PDA

View Full Version : AncestryDNA results from Asia, the Pacific & Native Americans



Don Felipe
09-05-2017, 10:30 PM
In order to improve correct interpretation of AncestryDNA’s regions i have been collecting AncestryDNA results of persons from Asia, the Pacific and Native Americans.
Follow these links for more details:


Online Spreadsheet with AncestryDNA results (https://docs.google.com/spreadsheets/d/1_sjsM56m-0ewGu1RlWbg2MtEwhWJrcbc4sRnvpkUquU/edit#gid=1613813682)
AncestryDNA results from Asia, the Pacific & Native Americans (blogpost incl. screenshots of individual results) (https://tracingafricanroots.wordpress.com/2017/09/05/ancestrydna-results-from-asia-the-pacific-native-americans/)



Below my main findings:

https://tracingafricanroots.files.wordpress.com/2017/09/stats-iran-n151.jpg


A total of 156 samples with backgrounds from atleast 37 different countries across Asia, the Pacific and the Americas has been used for my main survey findings (see chart above). Even when for some of the separate nationalities i was only able to obtain a minimal sample size. As can also be verified from Ancestry's own information (https://www.ancestry.com/cs/dna-help/ethnicity)as well as the above chart the prediction accuracy of each region reported by AncestryDNA is variable. The "Asia Central" region probably being least predictive. The other regions are sometimes very predictive for selected populations close to the samples used by AncestryDNA , such as Koreans, Saudi and Pakistani/NW Indians but also Polynesians and Native Americans from Latin America. However neighbouring populations tend to require additional regions which often seem eccentric if you fixate too much on the labeling.

This seems to be inevitable given increasing genetic differentation not to be captured within one single lump category. The particular constellation of AncestryDNA’s reference panel (https://www.ancestry.com/cs/dna-help/ethnicity/reference-panel) and its algorithm being focused on detecting origins from possibly “thousands of years ago” are important aspects to take into consideration. The chart above featuring my main survey findings is therefore best understood when you keep in mind overlapping geography/genetics and ancient migrations across the continents.

Don Felipe
09-06-2017, 10:27 AM
West Asian results

https://tracingafricanroots.files.wordpress.com/2017/09/iran.jpg

***

https://tracingafricanroots.files.wordpress.com/2017/09/armciraz.jpg

***

https://tracingafricanroots.files.wordpress.com/2017/09/turk.jpg

***

https://tracingafricanroots.files.wordpress.com/2017/09/iraqi.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/lebpal.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/saudjod.jpg

Don Felipe
09-06-2017, 10:33 AM
South Asian results



https://tracingafricanroots.files.wordpress.com/2017/09/paki.jpg


***


https://tracingafricanroots.files.wordpress.com/2017/09/india.jpg


***


https://tracingafricanroots.files.wordpress.com/2017/09/bangla.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/srinep.jpg

Don Felipe
09-06-2017, 10:35 AM
Central Asian & Native American results


https://tracingafricanroots.files.wordpress.com/2017/09/uzkal.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/esk.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/cad.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/usa.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/indio.jpg

Don Felipe
09-06-2017, 10:41 AM
Asian & Pacific results

https://tracingafricanroots.files.wordpress.com/2017/09/pacific.jpg

***


https://tracingafricanroots.files.wordpress.com/2017/09/korchi.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/vietlathai.jpg


***


https://tracingafricanroots.files.wordpress.com/2017/09/camindo.jpg


***

https://tracingafricanroots.files.wordpress.com/2017/09/pinoy.jpg

mwauthy
09-06-2017, 01:23 PM
Very Interesting! It would be interesting to see how close the averages are for Ancestry after thousands of samples as opposed to your averages. It would be a neat update if they did that. Ancestry tells us what other regions reference samples are found; however, if they put a real number next to that figure it would be very helpful. Thank you for your work because at least now we have an idea of what those proportions might be.

mwauthy
09-06-2017, 01:31 PM
I wonder how accurate it is to ascribe European Jewish to samples from Saudi Arabia, Iraq, or Turkey? Perhaps the genes popping up are more ancient from the Levant region and could even be Phoenician?

Don Felipe
09-06-2017, 07:00 PM
I wonder how accurate it is to ascribe European Jewish to samples from Saudi Arabia, Iraq, or Turkey? Perhaps the genes popping up are more ancient from the Levant region and could even be Phoenician?

I have seen some reactions of West Asians and also Egyptians after receiving such results and naturally they were surprised to say the least :biggrin1: I strongly suspect that in most cases it is not any genuine Jewish ancestry being reflected by these scores but more so shared ancient origins. Eventhough i imagine there will have been plenty of occasion of some minor absorption of Jewish genes in the general area throughout the ages. Then again i'm doubtful these Mizrahi Jews would really be that distinctive genetically speaking from their surrounding populations.

I have yet to see a result of either a Mizrahi or Sephardi Jew, would be really insightful to see how much "European Jew" they receive as opposed to "Middle East" and most likely also "Italy/Greece". Sephardi's might additionally also score some "Africa North" & "Iberian Peninsula" i suppose.

zeerichardson
09-06-2017, 07:15 PM
I am trying to figure out how this all works. Can you please help me, so i have my ancestry results but how are you taking the results and averaging them out to figure out other places your ancestors may of come from?

sktibo
09-06-2017, 07:46 PM
18584

My father's result. Metis descent. I always wondered what was up with that Central Asia percentage he got but your collection of results appears to explain it: Canada (mb) must be Manitoba, where his Native American ancestry came from. Cree, Salteaux, and Assiniboine.

Mystery solved! thank you Don Felipe!

zeerichardson
09-06-2017, 07:47 PM
Are you able to help me figure out what I'm looking for to finding out some information regarding results you came up with

Don Felipe
09-06-2017, 08:16 PM
18584

My father's result. Metis descent. I always wondered what was up with that Central Asia percentage he got but your collection of results appears to explain it: Canada (mb) must be Manitoba, where his Native American ancestry came from. Cree, Salteaux, and Assiniboine.

Mystery solved! thank you Don Felipe!


Very nice! Glad this overview could be useful! Yes i would combine those two regions as both being part of his Metis heritage. That Canadian result was indeed for someone from Manitoba, about whom i did not have any further details.

From what i have seen many people of partial Native American descent tend to score additional "Asia Central" scores. That is also including Hispanics even when Latin American Natives do not seem to have any detectable socalled Central Asian affinities yet compared with Natives from the US and especially Canada. The shared Siberian origins must be at the roots of this reporting but somehow for mixed people it might act up more strongly.


I am trying to figure out how this all works. Can you please help me, so i have my ancestry results but how are you taking the results and averaging them out to figure out other places your ancestors may of come from?

I'm not sure if i understand your question correctly, is it about my survey specifically or is it about how to interpret your own results?
In case it's the latter read this:

AncestryDNA Ethnicity Estimate Help and Tips (https://www.ancestry.com/cs/dna-help/ethnicity)

sktibo
09-06-2017, 10:37 PM
Very nice! Glad this overview could be useful! Yes i would combine those two regions as both being part of his Metis heritage. That Canadian result was indeed for someone from Manitoba, about whom i did not have any further details.

From what i have seen many people of partial Native American descent tend to score additional "Asia Central" scores. That is also including Hispanics even when Latin American Natives do not seem to have any detectable socalled Central Asian affinities yet compared with Natives from the US and especially Canada. The shared Siberian origins must be at the roots of this reporting but somehow for mixed people it might act up more strongly.

Thank you! I'm sorry I can't provide a "genetically true" Metis sample for you here, as my father is as far back as I am able to go, and his mother was Scottish/English/Welsh. His father's mother was also from the British Isles, His father's father's mother was the last person in his line who would have identified as Metis, (prior to our family today identifying as Metis with the cultural revival going on) his father's father's father being French Canadian. I suppose my father is about an eighth, genetically speaking. If his result is useful to you I would be happy to share, let me know though if you do want it so I can obtain his full permission if you would like.

jortita
09-07-2017, 01:57 AM
Not sure, what is the purpose of this exercise. I would suggest even include some of the Gedmatch calculator results as Ancestry DNA ancestry estimates do not tell you much

Don Felipe
09-07-2017, 09:45 AM
Not sure, what is the purpose of this exercise.

It's to determine how predictive each separate region might be judging by the scores reported for "natives". Like i mentioned in the OP i found that the predictive accuracy is variable and contextdependent.


Ancestry DNA ancestry estimates do not tell you much

I have to disagree. Instead of hasty dismissal i would argue for finding informational value despite imperfections. I understand that in individual cases, perhaps also yours, AncestryDNA might appear to get it wrong. Especially when you fixate too much on the regional labeling. However for other people the results will be much more in line with their known background. As i said this depends on the predictive accuracy for each separate region.

Despite inherent limitations I do believe that AncestryDNA can provide you with a great deal of informational value however you do need to be able to make correct interpretations. Being aware of these group averages (even when based on minimal sample size) might give you an idea how to judge your own results.

mwauthy
09-07-2017, 12:37 PM
It's to determine how predictive each separate region might be judging by the scores reported for "natives". Like i mentioned in the OP i found that the predictive accuracy is variable and contextdependent.



I have to disagree. Instead of hasty dismissal i would argue for finding informational value despite imperfections. I understand that in individual cases, perhaps also yours, AncestryDNA might appear to get it wrong. Especially when you fixate too much on the regional labeling. However for other people the results will be much more in line with their known background. As i said this depends on the predictive accuracy for each separate region.

Despite inherent limitations I do believe that AncestryDNA can provide you with a great deal of informational value however you do need to be able to make correct interpretations. Being aware of these group averages (even when based on minimal sample size) might give you an idea how to judge your own results.

I think a lot of people are having a hard time understanding that the results are not going to most likely correlate with your immediate paper trail. Rather, the results will shed light on the ancient and historical influences of those people in your paper trail.

Ancestry does state on their website that these results are thousands of years ago. I guess many people are ignoring that fact. I find the results very helpful in context because I already know my immediate paper trail.

Since History is an important hobby of mine the results make even more sense when you analyze historical migrations and empires. For the average layman with little historical background I could see how the results could be confusing. Ancestry does try to help out by giving historical info for each region if people bother to read it.

crossover
09-28-2017, 01:05 PM
i saw some guy post one of his in laws'(who happened to be from oaxaca) results- and the in law got 100% native american on ancestrydna

Bobby Martnen
10-01-2017, 12:36 AM
I found another Turkish result

44% Caucasus/Turkey
34% Italy/Greece
12% East Europe
5% Central Asia
3% Finnish/Volga-Ural
1% Iberian Peninsula
1% West Europe

Here (https://www.quora.com/Is-there-any-Turkish-person-that-done-Ancestry-DNA-test) is the link - this is from the answer of "Eda A"

Bobby Martnen
10-02-2017, 04:24 AM
Here's another Turkish result (https://www.youtube.com/watch?v=DIJnQt6JIsg)

59% Caucasus
26% Italy/Greece
6% Middle East
4% Asia Central
3% Asia East
< 1 % Finland/Northwest Russia
< 1% Melanesia

Bobby Martnen
10-02-2017, 06:47 AM
Two more Turk results (https://www.theapricity.com/forum/showthread.php?201003-Southeastern-European-AncestryDNA-results)

Turk 1:
76% Caucasus
14% Europe
9% Middle East
1% East Asian

Turk 2:
Caucasus 56%
Italy-Greece 25%
Middle East 7%
European Jew 5%
Central Asia 4%
Asia East 2%
Asia South 1%

Don Felipe
10-08-2017, 10:15 PM
Separate page for viewing South Asian AncestryDNA results:



South Asian & Melanesian AncestryDNA results (https://tracingafricanroots.wordpress.com/ancestrydna/south-asian-melanesian-ancestrydna-results/)



Below chart based on my own preliminary survey findings:

https://tracingafricanroots.files.wordpress.com/2017/10/stats-desi.jpg




It seems likely that Pakistani samples from HGDP are being used for the "Asia South" region. Below chart based on the attachment (3rd) (https://www.nature.com/articles/ncomms14238#supplementary-information) to a recent paper done by the Ancestry research team and published in Nature (“Clustering of 770,000 genomes reveals post-colonial population structure of North America”, 2017).




https://tracingafricanroots.files.wordpress.com/2017/10/hgdp-1k-samples.jpg

jortita
10-09-2017, 06:32 AM
Separate page for viewing South Asian AncestryDNA results:



South Asian & Melanesian AncestryDNA results (https://tracingafricanroots.wordpress.com/ancestrydna/south-asian-melanesian-ancestrydna-results/)



Below chart based on my own preliminary survey findings:

https://tracingafricanroots.files.wordpress.com/2017/10/stats-desi.jpg




It seems likely that Pakistani samples from HGDP are being used for the "Asia South" region. Below chart based on the attachment (3rd) (https://www.nature.com/articles/ncomms14238#supplementary-information) to a recent paper done by the Ancestry research team and published in Nature (“Clustering of 770,000 genomes reveals post-colonial population structure of North America”, 2017).




https://tracingafricanroots.files.wordpress.com/2017/10/hgdp-1k-samples.jpg

I am curious on your NE India samples, as I am from NE India.

Don Felipe
10-09-2017, 08:24 AM
I am curious on your NE India samples, as I am from NE India.

Yes you are one of the two samples i used for calculating the NE Indian group average. The other person is from West Bengal. You can see his full results on the page i referred to above. Obviously the samplesize is minimal ;) But combined with the Bangladeshi samples and just based on what's already known from genetic studies, geography etc. the tendency towards more "Asia East" seems to be a valid finding.

jortita
10-09-2017, 10:50 AM
Yes you are one of the two samples i used for calculating the NE Indian group average. The other person is from West Bengal. You can see his full results on the page i referred to above. Obviously the samplesize is minimal ;) But combined with the Bangladeshi samples and just based on what's already known from genetic studies, geography etc. the tendency towards more "Asia East" seems to be a valid finding.

West Bengal is actually not NE India, but East India

Don Felipe
10-15-2017, 11:06 PM
Separate page for viewing East Asian (incl. Southeast Asian) & Polynesian AncestryDNA results:



East Asian & Polynesian AncestryDNA results (https://tracingafricanroots.wordpress.com/ancestrydna/east-asian-polynesian-results//)



Below chart based on my own preliminary survey findings:

https://tracingafricanroots.files.wordpress.com/2017/10/stats-mg3.jpg




It seems likely that Japanese samples from HGDP & 1000 Genomes are being used for the "Asia East" region (among others). Below chart based on the attachment (3rd) (https://www.nature.com/articles/ncomms14238#supplementary-information) to a recent paper done by the Ancestry research team and published in Nature (“Clustering of 770,000 genomes reveals post-colonial population structure of North America”, 2017).




https://tracingafricanroots.files.wordpress.com/2017/10/hdgp-1k-samples1.jpg

Bobby Martnen
10-19-2017, 05:19 AM
Thank you to Don Felipe for doing this - it's interesting to see.

Don Felipe - have you tested your own DNA?

Don Felipe
10-19-2017, 10:42 AM
Thank you to Don Felipe for doing this - it's interesting to see.

Don Felipe - have you tested your own DNA?

Thanks! See this post (http://www.anthrogenica.com/showthread.php?1473-AncestryDNA-Ethnicity-Estimates&p=277819&viewfull=1#post277819)for my AncestryDNA results.

Bobby Martnen
10-19-2017, 10:14 PM
Thanks! See this post (http://www.anthrogenica.com/showthread.php?1473-AncestryDNA-Ethnicity-Estimates&p=277819&viewfull=1#post277819)for my AncestryDNA results.

I'll try to find more results for you to incorporate

Don Felipe
11-15-2017, 10:32 AM
My results (Afghan)19800

Many thanks! Very interesting given your Taijk/Persian background. I have no confirmation for it but it seems to be quite likely that Ancestry uses Kalash and other Pakistani samples for their "Asia South" region (see post 21 (http://www.anthrogenica.com/showthread.php?11885-AncestryDNA-results-from-Asia-the-Pacific-amp-Native-Americans&p=294992&viewfull=1#post294992)). While Hazara samples are being used for their socalled "Asia Central" region. This might help to put your breakdown in better perspective. The minor socalled Native America is obviously just some shared Siberian like DNA from ancient times. Other Central Asians also show it:



https://tracingafricanroots.files.wordpress.com/2017/09/uzkal.jpg

murtazasayeed
11-15-2017, 11:27 AM
Yes I think so as well. Thanks for showing your results also.