghan Pashtun, then in Pahari Jatt, then Indians (Haryanvi Jatts, North Indian Brahmins). According to MDLP K23b. According to Harappa it peaks in Punjabi Gujjar, then Jatt Sikh, then Pashtun Pakistan area and Pahari Jatts, then Afghan Pashtun and Indians ("Western" ones, such as Haryanvi Jatts or South Indian Brahmins, then Eastern ones with significant East Eurasian admixture like Nepal/North India).

So if WHG-UHG is not related to this unknown ENF-like component, then this could be Gedrosian. If WHG-UHG is related to it, it's still Gedrosian, but it just means Haryanvi Jatts are just Punjabi Jatts who for whatever reason (relative inbreeding due to less diversity than in Punjab, endogamy, genetic recombination) have their ENF resembling WHG-UHG and this unknown ENF component is peaking in them, Punjabi Jatts, and Afghan Pashtun.

Dr_McNinja
09-15-2014, 06:00 PM
It is more likely that the East Eurasian (the non-ANE portion, Y-O) at least in the north is of a more recent (relatively) origin, we really need ancient DNA to confirm.

The south is difficult to figure out. The southern part of India, ironically may be India's most west Eurasian region. On the Y side, the F-M89(xK) has its strongest representation in the south. It is just that the same east Eurasian populations have swamped both northern/NW India (R1a+R2) and Europe (R1b+R1a) making northern India look more west Eurasian in modern terms. Then the eastern and Himalayan parts had a subsequent influx of Y-O, which did not reach Europe making these regions look East Eurasian in comparison.According to this, it is the most West Eurasian part of India since the India-specific West Eurasian/Caucasian/ENF-like component is highest in them just barely. I'd actually group South Indian Brahmins into "West Indian" with Haryanvi Jatts, and North Indians into "East Indians" because of their (perhaps recent) East Eurasian which limits their West Eurasian.

Dr_McNinja
09-15-2014, 06:04 PM
Updated MDS

https://docs.google.com/file/d/0B0cAhtjgcN_yWDJnRVk4Y21sTXM/edit

Dendrogram

http://s21.postimg.org/5hg2oak8n/Eurogenes_ANE_K_7_Dendrogram.pngWould it be possible to run this with my mother, grandmother and father in addition to myself? I'm really curious if they'd wind up on the same branch as me (I don't think they would).