Search This Blog

Loading...

Thursday, September 27, 2012

Jtest K14 - the Eurogenes Ashkenazi ancestry test


Update 27/11/2013: I've made the new K13 the default Eurogenes admix test at GEDmatch. It seems to hit the spot for most people. See here.

Update 07/10/2013: An upgraded version of the popular EUtest is now available at GEDmatch. See here.

...

I recently learned that the new Ancestry Painting at 23andMe will include an Ashkenazi reference group. To be honest, I’m not sure there’s much value in using a genetically bottlenecked population of varied biogeographical origins as a reference in such things. Indeed, the Ashkenazi mainly descend from a few hundred founders, but carry Central European, Eastern European, Middle Eastern, African and probably many other admixtures, as evidenced by their genome-wide and uniparental markers.

That’s quite a problem, because due to their relative inbreeding, they produce strong ancestral clusters in many analyses, like in ADMIXTURE runs. However, these clusters are made up of allele frequencies from a wide range of sources and, paradoxically, it’s the relatively more outbred populations which contributed to the Ashkenazi gene pool at its formative stages that often end up showing Ashkenazi admixture in such tests, despite not having any. I've seen this happen regularly in my experiments with ADMIXTURE and STRUCTURE, and I'm pretty sure I could find an example in a peer reviewed study if I tried.

That’s just how things work with the algorithms we have available to run these sorts of tests. Nevertheless, since 23andMe is incorporating an Ashkenazi cluster into its new painting, I thought I’d try and come up with an Ashkenazi ancestry test to perhaps get a rough idea of what we might expect. I'm using ADMIXTURE in supervised mode, and basically trying to recreate clusters that have shown up in a variety of fine-scale analyses, including my ChromoPainter run of Northern European samples. It’s still a work in progress, but below are links to files that many of you might find useful..

Jtest K14 files

Jtest averages for selected populations

EUtest K13 files

EUtest averages for selected populations

The Jtest folder contains files that can be used to make an Ashkenazi ancestry test/chromosome painting with 14 Eurasian and African clusters. The EUtest folder contains the same files, except that the Ashkenazi allele frequencies have been removed. It’s useful to cross check results from both tests, mainly to see what’s hiding under the Ashkenazi admixture if it shows up in the Jtest.

Based on a few test runs today, I’d say that the noise level for the continental clusters is much less than 1%. But it rises to a few per cent for the intra-West Eurasian clusters. In other words, if you’re European, then you might score something like 0.02% in the Sub-Saharan cluster, which basically means 0%. However, you might get around 2% in the Middle Eastern cluster, even though you’re from Central Europe, and you don’t have any recent Middle Eastern ancestry. You can blame various prehistoric and historic migrations into Europe for these seemingly quirky results, and also the fact that Mesolithic Europeans were significantly Eurasian (i.e. Siberian, Amerindian and South Asian-like).

The Ashkenazi cluster is very similar to the Middle Eastern cluster in that regard. So anyone who gets an Ashkenazi score of around 2-3% either has very distant Jewish ancestry or, more likely, none at all. However, those who show more than 25% membership in that cluster are almost certainly of fully Ashkenazi ancestry, and their genomes peppered with Ashkenazi-specific chromosomal segments.

Wow, there’s really not much difference between 2% and 25%, you might say. In fact, there is if we say there is. As always, the main thing to remember is that these clusters don’t really exist, because genetic variation is clinal, so the cluster names are basically arbitrary and it’s always the relative results that matter. That’s why to really understand what your scores mean, you need to compare them with those of other users.

Obviously, it's best to compare with people from the same ethnic and/or regional groups. If the Ashkenazi + East Med scores look relatively inflated, that's a sign of recent Ashkenazi ancestry.

Feel free to use the files above for anything you want, except commercial stuff. Please note, I make no guarantees that they’ll provide accurate results for everyone. I might update this post early next week with new and/or additional files and more tips.

...

Update 6/10/2012: The Jtest K14 and EUtest K13 will soon be available at GEDmatch, accompanied by an "Oracle" population matching analysis and maybe even a 3D genetic map. If all goes to plan, the population matching test should be able to give a decisive yay or nay to anyone wondering whether they have recent Ashkenazi ancestry.

By the way, below is a PCA based on the Jtest averages for selected populations. It was produced by one of my project members so that we could check the reliability of the 14 "ancestral" components. The samples were classified into clusters based on their highest peaking component. So, for instance, the Scots are in the light blue Atlantic cluster, along with French Basques, because the Atlantic component dominates in both groups. However, overall, they're more similar to other samples than to each other.

As per above, the plan is that GEDmatch will soon offer a 3D genetic map based on the loadings from this PCA analysis.

Update 11/10/2012: The Jtest and EUtest are now on offer at GEDmatch. The quickest way to get there is via this link to the Ad-Mix page. Then, from the drop down menus, choose Eurogenes, followed by Jtest.

First run the Admix test to check whether your Ashkenazi admixture is significantly higher than expected for your part of the world (as per above, Jtest averages for selected populations are available here). Then move on to the Oracle analysis by pressing the relevant button at the bottom of the page.

If your Ashkenazi admixture is clearly elevated, and the top 20 single and/or mixed mode Oracle results show AJ (Ashkenazi Jews) as one of your potential matches, then it’s likely you have recent Ashkenazi ancestry.

Whether that’s the case or not, you can then move on to the Chromosome Painting feature to see where the potential Ashkenazi admixture is located in your genome. It’s useful to cross check the results with those from the Ancestry Finder at 23andMe to assess their accuracy.


As already mentioned, the EUtest is exactly the same as the Jtest, but with the Ashkenazi allele frequencies taken out. You can use this option to see what’s hiding under your Ashkenazi admixture in the Jtest. To compare your results with those of selected populations from Europe, Asia and Africa, refer to the EUtest averages sheet.

Please note: it's important to interpret the results with insight. You need to learn how the system works, pay attention to the types of populations that appear in your results, consider carefully why they might be paired with other populations, and of course study the statistics in detail. Expecting a bullseye classification at the top of the Oracle list is likely to lead to major disappointment for many people, simply because I don't have enough samples to represent all of the substructures that exist around the world, especially within countries.

I’ll try and update both tests in a few weeks, after seeing how successful the whole set up is at predicting Ashkenazi admixture and locating it in the genome. One of the main goals will be to improve the accuracy of the Oracle analysis for everyone, including New World people with Amerindian admixture.

Update 21/10/2012: Below are gradient maps of a few of the ancestral clusters from the Jtest, courtesy of project member FR7.


Update 4/12/2012: The Jtest and EUtest at GEDmatch now include a new tool called the 4-Ancestors Oracle (aka. Oracle-4), as well as the 3D PCAs I promised earlier. Oracle-4 will attempt to pinpoint your ethnic group of origin, and then also work out the most likely combinations of two, three and four ancestral populations which make up your genome. However, this doesn't mean the results will actually show your ethnic group, or those of your parents (in dual mode) or grandparents (4-way mode). They might for many people, but for others they'll reflect the best possible outcomes from the reference samples available.

Enjoy, and feel free to give feedback to John at GEDmatch if you think it might be useful (but please don't spam his account).


168 comments:

ldnt said...

Thanx! How do we run those calculators? There are no .par files.

David LaTour said...

Awesome! The files worked great for me. Based on my documented ancestry, I would be about 6.25% Sephardi Jew. The Jtest indicated 3.59% Ashkenazim, and the EUtest shifted most of this into East_Med, West_Asian, and Middle_Eastern.

Ted Kandell said...

Would it be possible to filter out the Ashkenazi segments, masking the rest, and then do the EUTest painting on just those? In other words, a "two-stage test" which shows what lies behind the Ashkenazi segments automatically?

Also, perhaps you can phase the Ashkenazi segments using FastIBD, then identify them using phased sequences as either haploid, compound heterozygous, or homozygous Ashkenazi segments?

Davidski said...

^ OK, that sounds a bit complicated. But I have a feeling that if you can get enough people to use these tests, and then put together an oracle type application from ethnic averages, then you'd have a very precise Ashkenazim test for everyone.

BTW, new calc files...

https://docs.google.com/open?id=0B9o3EYTdM8lQNnRNamU5cVhrTW8

petar martinic said...

Polako,
I get the following messaage: 14 ancestral populations
159620 total SNPs
At line 121 of file DIYDodecad.f90 (Unit 30 "Jtest.14.F")
Traceback: not available, compile with -ftrace=frame or -ftrace=full
Fortran runtime error: End of file
Warning message:
running command 'DIYDodecadWin jtest.par' had status 2


https://docs.google.com/file/d/0B9o3EYTdM8lQQ21IeTZJQ2c5eVk/edit
When I put files above everything works ok, but not with the new ones.

Davidski said...

^ It'll be available at GEDmatch soon, with population matching and maybe a 3D PCA plot.

George said...

Davidski should be able to run his composite AJ specimen through the EUtest to tell us that. I've posted the composition of my AJ portion (which is only 7%, and not genealogically attested, btw) at 23andme. I'll do it here:
2% SOUTH_BALTIC
7% EAST_EURO
6% NORTH-CENTRAL_EURO
14% ATLANTIC
13% WEST_MED
25% EAST_MED
16% WEST_ASIAN
18% MIDDLE_EASTERN

David La Tour, can you state similarly where your 3.59% went? I'd like to see if your proportions are similar. Ted, how about you? You have a lot more AJ to distribute. What percentage did you get from Jtest?

Aussie said...

Had a few issues doing it caused by having too many calculators in the same R directory most likely. Forced me to sort that out. My guess is that 3.33% does represent AJ ancestry and not high level noise.

The West Asian? well perhaps there was a real wanderer in the family....

Jtest
---------

14.07% SOUTH_BALTIC
9.21% EAST_EURO
23.97% NORTH-CENTRAL_EURO
30.16% ATLANTIC
11.85% WEST_MED
3.33% ASHKENAZIM
2.13% EAST_MED
5.07% WEST_ASIAN
0.03% MIDDLE_EASTERN
0.18% SOUTH_ASIAN
0.00% EAST_AFRICAN
0.00% EAST_ASIAN
0.00% SIBERIAN
0.00% WEST_AFRICAN

EUtest
----------
14.14% SOUTH_BALTIC
9.43% EAST_EURO
24.18% NORTH-CENTRAL_EURO
30.61% ATLANTIC
12.35% WEST_MED
3.15% EAST_MED
5.64% WEST_ASIAN
0.24% MIDDLE_EASTERN
0.26% SOUTH_ASIAN
0.00% EAST_AFRICAN
0.00% EAST_ASIAN
0.00% SIBERIAN
0.00% WEST_AFRICAN

Jeremy Knowles said...

I have 3 British/Irish grandparents, and 1 Chinese grandparent. The West Asian component always seems to be there in most British Isles based populations, so I assume this is part of British side. I attribute Siberian, South Asian and East Asian to the Chinese grandparent, while West Asian to the European, but until I see the populations this is only an assumption.

My results below:

EUTEST
SOUTH_BALTIC 8.57%
EAST_EURO 9.09%
NORTH-CENTRAL_EURO 20.50%
ATLANTIC 23.63%
WEST_MED 8.80%
EAST_MED 0.05%
WEST_ASIAN 6.27%
MIDDLE_EASTERN 0.00%
SOUTH_ASIAN 3.41%
EAST_AFRICAN 0.00%
EAST_ASIAN 19.68%
SIBERIAN 0.00%
WEST_AFRICAN 0.00%


JTEST
SOUTH_BALTIC 8.54%
EAST_EURO 8.91%
NORTH-CENTRAL_EURO 20.31%
ATLANTIC 23.25%
WEST_MED 8.27%
ASHKENAZIM 2.15%
EAST_MED 0.03%
WEST_ASIAN 5.55%
MIDDLE_EASTERN 0.00%
SOUTH_ASIAN 3.31%
EAST_AFRICAN 0.00%
EAST_ASIAN 19.66%
SIBERIAN 0.01%
WEST_AFRICAN 0.00%

zardos said...

A friend of mine has significantly higher Ashkenazim in the JTEST compared to the average of all related populations, but no Ashkenazi as proven by AF with not a single Ashkenazi relative. On the other hand some pals with known Ashkenazi admixture have no higher Ashkenazim in this test.


Davidski said...

^ Well, that's just one part of the test. The other parts he has to check are the population matching and chromosome painting.

zardos said...

We looked at that too, but it doesn't change the result, which is somewhat odd. Eastmed is lower than the average by the way.
In Oracle mixed mode, the first match is a related population with 12,3 Ashkenazi.
The other population matches are not exact, but still make sense.

ronen said...

this is my jtest results

SOUTH_BALTIC 1.59%
EAST_EURO 1.57%
NORTH-CENTRAL_EURO 5.49%
ATLANTIC 6.26%
WEST_MED 14.54%
ASHKENAZIM 18.39%
EAST_MED 26.23%
WEST_ASIAN 9.92%
MIDDLE_EASTERN 14.45%
SOUTH_ASIAN -
EAST_AFRICAN 0.76%
EAST_ASIAN 0.05%
SIBERIAN -
WEST_AFRICAN 0.75%

is this mean i'm half ashkenazi in reality?(i'm fully jewish but not fully ashkenazi)

Guillaume Breteuil - Porlier said...

in the Jtest I would like please to know what is IE , GR , GE ?

Will

Davidski said...

I always use official codes for nations, so IE = Ireland, GR = Greece, and GE = Georgia.

Guillaume Breteuil - Porlier said...

thank you so about the Jtest in the Atzmon et al. and Behar et al. studies the Jewish populations are much more proportionately sampled, so this is probably why in those studies West Eurasian Jews (Ashkenazim or not) appear fully aligned in the European-Middle Eastern genetic continuum in all analysis. If you took "Ashk SNPs" I mean the people who are tested in 23 andMe or FTDNA that you used for that test are of course "fully Ashkenazim" in the Jtest .

Ben d 'Rivera said...

What's the difference between East Med, West Asian and Middle Eastern ?

Davidski said...

East Med = Eastern Mediterranean coastal areas, West Asian = Caucasus, Middle Eastern = Arabian Peninsula.

cjm said...

What does each component represent? Is South Baltic Lithuanian while East Euro is Ukraine? What is North Central Euro and Atlantic? If that makes sense. Thanks!

Davidski said...

^ The ethnic averages and PCA plot above will tell you what the components represent.

Levitylab said...

Your Jtest tool was uncannily accurate for me. I have no known Jewish ancestry, but do have 7 part-Jewish matches in 23andme's Ancestry Finder feature. Six of these had only one Ashkenazi blue bar, with one having two blue bars. Thus 8 out of 28 grandparents were Ashkenazi. So, taking the fraction 8/28 and multiplying it by the number of part-Jewish matches (7), should roughly correspond to the number of these individuals who show light green in their chromosone painting. Amazingly 8/28 times 7 is exactly 2, and exactly 2 of my 7, part-Ashkenazi, matches show light green at their chromosone positions.

I'm wondering now if other chromosone regions that show light green, might happen to be people who are unaware of Ashkenazi heritage, or are Ashkenazi's who didn't fill out their ancestry forms. I have one individual whose surname is Levy, with an old testament first name, which I won't mention for his privacy. I'm going to check if he has any light green at his chromosone position. I also have at least a dozen Ashkenazi-specific, or common, haplotypes matches, so those will be checked also.

Guillaume Breteuil - Porlier said...

Good ! I am in the same case so what is the answer ? your question is do you have Jewish ancestry ? may be thats possible ?

weirdbeyondreason said...

My dad is Irish, my mom is Polish, possible distant ASHKENAZI not sure.

Population Percent
1 ATLANTIC 24.23
2 NORTH-CENTRAL_EURO 21.71
3 SOUTH_BALTIC 20.73
4 EAST_EURO 13.22
5 WEST_MED 10.47
6 EAST_MED 3.98
7 ASHKENAZI 3.01
8 WEST_ASIAN 2.41
9 SIBERIAN 0.25

Cyndi said...

Very cool, thank you so much for doing this. Before doing my test (23andme) I thought I was 100% Ashkenazi (with obvious admixture from more than a few generations ago since I have light skin and blue/green eyes). Then I discovered my mtDNA is a Sephardic cluster (U6a7a1b), I'm a carrier for a disease that points to being Sephardic, and my various admixture tests come up with lots of Middle-Eastern/West Asia/Mediterranean/Sephardic/etc results. I love it.

My jtest results are:
SOUTH_BALTIC 5.16%
EAST_EURO 2.98%
NORTH-CENTRAL_EURO 6.84%
ATLANTIC 6.47%
WEST_MED 8.36%
ASHKENAZI 31.57%
EAST_MED 21.62%
WEST_ASIAN 6.32%
MIDDLE_EASTERN 9.51%
SOUTH_ASIAN 0.06%
EAST_AFRICAN 0.49%
EAST_ASIAN 0.62%
SIBERIAN -
WEST_AFRICAN -

My mom's side is Slovakian (Hungarian) and my dad's is Polish and Belorussian. Based on your test, I'm pretty clearly Jewish all around, which was what I assumed. I'm still trying to figure out specific locations and also how deep the Sephardic goes. It doesn't seem that it's just the maternal line back thousands of years, but it must be a pretty significant portion of my great-grandma's heritage (as far back as I can trace on her line...her surname is standard Ashkenazi I think). My various Oracle results (which I don't know how to interpret yet) are a mix of Sephardic and Ashkenazi. (Hints on where to go next are welcome.)

Annie said...

I'm new at this so please bear with me. I know you warn to interpret the results with insight and to learn how the system works. I've tried searching for some type of description that will help me to interpret the results from the EUTest I've done on my adopted daughter's data (orig. from 23&Me) but can't seem to find the right page. I'm just looking for something to tell me what the difference is between looking at the results of the Single Population Sharing and the Mixed Population Sharing and what units the Distance is given in (i.e. generations?). If there is a page I can access and read for myself, could someone point me in the right direction? Thanks!

Davidski said...

^ The actual distance isn't very important. The main thing to look at are the 3-4 populations at the top in the single pop mode, provided your daughter isn't of recently mixed origin. This will tell you her general geographic origins, and perhaps even very specific ethnic origins if the right reference group happens to be available.

However, if she's recently mixed, then the mixed pop mode is the one to focus on.

You can get an idea of whether she's of recently mixed origin by running my SPA model files. See here...

http://bga101.blogspot.com.au/2012/11/spatial-ancestry-analysis-spa-model.html

Recently mixed people show that they're mixed very clearly in the dual ancestry mode.

The only other advice I can give is for you to get used to using all the various tools, like SPA and the stuff at GEDmatch and 23andMe, and cross check all the results. After a while you'll be able to hone in on her ancestry with quite a bit of precision, unless she's a really complex mix of exotic populations that are still very poorly sampled.

LeviZoe said...

First, thank you for your clear and concise explanations.

Next, I wanted to let you know I think the J test comes closer to my known and professed ancestry than other tests previously used. With the exception of Ashkenazi, which comes at about 2-3% for me, husband and dad. (none professed or previously expected, though family history suggests sympathies and sympathetic groups).

Also, what I don't get is the differences between your results and Dodecad. Dodecad K7b has me at about 8-10% Arab....Palestinian, Bedouin, Jordanian.....Harappa world suggested about that same amount as Ashkenazy, Cypriot, etc.

But, you say 0% middle eastern. Can you explain this difference? Thanks,

Heather

Davidski said...

The Middle Eastern clusters in the different ancestry projects aren't the same clusters, because they're made up of somewhat different allele frequencies. But they're similar, based on similar samples and peaking in about the same region of the world.

So that's one potential reason for the discrepancies you're seeing. Another might be what I call the "Calculator Effect". This skews the results of people who aren't project members of Dodecad, Harappa etc. but it's not an issue with the Jtest and EUtest, because of the way these analyses were designed. In other words, you don't have to be a Eurogenes project member to get accurate results with the Jtest and EUtest.

EDP said...

Davidski,

My Jtest results indicate my Ashkenazi score is 7.89%, which is higher than the average S. Italians and North Africans get on this test. My East Med is 8.61% and my Middle Eastern is 7.42%. However, my Caucasus or West Asian is only 0.24%. I do not have any known AJ ancestry. I do have Iberian, Canarian and French as my main ancestral sources. My question is: Can my results be tied to those population groups or do they indicate admixture from another population besides those mentioned? I do not have an AJ pop. listed here. I do get Algeria four times in the Mixed Mode, along with Morocco and Mozabite Berber, three times each. Only on the Eutest do I get AJ at #20, single pop. list.

My top, single populations for both tests are Portugal, Spain, North Italian, French and Tuscan.

Henrik Alstad said...

I'm Norwegian,but my father is 25% Finn.
I got 6.65% West Asian, isnt that quite high for being Norwegian?
My father(with some finnish and lapp ancestry got 3.63% West Asian).

So my father: 3.63% West Asian
Me: 6.65% West Asian
That means my mother likely got even more west-asian admixture... how normal is this for being Norwegian?

Davidski said...

That's very unusual for a Norwegian, because the national average is much lower. The averages are listed below - look for NO...

https://docs.google.com/file/d/0B9o3EYTdM8lQOTJFaUlKWkd5TW8/edit?authkey=CO6P06UH&authkey=CO6P06UH

But I think you probably made a mistake and confused your West Med score for your West Asian score. Or perhaps you're referring to a different test? If you really do have over 6% of West Asian in the Jtest, then I'd say your mother isn't Norwegian.

Anders PĂ„lsen said...

What software was used to make these maps?

Davidski said...

I have no idea, because FR7 (Loxias at ABF) made them. You can try and PM him at ABF, or at 23andMe, but I think he's on holidays in Europe at the moment.

Henrik Alstad said...

I did one mistake: I used the EUTest and not JTest.
With the JTest, a tiny bit of the west-asian is considered ashkenazi. Not much though.
With the JTest it dropped just right below 6%.

My results:
SOUTH_BALTIC 16.54%
EAST_EURO 14.51%
NORTH-CENTRAL_EURO 30.29%
ATLANTIC 23.38%
WEST_MED 6.21%
ASHKENAZI 1.98%
EAST_MED -
WEST_ASIAN 5.95%
MIDDLE_EASTERN -
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN 1.15%
WEST_AFRICAN -


My dads results: (partly Finn & lapp)
SOUTH_BALTIC 19.26%
EAST_EURO 14.27%
NORTH-CENTRAL_EURO 26.65%
ATLANTIC 22.59%
WEST_MED 7.64%
ASHKENAZI 2.78%
EAST_MED 0.06%
WEST_ASIAN 3.15%
MIDDLE_EASTERN -
SOUTH_ASIAN 0.70%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN 2.89%
WEST_AFRICAN -

Davidski said...

That's still very high for a Norwegian, and a couple per cent higher than in Orcadians, who have the highest level of this component in Northern Europe.

If your mother is an average Norwegian, then maybe you inherited large half-segments of it from each parent, so that the combined total for you is higher than for each of your parents?

Henrik Alstad said...

Possibly. I have ordered a test for my mom as well, but it's taking forever. But I'll see when it's done.
The oracle puts me at:
95.3% South_&_Central_Swedish + 4.7% Lezgin @ 2.72
96.3% South_&_Central_Swedish + 3.7% GE @ 2.8
94.9% NO + 5.1% Lezgin @ 2.91
96.1% NO + 3.9% GE @ 3.04
97.2% South_&_Central_Swedish + 2.8% IR @ 3.22
97% South_&_Central_Swedish + 3% TR @ 3.23
97.3% South_&_Central_Swedish + 2.7% Kurdish @ 3.24
97.5% South_&_Central_Swedish + 2.5% Armenian @ 3.25
97.7% South_&_Central_Swedish + 2.3% Kalash @ 3.27
97.9% South_&_Central_Swedish + 2.1% KZ @ 3.3
97.9% South_&_Central_Swedish + 2.1% Balochi @ 3.31

I assume GE is Georgian, IR is Iranian and TR are Turkish?
There is a pattern here, as I get these kind of results in almost all the admixture calculators. Very often different weird caucasian ethnicities like Lezgin, Adyghe, Kalash, Georgian, Avars, Tabasarans, Ingush etc.

However, I wonder if this is actually an indication of ancestra from that area, or if it's just a natural variation above the normal Norwegian average

Eric974 said...

Hello, I am reunionese (french island in the Indian Ocean Reunion), mainly descendants from french, indo-portuguese (mix of portuguese + indian/north indian), malagasy (mix of african + asians). We are the result of 350 years of admixing.

Here are my results

Jtest EUtest
SOUTH_BALTIC 6.44 6.50 -0.06
EAST_EURO 4.75 5.04 -0.29
NORTH-CENTRAL_EURO 10.77 11.11 -0.34
ATLANTIC 17.37 18.12 -0.75
WEST_MED 12.06 12.69 -0.63
ASHKENAZIM 5.10
EAST_MED 5.14 6.54 -1.4
WEST_ASIAN 4.05 4.77 -0.72
MIDDLE_EASTERN 0.27 1.05 -0.78
SOUTH_ASIAN 7.69 7.70 -0.01
EAST_AFRICAN 3.93 4.04 -0.11
EAST_ASIAN 8.51 8.56 -0.04
SIBERIAN 1.19 1.17 -0.02
WEST_AFRICAN 12.74 12.71 -0.03

My Ashkenazim seem to have picked up from the West/East Med + Atlantic and West Asian in EUtest. Is this significant ?

FTDNA saw nothing mideast in me. Doug Mc Donald had a bit of mideast but could not tell. Globe13 sees 3.8% southwest_asian and 8.61% west_asian. Dodecad dv3 sees 7.90 west_asian and 2.14 southwest_asian.

Thank you!

Eric974 said...

Sorry if posted twice!

Hello, I am reunionese (french island in the Indian Ocean Reunion), mainly descendants from french, indo-portuguese (mix of portuguese + indian/north indian), malagasy (mix of african + asians). We are the result of 350 years of admixing.

Here are my results

Jtest
SOUTH_BALTIC 6.44
EAST_EURO 4.75
NORTH-CENTRAL_EURO 10.77
ATLANTIC 17.37
WEST_MED 12.06
ASHKENAZIM 5.10
EAST_MED 5.14
WEST_ASIAN 4.05
MIDDLE_EASTERN 0.27
SOUTH_ASIAN 7.69
EAST_AFRICAN 3.93
EAST_ASIAN 8.51
SIBERIAN 1.19
WEST_AFRICAN 12.74

EUtest
SOUTH_BALTIC 6.50 +0.06
EAST_EURO 5.04 +0.29
NORTH-CENTRAL_EURO 11.11 +0.34
ATLANTIC 18.12 +0.75
WEST_MED 12.69 +0.63
EAST_MED 6.54 +1.4
WEST_ASIAN 4.77 +0.72
MIDDLE_EASTERN 1.05 +0.78
SOUTH_ASIAN 7.70 +0.01
EAST_AFRICAN 4.04 +0.11
EAST_ASIAN 8.56 +0.04
SIBERIAN 1.17 +0.02
WEST_AFRICAN 12.71 +0.03

My Ashkenazim seem to have picked up from the West/East Med + Atlantic and West Asian in EUtest. Is this significant ?

FTDNA saw nothing mideast in me. Doug Mc Donald had a bit of mideast but could not tell. Globe13 sees 3.8% southwest_asian and 8.61% west_asian. Dodecad dv3 sees 7.90 west_asian and 2.14 southwest_asian.

I have no known Ashkenazim in my family (genealogy going back to 1600). only have a few portuguese, spanish, italian and south french that could stick out???

Thank you!

Davidski said...

You should compare your Jtest and Oracle results with those of other people of very similar ancestry. If your score is consistently higher, and you show Ashkenazi in the Oracle results, but they don't, then that would suggest some minor Ashkenazi ancestry.

You should also check on the Jtest chromosome painting if the Ashkenazi influence is concentrated in any particular parts of the genome, and then see whether you match Ashkenazi users in those areas at GEdmatch.

EDP said...

Davidski, you never answered my question or commented on it.

EDP

Davidski said...

I don't know what your ancestry is precisely, but if its mostly South Italian and North African, then you can expect an unusually high Ashkenazi score without any Ashkenazi ancestry.

The only time users of South Italian, Sicilian, North African, Greek, Cypriot, Maltese, etc. ancestry should wonder about Ashkenazi admixture based on this test is if the Oracle ouctomes repeatedly scream out Ashkenazi.

EDP said...

No, I don't have South Italian ancestry, and whatever North African ancestry I may have is via the Canary Islands, which is already mixed with Iberian. I'm Cuban.

Henrik Alstad said...

I tried the K36 calculator, and it put me at:
22.0% Fennoscandian
18.86% North sea
15.02% North_Atlantic
8.34% French
7.94% East_Central_Euro
7.38% Iberian
7.20% Eastern Euro
6.38% North_Caucasian
5.19% Central_Euro
1.07% Basque
0.41% Amerindian
0.19% Italian
0.01% South_Central_Asian

Note the high North_Caucasian percentage...I think it match the west-asian component in EU/JTest...but from what I've seen, British and Orcadians score significantly less in Northern-caucasus than me

AJ75 said...

Hi,

These were my results on the EUtest (sorted by Oracle). I'm wondering what the regions represent (my guesses in brackets on the basis of whats been said above):

1 EAST_MED 26.22 (Levant)
2 WEST_MED 15.8 (Is this North Africa / Spain?)
3 MIDDLE_EASTERN 14.18 (Arabian Peninsula)
4 NORTH-CENTRAL_EURO 13.8 (Germany / Czech Republic)
5 ATLANTIC 10.58 (?)
6 WEST_ASIAN 9.63 (Caucaus)
7 SOUTH_BALTIC 4.23 (?)
8 EAST_EURO 3.4 (Russia / Ukraine?)
9 SIBERIAN 0.81 (Siberia)
10 EAST_AFRICAN 0.74 (Kenya / Tanzania)
11 EAST_ASIAN 0.5 (China)
12 SOUTH_ASIAN 0.11 (Indonesia)

Cuah123 said...

Hello

I compared both tests EU and J test then substracted the differences

SOUTH_BALTIC 0.14
EAST_EURO 0.55
NORTH-CENTRAL_EURO 0.47
ATLANTIC 0.8
WEST_MED 0.81
ASHKENAZI 0
EAST_MED 1.76
WEST_ASIAN 0.93
MIDDLE_EASTERN 1.18
SOUTH_ASIAN 0
EAST_AFRICAN 0.12
EAST_ASIAN 0.06
SIBERIAN 0.04
WEST_AFRICAN 0.01

It appears a most fractional make up other than the obvious Host populations contributing to the admixture, namely Middle East and East_Med. I do have E1b1 (predicted) and J2 M92 (mmf) in the mix, as well as I1 (mf) and R1b L48 (father).

I wonder how does your test compare to the recent (2 year) studies Behar et al and Avshalom Zoossmann-Diskin 2010, pointing to the recent migration of Italians who became a bulk of Ashkenazi Jews.

In other words, the inverse, a large I believe it was 40% from memory of Italians are autosomally comparable to EEJ. Italians in other words were the host population.
http://www.biology-direct.com/content/5/1/57

Cuah123 said...

one more note of interest I took a similar test using one of Dienekes tools and should up 0% Ashkenazi...

AJ75 said...

I don't think that the geneticists writing about the 'Jewish' origins add much to the debate anymore - anyone can replicate the results of their research using the admixture tools we ourselves are using. Most of them have a pretty poor grasp of history and reach very black and white conclusions - Israelites, Italians, Germans, Khazar, from the moon etc.

I think Davidski's ad-mixture results give a very clear picture for example of Ashkenazi Jews - a combination of near eastern and mediterranean DNA with more limited northern European ancestry. If I had to analyse the results the most plausible conclusion would be a core population comprised of Jews from Judea mixed with early converts from the greco-roman world supplanted over time by converts from the host populations in which the Ashkenazi Jews lived in. As this pretty much ties in with known history I can't see any reason to reinvent history to support a single origin theory.

Ronda S said...

Hi. I did my testing on Familytreedna and their system categorized my DNA as 100% Western European - French. That's it. Just the one thing. But when I try your tests I'm all over the place. I do know that I'm 25% Northern Italian, probably 15-20% Welsh, some English, German, Scotch-Irish, Dutch.... Wish I could tell you percentages better but I'd have to be able to prove back to all of my immigrant ancestors to do it. I would give you my kit # but I prefer not to post it on open forum. The numbers that show up in the Jtest & EUtest are similar to each other, so here are the EUtest figures. I sure hope you can give me some insight on what they might mean:

South Baltic: 10.03%
East Euro: 9.09%
North Central Euro: 24.69%
Atlantic: 27.45%
West Med: 16.23%
Ashkenazi: 2.35%
East Med: 6.01%
West Asian: 1.79%
Middle Eastern: 1.64%
South Asian: 0.71%
East African: --
East Asian: --
Siberian: --
West African: --

Davidski said...

You should run the various Oracles linked to the EUtest and Jtest. They'll give you a good idea of what the results mean in terms of ethnic affinities.

Ronda S said...

Thank you for the suggestion. To be frank, it still doesn't make a lot of sense to me why you've got so many different sorting methods / tests. Is there a blurb on that somewhere that I could read to better familiarize myself with the utility? Also, some of the 2-letter abbreviations that come back after Oracle searches don't have easy translations (at least for me). It would really help me out if there were an abbreviations key.

J Sten said...

Hi Davidski,

The AJ percentage in my results seems elevated, but I'm not sure if that's because I have some Hungarian ancestry (about 12.5%). Below are the results and the oracle-x; what do you think?

Admix Results:

# Population Percent
1 SOUTH_BALTIC 18.88
2 EAST_EURO 14.57
3 NORTH-CENTRAL_EURO 21.69
4 ATLANTIC 22.32
5 WEST_MED 7.33
6 ASHKENAZI 4.81
7 EAST_MED 2.79
8 WEST_ASIAN 3.98
9 MIDDLE_EASTERN 0.00
10 SOUTH_ASIAN 0.00
11 EAST_AFRICAN 0.05
12 EAST_ASIAN 0.00
13 SIBERIAN 0.00
14 WEST_AFRICAN 3.58


Pct. Calc. Option 1

1 Scottish 44.76%
2 Northwest_Russian 25.76%
3 South_&_Central_Swedish 11.84%
4 AJ 6.23%
5 Yoruba 4.05%
6 AT 3.51%
7 HU 3.26%
8 Lezgin 0.59%
9 RO 0.00%
10 French_Basque 0.00%

Total RMSD: 0.419273

Davidski said...

It's hard to say from those results. The oracles that show multiple ancestries are probably better. In any case, an Ashkenazi score of 4.81% is unlikely to indicate anything but very minor Ashkenazi ancestry.

But you can check where your Ashkenazi segments are located in your genome via the chromosome painting at GEDmatch. This will tell you whether they're proper segments, rather than just small dispersed fragments as a result of noise, and if they overlap with Ashkenazi segments at 23andMe.

J Sten said...

I received 0% AJ from the chromosome painting on 23andMe, but I did have several part-Jewish matches from the ancestry finder. I ran a "by segment" analysis in DIYDodecad for two half-Jewish matches, and got back about 20% AJ in both segments (using Jtest). It appears that these matches from the ancestry finder could be real.

Also, I found on the Anthrocivitas forum that you provided some extra files for the K36 calculator for Ashkenazi and Sephardic clusters. I only got about half a percent of each overall, but there were elevated percentages in the segments above.

CatchingAllTheSouls said...

So is this basically saying that I am Scottish?

1 50% Cornish +50% FR @ 4.101
2 50% FR +50% Scottish @ 4.127
3 50% FR +50% Orcadian @ 4.405
4 50% FR +50% IE @ 4.570
5 50% English +50% FR @ 5.031
6 50% ES +50% Scottish @ 5.178
7 50% ES +50% Orcadian @ 5.340
8 50% PT +50% Scottish @ 5.346
9 50% FR +50% NL @ 5.375
10 50% ES +50% IE @ 5.671
3081 iterations.

Davidski said...

No, it actually says you're somewhere between most western Brits and most French. So maybe something like ancient Bretonnian?

http://en.wikipedia.org/wiki/Brittany_%28administrative_region%29

Although that might be a coincidence if you're not actually from ancient Britain or Brittany, but rather someone of mixed European descent from the US.

CatchingAllTheSouls said...
This comment has been removed by the author.
Davidski said...

If you're a really mixed European American, with several ethnic groups in your wood pile, then pretty much any combination of European reference samples is possible. That's because there isn't much genetic differentiation across most of Europe to begin with, and when things get broken up a bit within the genome then all hell breaks loose.

CatchingAllTheSouls said...
This comment has been removed by the author.
TribalCat said...

Here are my EUtest and Jtest results:

EUtest:
45.41% East Asian
11.87% Atlantic
10.61% West Mediterranean
7.45% North Central European
6.42% South Asian
5.76% East Mediterranean
4.91% West African
3.29% South Baltic
1.98% East European
1.84% Middle Eastern
0.46% East African

Jtest:
45.40% East Asian
11.54% Atlantic
10.30% West Mediterranean
7.25% North Central European
6.37% South Asian
4.925 West African
4.90% East Mediterranean
3.25% South Baltic
2.46% Ashkenazi
1.80% East European
1.38% Middle Eastern
0.42% East African

With the Ashkenazi only showing at 2.46%, is it more likely that the number comes from the Mediterranean, and that I don't have any Ashkenazi ancestry?

Levitylab said...

Here is my EUtest results with 3 populations. Another 23andme user provided a link to the IOC codes at Wikipedia, pointing out that HU was for Hungary. I surmised that NO was for Norway and PL for Poland, as both countries would make perfect sense with my 3/4 British Isles and 1/4 German ancestry. But then I noticed that while the three letter abbreviation for Poland is POL, there's two other countries Palau (PLW) and Palestine (PLE) that also have the letters P and L in them. Palau, of course, is ruled out, but I wondered if Palestine might be in position 10, the weakest, because of "calculator effect".


1 50% Scottish +25% HU +25% South_&_Central_Swedish @ 1.391
2 50% Scottish +25% HU +25% NO @ 1.434
3 50% IE +25% HU +25% South_&_Central_Swedish @ 1.715
4 50% IE +25% HU +25% NO @ 1.774
5 50% Orcadian +25% HU +25% NO @ 1.795
6 50% Scottish +25% DK +25% HU @ 1.859
7 50% Orcadian +25% HU +25% South_&_Central_Swedish @ 1.899
8 50% Orcadian +25% AT +25% South_&_Central_Swedish @ 1.926
9 50% Orcadian +25% HU +25% Orcadian @ 1.976
10 50% IE +25% English +25% PL @ 2.006


Davidski said...

PL is Poland.

Davidski said...

And by the way, there's no calculator effect with the EUtest and Jtest.

CatchingAllTheSouls said...

Doug McDonald's results were very similar to EUTest so I would say that between the two of you I have basically found my ancestry. He has me at mostly English but with the rest in any of 5 other places. Now the most logical one for me is 72 percent English and 28 percent French. I have no known Italian, Spanish, or Romanian ancestry. The French ancestry could have easily come from colonial times as my family has been here since the 1600s. That said we have remained relatively "pure" ancestry wise in that we are nowhere near as mixed as other Americans. Now would you say, in your expert opinion, that these results on the EUtest reflect a mixed English and French lineage? I would say they do but I am no expert:

Using 1 population approximation:
1 Cornish @ 6.101
2 FR @ 7.311
3 English @ 8.276
4 NL @ 8.570
5 West_&_Central_German @ 9.043
6 Orcadian @ 9.325
7 IE @ 9.769
8 Scottish @ 10.140
9 DK @ 12.299
10 AT @ 13.708
78 iterations.



Using 2 populations approximation:
1 50% Cornish +50% FR @ 4.086
2 50% FR +50% Scottish @ 4.241
3 50% FR +50% Orcadian @ 4.489
4 50% FR +50% IE @ 4.761
5 50% English +50% FR @ 4.982
6 50% ES +50% Scottish @ 5.308
7 50% FR +50% NL @ 5.383
8 50% PT +50% Scottish @ 5.471
9 50% ES +50% Orcadian @ 5.502
10 50% FR +50% West_&_Central_German @ 5.778
3081 iterations.



Using 3 populations approximation:
1 50% English +25% AT +25% French_Basque @ 3.579
2 50% West_&_Central_German +25% French_Basque +25% West_&_Central_German @ 3.858
3 50% West_&_Central_German +25% English +25% French_Basque @ 3.875
4 50% West_&_Central_German +25% French_Basque +25% NL @ 3.895
5 50% West_&_Central_German +25% Cornish +25% French_Basque @ 3.919
6 50% FR +25% Cornish +25% Scottish @ 3.951
7 50% NL +25% French_Basque +25% West_&_Central_German @ 4.006
8 50% Cornish +25% Cornish +25% ES @ 4.050
9 50% Cornish +25% FR +25% FR @ 4.086
10 50% FR +25% Cornish +25% Cornish @ 4.086
71107 iterations.



Using 4 populations approximation:
1 AT + English + English + French_Basque @ 3.579
2 AT + Cornish + English + French_Basque @ 3.786
3 AT + English + French_Basque + NL @ 3.815
4 FR + French_Basque + South_&_Central_Swedish + West_&_Central_German @ 3.841
5 French_Basque + West_&_Central_German + West_&_Central_German + West_&_Central_German @ 3.858
6 English + French_Basque + West_&_Central_German + West_&_Central_German @ 3.875
7 AT + English + French_Basque + West_&_Central_German @ 3.891
8 French_Basque + NL + West_&_Central_German + West_&_Central_German @ 3.895
9 Cornish + French_Basque + West_&_Central_German + West_&_Central_German @ 3.919
10 Cornish + FR + FR + Scottish @ 3.951
11 AT + English + French_Basque + Orcadian @ 4.003
12 French_Basque + NL + NL + West_&_Central_German @ 4.006
13 FR + French_Basque + NL + South_&_Central_Swedish @ 4.018
14 AT + Cornish + French_Basque + NL @ 4.031
15 English + French_Basque + NL + West_&_Central_German @ 4.045
16 Cornish + Cornish + Cornish + ES @ 4.050
17 English + FR + French_Basque + South_&_Central_Swedish @ 4.057
18 AT + Cornish + French_Basque + West_&_Central_German @ 4.085
19 Cornish + Cornish + FR + FR @ 4.086
20 FR + French_Basque + West_&_Central_German + West_&_Central_German @ 4.091

CatchingAllTheSouls said...

And here are the "raw" admixture results before the oracle examines it:

SOUTH_BALTIC 13.33%
EAST_EURO 7.14%
NORTH-CENTRAL_EURO 24.64%
ATLANTIC 29.65%
WEST_MED 14.92%
EAST_MED 6.45%
WEST_ASIAN 1.96%
MIDDLE_EASTERN -
SOUTH_ASIAN 1.69%
EAST_AFRICAN 0.06%
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN 0.14%

CatchingAllTheSouls said...

Do you think 3 populations gives the "best" overall possible ancestry? My oracle-4 results are inconsistent in that 2 populations will say one thing, 3 will say another, and then 4 will throw in ethnicities that the other do not have. Well I posted my results not too long ago so you can see what I mean!

Chad Rohlfsen said...

Davidski. I'm rather confused about my numbers, especially the west Asian component. My numbers are like the Basque in West Asian and more like central to eastern European in a high east Mediterranean. My ancestry is a bit questionable. Mostly British Isles/Irish as far as have been told. I'm guessing theres less of that and more Central European than was known. 23andme picked up some Ashkenazi in me. Not a surprise with rumors about family lines. Can you take a stab at where my ancestry must lie besides 40-50% british/irish, the rest as far as I know is French, German and Danish. Maybe that's high too. Im kind of stuck here. How does someone get higher east med scores and basically zero west Asian? My west med seems rather low too. It seems very odd as my numbers aren't even close to known ancestry when comparing to the EUtest population averages. Any light you can shed on this or tips on what to search for in my family tree are greatly appreciated.
EU Test
Population
SOUTH_BALTIC 12.98%
EAST_EURO 11.98%
NORTH-CENTRAL_EURO 26.67%
ATLANTIC 28.35%
WEST_MED 9.81%
EAST_MED 9.77%
WEST_ASIAN 0.15%
MIDDLE_EASTERN -
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN 0.28%
WEST_AFRICAN -

JTest

Population
SOUTH_BALTIC 12.95%
EAST_EURO 11.85%
NORTH-CENTRAL_EURO 26.44%
ATLANTIC 27.89%
WEST_MED 9.37%
ASHKENAZI 2.90%
EAST_MED 8.46%
WEST_ASIAN -
MIDDLE_EASTERN -
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN 0.13%
WEST_AFRICAN -


Davidski said...

Have you run the oracles? Do they show what you expect (more or less, based on the references available)?

Chad Rohlfsen said...

I find the oracles off. Maybe my understanding of my family is off. I usually show as central/west german or mixed Germanic at the top. Ill see a common use of Romanian or Serbian or Italian mixing with English to get my results in the population mix section. Arent my east med and west Asian at odds with each other? My atlantic is nw European like but other stuff looks eastern and south eastern. It's rather confusing as I can only go back so far on my family. I am recently finding that the further back I go on some lines, the more eastern and southern they get. Maybe it's right. I really don't know. I just wish we knew now what we probably will about autosomal dna in a few years. I know my family is mixed, but with what is the question. No paper trail beyond the mid-late 1800s for most lines

Davidski said...

It's always difficult to interpret results when someone is mixed in a complex way, and most of their ancestry comes from genetically similar ethnic groups. I don't think that will ever change.

The best course of action for you would be to test your parents. This might give you more details about the unexpected admixture you're seeing in your own genome.

Chad Rohlfsen said...

Here's my Oracle results.
# Population (source) Distance
1 West_&_Central_German 6.24
2 Cornish 6.52
3 NL 6.52
4 English 6.53
5 Orcadian 7.83
6 DK 7.89
7 Scottish 8.83
8 IE 8.84
9 NO 10.14
10 South_&_Central_Swedish 10.63
11 FR 10.98
12 AT 11.05
13 North_Swedish 13.01
14 HU 16.7
15 PT 17.32
16 ES 17.71
17 Serbian 17.72
18 North_Italian 20.47
19 South_Finnish 20.54
20 RO 20.73

Mixed Mode Population Sharing:


# Primary Population (source) Secondary Population (source) Distance
1 93.8% Cornish + 6.2% Druze @ 5.47
2 94.1% English + 5.9% Druze @ 5.61
3 90.3% West_&_Central_German + 9.7% French_Basque @ 5.72
4 56.2% West_&_Central_German + 43.8% Cornish @ 5.78
5 84.8% Cornish + 15.2% Serbian @ 5.79
6 85.2% English + 14.8% Serbian @ 5.85
7 87.3% English + 12.7% RO @ 5.86
8 91.2% Orcadian + 8.8% Druze @ 5.86
9 76.8% English + 23.2% AT @ 5.89
10 77.4% Cornish + 22.6% AT @ 5.92
11 95.1% English + 4.9% Samaritan @ 5.92
12 95.2% Cornish + 4.8% Samaritan @ 5.93
13 79.1% DK + 20.9% North_Italian @ 5.96
14 88.6% Cornish + 11.4% RO @ 5.99
15 90.6% West_&_Central_German + 9.4% ES @ 6
16 79.6% West_&_Central_German + 20.4% Scottish @ 6.02
17 93.2% English + 6.8% GR @ 6.02
18 76.6% DK + 23.4% ES @ 6.03
19 95.6% NL + 4.4% Druze @ 6.05
20 61.6% West_&_Central_German + 38.4% English @ 6.05

Chad Rohlfsen said...

Here's my Oracle X
Pct. Calc. Option 2


0 Unable to determine 0.01%
1 West_&_Central_German 49.38%
2 Scottish 10.88%
3 Cornish 10.17%
4 DK 9.01%
5 French_Basque 7.75%
6 Druze 5.89%
7 South_Finnish 2.82%
8 North_Swedish 2.05%
9 English 2.01%
10 ES 0.02%

What in here jumps out at you for someone with my ancestry if theres any way to tell?

Chad Rohlfsen said...

Ok. My father is alive but my mother passed away. If I just test my father is that enough?

Davidski said...

It would still help.

CatchingAllTheSouls said...

What is AT in the oracle results? This is showing up in the oracle.

CatchingAllTheSouls said...

I am also wondering if having significant Irish ancestry would show up as Cornish? I asked my father what he knows of his side of the family and he said that he knows of Irish/English. My mother says English.

Efim45 said...

Hello, I am RU19, M540435 on Gedmatch. David, a recent conversation with my father and new Oracle results have shed some new light on what was once a speculated minor component of my ancestry. When my father was young, he was at a family meet/greet (from his paternal grandfather's side; our R1a1a line). They were discussing the origins of our people, and they really stressed that we were Mordvins(my father doesn't remember whether they said Moksha or Erzya)

All of my paternal grandfather's ancestors were from a village not far from the historical Finno-Ugric settlement area. In the 1700s, peasants were brought in to the village from all around to work in a factory. Some were brought from Mordovia itself.

Fast forward to my oracle results, and I don't know what to think anymore. Could it be that my grandfather is more Mordvin than I thought?:

JTest Mixed-Mode,2,3,4 Oracle Results (all that include Erzya):

Mixed Mode:
13. 69.4% Erzya 30.6% AJ @ 6.903

Two Populations:
3 50% Erzya +50% RO @ 8.805

Three Populations:
4 50% Belorussian +25% AJ +25% Erzya @ 3.479
5 50% West_Russian +25% AJ +25% Erzya @ 3.486
6 50% PL +25% AJ +25% Erzya @ 3.527
8 50% Northwest_Russian +25% AJ +25% Erzya @ 3.615

Four Populations:
1 AJ + Erzya + Northwest_Russian + PL @ 2.865
2 AJ + Erzya + Northwest_Russian + West_Russian @ 3.134
4 AJ + Erzya + LIT + PL @ 3.208
5 AJ + Belorussian + Erzya + West_Russian @ 3.217
6 AJ + Belorussian + Erzya + PL @ 3.227
7 AJ + Erzya + LIT + West_Russian @ 3.236
8 AJ + Erzya + PL + West_Russian @ 3.311
10 AJ + Erzya + Northwest_Russian + UA @ 3.346
13 AJ + Belorussian + Erzya + Northwest_Russian @ 3.415
19 AJ + Erzya + LIT + Ukrainian-Russian @ 3.470
20 AJ + Erzya + Northwest_Russian + Ukrainian-Russian @ 3.470

Oracle X (Pct. Calc. Option 1):
1 AJ 28.39%
2 Erzya 26.33%
3 LIT 21.88%
4 Northwest_Russian 19.87%
5 French_Basque 3.28%
6 Druze 0.22%
7 Bedouin 0.03%
8 Ukrainian-Russian 0.00%
9 RO 0.00%
10 Udmurt 0.00%

Since the Oracle X result for AJ is pretty much spot on, could it be that my paternal grandfather is mostly Erzya/Moksha?

CatchingAllTheSouls said...

Your oracle results are quite consistent throughout. My results seem to show that I am a mix between French and English but the oracle X thinks I am Cornish. I think that with me I am of more than one European ethnic group so my results are just not consistent. AncestryDNA is updating their autosomal admixture results so maybe they can properly break everything down clearer? I suppose only time will tell. Doug McDonald indicates that I am mixed with French.. maybe 30 percent? Still everyone's results seem consistent throughout the oracle-4 except for me. It's like it shows different groups at 2, 3, and 4 populations. Very strange.

Efim45 said...

I think it'll take a

anen said...

I'm mexican and ran the EUtest and got 0.03 WEST_MED and 0 EAST_MED so it's definitely not Spain.


The highest Caucasian populations for me was ATLANTIC and NORTH-CENTRAL_EURO which would be from my spanish side.

Liz Davenport said...

Hello! My name is Liz, a.k.a. Kit #742089 :-). I have been wondering for decades about the hinted-at possibility of some Jewish ancestry in my family. My paternal grandmother's family name was Hartberger, although I realize that that doesn't prove anything. It seems that that family arrived in America in the first half of the 18th century, or a little later in that century. I have always had the impression, and it is only that, that perhaps I did have a Jewish ancestor a number of generations back (the first one who arrived in America, perhaps) and that intermarriage might have begun right away (my gg-grandfather was a Southern Baptist minister, for example).

Here are my JTest results. I would be grateful if you would give me your opinion of them as related to possible Jewish ancestry:

JTest:

Population
SOUTH_BALTIC 12.75%
EAST_EURO 9.17%
NORTH-CENTRAL_EURO 24.69%
ATLANTIC 28.14%
WEST_MED 12.39%
ASHKENAZI 3.03%
EAST_MED 3.12%
WEST_ASIAN 6.41%
MIDDLE_EASTERN -
SOUTH_ASIAN 0.17%
EAST_AFRICAN 0.06%
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN 0.06%

Oracle X Pct. Calc. Option 2:

Pct. Calc. Option 2

0 Unable to determine 0.01%
1 Cornish 37.07%
2 Orcadian 26.46%
3 Scottish 11.45%
4 IE 7.31%
5 GE 5.42%
6 Sardinian 4.25%
7 French_Basque 3.89%
8 LIT 3.08%
9 North_Italian 1.05%
10 Lezgin 0.00%

I'm a bit new to this (and loving it!!), so not sure if you could use my EUTest results also, so here they are:

EUTest results:

Population
SOUTH_BALTIC 12.78%
EAST_EURO 9.32%
NORTH-CENTRAL_EURO 24.93%
ATLANTIC 28.56%
WEST_MED 12.86%
EAST_MED 4.24%
WEST_ASIAN 6.88%
MIDDLE_EASTERN -
SOUTH_ASIAN 0.30%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN 0.10%


OracleX Pct. Calc. Option 2:

Pct. Calc. Option 2

0 Unable to determine 0.01%
1 Cornish 46.54%
2 Scottish 13.95%
3 South_&_Central_Swedish 8.16%
4 IE 6.27%
5 GE 6.26%
6 Orcadian 6.24%
7 Sardinian 4.84%
8 French_Basque 4.38%
9 LIT 3.34%
10 Lezgin 0.00%

Thank you so much for your time, and for all your good work.

Davidski said...

Unfortunately, I don't think you have any Ashkenazi ancestry with a percentage score of just over 3%.

Liz Davenport said...

OK! Thank you for your quick reply. :-)

Seinundzeit said...

I had an idea for a supervised Admixture run involving all of your South Asian participants and populations. Basically, everyone from Afghanistan to Bangladesh. Perhaps you could set the Georgians, the Onge, one Northeast Asian/Siberian population, one Southeast Asian population, and the Yoruba, as "pure" references. It would be a lot of fun, and you could compare the results with the Moorjani et al paper (I understand that the results are not comparable, but it would still be fun). You also don't have any of the limitations of these authors. Now, I do understand the Georgians are not a perfect proxy for the "ANI". To my knowledge (correct me if I am wrong), the authors agree with Metspalu et al. West Eurasian ancestry in South Asia is distinctive. But it seems you can't find a better living proxy than Georgians for something of this nature. Same caveat with the Onge, but it will still be really interesting. It would be very fascinating to see how much recent East Eurasian ancestry is being construed in these academic papers as "ASI". And the minor African ancestry in some individual South Asians could also be separated out. Even if one ignores all of this, it would just be a truly enjoyable experiment, and a lot of fun, especially considering the samples you have. I hope you do consider this worth your time and effort. If not, that is absolutely okay. But I do hope you consider this. You could even create a DIY version for South Asian users of GEDMatch. Thank you for your time.

Waviking26 said...

I am American of French/German/British descent as far as the paper trail is goes. I am curious of what to make of my south asian percentage. Here are my results:



# Population Percent
1 ATLANTIC 28.95
2 NORTH-CENTRAL_EURO 24.67
3 WEST_MED 12.92
4 EAST_EURO 11.74
5 SOUTH_BALTIC 9.63
6 EAST_MED 7.03
7 SOUTH_ASIAN 2.67
8 WEST_ASIAN 1.45
9 WEST_AFRICAN 0.94

Mike Shake said...

Dear Davidski,
Thanks for all this great work!
Could you cast a swift eye on my results? They do look like some results above and my question is somewhat similar: what Jewish part could there be in it?
The thing is, my family history has it I should have something of 1/8 Armenian (or more, that's for sure, but maybe mixed with some Ossetia); 1/8 Jewish or less (not clearly documented) and 3/4 Slavic (of very vague origins, mostly Russian I guess from East/West/North/South...).
FTDNA showed bunches of matches with Jews, most of them in just two regions: chr 3 that paints in Ashkenazi or North_Central_Europe in Jtest and clear Jewish matches 12cM in chr 15 that paints in South_baltic and East_Europe.

Here are my Jtest results:
# Population %
1 SOUTH_BALTIC 26.69
2 EAST_EURO 19.99
3 NORTH-CENTRAL_EURO 16.28
4 ATLANTIC 11.10
5 WEST_MED 5.71
6 ASHKENAZI 3.38
7 EAST_MED 8.62
8 WEST_ASIAN 5.65
9 MIDDLE_EASTERN 0.29
10 SOUTH_ASIAN 1.04
11 EAST_AFRICAN 0.07
12 EAST_ASIAN 0.14
13 SIBERIAN 1.02
14 WEST_AFRICAN 0.02


Pct. Calc. Option 1
1 UA 63.28%
2 LIT 9.42%
3 East_Finnish 8.46%
4 Druze 7.42%
5 Ukrainian-Russian 4.40%
6 Lezgin 3.68%
7 AJ 1.82%
8 IR 1.52%
9 Assyrian 0.00%
10 Selkup 0.00%

What is UA, actually? Ukrainian?
I would be very surprised in having so much non-Russian Slavic genes, but who knows… Finnish though??

And is it compatible with having a Jewish great-great-grand mom?

And last, what are all these Druze and Lezgin genes add up to? I could be Armenian or Osset but Lezgin? Druze? :)

Would appreciate any answer!!

Thanks a lot again,

Mike

Davidski said...

What you're seeing is the effect that your mixed ancestry produces. I'd say the reason you're getting Ukrainian (UA) instead of Russian, is because Ukrainians are very similar to Russians, except they're more southern genetically. The algorithm then has to account for the rest of your southern ancestry, and in this case the Druze and Lezgins are a better fit than Armenians and Ossets.

In other words, there are many ways to describe the same result when someone's mixed. If you try the other calculators they'll give you plenty of other options, and it's likely that Armenian and Ossetian will feature in some of them.

Mike Shake said...

Thanks for such a rapid response!
Given that the Slavic component in my case can also reasonably be anything else (polish, ukrainian), I understand that it bogs the algorithm.

I was mostly wondering though why Armenian or Ossetian (historically 1/8th so about 12%) practically never come up in my results, whereas most calculators show 20-35% Middle East/Mediteranean or similar Southern admixtures.
Can it be accounted for by Jewish admixtures? Or, given that Ashkenazi part appears so low (3.3%), is it something else?
In other words, is Jewish or a different Non-slavic component more likely? Or will it be impossible to tell?
Cheers,
Mike

Davidski said...

The ethnic groups that the algorithm chooses to describe your ancestry shouldn't be taken too literally for a number of reasons. For instance, there's no guarantee that your Armenian ancestor had exactly the same admixture proportions as the Armenian reference panel here. That's because there are differences between Armenian populations and individuals, with some being more Middle Eastern and others more Transcaucasian or even European.

So the algorithm has to compensate to make everything fit. That's what it did here, by picking ethnic groups that you don't have any direct ties to, but that you do resemble in terms of various ancestral components.

Your Jewish ancestry could well be putting a bit of a spanner in the works, but it's impossible to tell.

Sam Dee said...

So here is my post/question...the original test I took at Ancestry said that I was European (various areas) with some Ural connections. I am not seeing any difference between conclusions. But, I do have a strange situation arising: I have a lot of Ancestry matches to Jewish people, and we are talking about a third of all predicted cousin matches, and some are like 4th cousins. My family came to North America at least 300 years ago, 80% were the founder families in Quebec, and the other 20% were most likely pilgrims. A small percent were native. Below are my results:
J test
SOUTH_BALTIC 10.66%
EAST_EURO 9.65%
NORTH-CEN_EURO 24.23%
ATLANTIC 25.09%
WEST_MED 16.37%
ASHKENAZI 2.36%
EAST_MED 6.81%
WEST_ASIAN 3.94%
MIDDLE_EASTERN 0.81%

EU test
SOUTH_BALTIC 10.67%
EAST_EURO 9.84%
NORTH-CEN_EURO 24.43%
ATLANTIC 25.43%
WEST_MED 16.65%
EAST_MED 7.39%
WEST_ASIAN 4.28%
MIDDLE_EASTERN 1.21%
That is about 60% European areas and 40% all around the Mediterranean. Is it possible that I have Sephardic ancestors? When I run some of the other models in Gedmatch I get a lot of results right around the area of the middle east, one said Baloch, also Iberian...

Unknown said...

Maybe i can help with that. I have about the EXACT same as you do. but what i'm finding out, is that Ashkenazi Jews have a lot of mediterranean in the admixture. I was also troubled along the lines because my ancestors on my moms maternal side all were German Jews. Now on my dad's side, there was a question of Sephardic lines, but a 2nd cousin i have has no Jewish admixture according to FTDNA. So we are waiting for an Uncle's test results to see what happens. Anyways, i also was wondering if maybe i had both Sephardic AND Ashkenazim admixture, but after reading some info, i think maybe its just Ashkenazim. but the problem i have with just Ashkenazim comes from the fact that my fathers line is Portuguese/Spanish/Mexican, and my Mothers line is Scottish/English/German......all lines have been traced pretty well. now the problem lies in the fact that at FTDNA, i come out with 53.52% NorthEast European, 11.22% Mayan/Colombian (the mexican side), and 35.25% MiddleEast(Jewish)....specifically jewish...no other middle eastern......so if all the Jewish was Ashkenazim, then where is my fathers DNA besides the Mayan/Colombian? being half Scottish/English/German, would explain the NorthEast Europe, but then what about the 35% Jewish.......so i guess as soon as i have my parents and uncles tests come in, i'll be able to better see how their admixture is, and maybe the numbers can be seen better. I wish we could get more Sephard and Ashkenaz and mixed Sephard/Ashkenaz to post their results. it would really help those like us.

here's the link for the Middle East in Ashkenazim other Jews

www.wikipedia.org/wiki/Genetic_studies_on_Jews

and here's my results

JTest
SOUTH_BALTIC 8.61%
EAST_EURO 12.62%
NORTH-CENTRAL_EURO 20.98%
ATLANTIC 20.97%
WEST_MED 12.40%
ASHKENAZI 2.21%
EAST_MED 6.93%
WEST_ASIAN 1.23%
MIDDLE_EASTERN 0.83%
SOUTH_ASIAN 1.58%
EAST_AFRICAN -
EAST_ASIAN 3.79%
SIBERIAN 6.18%
WEST_AFRICAN 1.68%


EUTest
SOUTH_BALTIC 8.64%
EAST_EURO 12.76%
NORTH-CENTRAL_EURO 21.13%
ATLANTIC 21.27%
WEST_MED 12.68%
EAST_MED 7.48%
WEST_ASIAN 1.56%
MIDDLE_EASTERN 1.21%
SOUTH_ASIAN 1.60%
EAST_AFRICAN -
EAST_ASIAN 3.81%
SIBERIAN 6.17%
WEST_AFRICAN 1.69%

You can see the Ashkenazi in my tests are coming from the Middle Eastern, Mediterranean, and European segments.

Sam Dee said...

Thank you. You are helping me. In the big picture most people have crossed through many regions in life and all of us at one point could have been in the Middle East and Africa, but the tests I had done was a basic one and should just mean recent past, about two thousand years or so. As far as I know all of my ancestors came to the New World in 1600s and stayed put in very specific places, Quebec, and New England. And from the names I found married to each other, they didn't mix too well with other families. The New England group have like only four surnames marrying into each other. With the people in Quebec, they married within as well with other founder families. There is another test Dodecad Africa 9, these are my results:
Admix Results
# Population Percent
1 Europe 71.54
2 SW_Asia 24.76
3 NW_Africa 3.71

Using 3 populations approximation:
1 50% Tuscan +25% French Basque +25% Morocco Jews @ 0.934
2 50% French Basque +25% Druze +25% North_African_Jews @ 1.282
3 50% North Italian +25% French Basque +25% North_African_Jews @ 1.527
4 50% French Basque +25% Jordanians +25% Morocco Jews @ 1.646
5 50% French Basque +25% Druze +25% Morocco Jews @ 1.724
6 50% North Italian +25% North Italian +25% North Italian @ 2.171
7 50% Tuscan +25% French_Basque +25% North_African_Jews @ 2.195
8 50% French_Basque +25% North_African_Jews +25% North_African_Jews @ 2.239
9 50% North_African_Jews +25% French_Basque +25% French_Basque @ 2.239
10 50% North_Italian +25% North Italian +25% Tuscan @ 2.353
As seen, it is trying to mix in Jewish origins from North Africa. That would match results from another genetic program that said I have ties to Morocco. Also, that other program I used said that I have ties to the group-Mexicans from Los Ang. and Indians from New Mex.. I can only assume that is the crossing of Spain and Native American, and that is the only place where my Native American has shown up.
Maybe some of our Ancestors are from the time before the Second Temple destruction, "old Jewish stock". We had more time to dissolve in the stream. Thank you again.

RSoza said...

I was the previous post marked as "Unknown September 9, 2013 at 7:54 PM"


Anyways, there should be a FAQ made on how to use the admixture utilities for some who are just beginning. I've got a grasp, but what exactly is the number at the end, when doing the Oracle-4?

example

6 50% Morocco_Jews +50% Tuscan @ 10.842
50% North_African_Jews +50% North_Italian @ 10.974

What is the @10.842 and @10.974, and what is the least squares method? And if its just an algorithm, then how does that relate in Oracle-4 to the other results?

Thanks.

Sam Dee said...

I also would like to know, I was assuming that it was saying that:
1) 50% Tuscan +25% French Basque +25% Morocco Jews @ 0.934 is more closely related (probable) then say,
10) 50% North Italian +25% North Italian +25% Tuscan @ 2.353.

Like a variation on probable, the smaller the number, the more likely the scenario. Hopefully we get a response from the HOST.

Davidski said...

Yes, the smaller number indicates a lower genetic distance, and thus a better fit (based on the reference individuals currently available).

Sam Dee said...

Okay, I have one more question...maybe two: as stated above, I fit into a very tight community in Quebec, and they mingled uninterrupted for at least 300 years, obviously that is why I have such a strong European connection, but can that actually blur an original ethnic background out of context? Can it over saturate results?
I have been running all the Gedmatch scenarios (as well as other sites),I was even tested for carrier status and am a carrier for something rare but most prevalent in Iranian Jews. There is an undertone in all of it, that of the Mediterranean, North Africa, and Middle East: when I run the Oracle mixes the closest comes in for .8_ _ grouping with Iberian/Tuscan, Basque, and then some Middle east origin like Morocco, or Druze. When I do the Oracle on the European populations the nearest matches are 1.5_ _ or larger. Do you think it is safe to assume that my ancestors migrated out of West Asia?

Davidski said...

Unfortunately, it's not safe to assume that, especially if in large part you're basing this on data from calculators that suffer the calculator effect.

http://bga101.blogspot.com.au/2012/05/beware-calculator-effect.html

There’s actually a lot of ancient Mediterranean and even West Asian admixture in the French population, and when the calculators aren’t performing at their best these fairly exotic ancestry proportions can increase significantly for most people.

RSoza said...

I would like to add though, that if the community you come from is that completely isolate, "tight knit", then there won't be much dilution of the DNA. So the genes will be fairly similar to the same ones they came over with, because they are just giving the same genes over and over.

RSoza said...

FTDNA says i have 35% Jewish Ancestry.....specifically Jewish ancestry. But FTDNA has my cousin at 0% Jewish ancestry. Why is his Ashkenazi so much higher than mine?

His:

SOUTH_BALTIC 7.03%
EAST_EURO 8.30%
NORTH-CENTRAL_EURO 12.41%
ATLANTIC 19.10%
WEST_MED 13.91%
ASHKENAZI 6.14%
EAST_MED 4.40%
WEST_ASIAN 0.19%
MIDDLE_EASTERN 5.88%
SOUTH_ASIAN 3.76%
EAST_AFRICAN 2.46%
EAST_ASIAN 5.18%
SIBERIAN 10.98%
WEST_AFRICAN 0.26%



Mine:

SOUTH_BALTIC 8.61%
EAST_EURO 12.62%
NORTH-CENTRAL_EURO 20.98%
ATLANTIC 20.97%
WEST_MED 12.40%
ASHKENAZI 2.21%
EAST_MED 6.93%
WEST_ASIAN 1.23%
MIDDLE_EASTERN 0.83%
SOUTH_ASIAN 1.58%
EAST_AFRICAN -
EAST_ASIAN 3.79%
SIBERIAN 6.18%
WEST_AFRICAN 1.68%

Davidski said...

I don't know much about the FTDNA test. But from the little I do know about it, I'd say that it sometimes produces fairly coarse results. So if you've been classified as 35% Jewish and that doesn't look right, then you should investigate this result with as many reliable tools as you can find.

RSoza said...

Actually, the 35% from FTDNA was fairly accurate considering i have fairly recent Ashkenazi admixture from my grandparents who were 100% German Jews. Both my cousin and I can trace our families pretty well, and he does not have any Jewish relatives. So i was surprised by JTest. But when i run through the Oracles-4, on other tests like Dodecad, no Jewish populations show up in his results, but they show up in mine.

David Joyce said...

Hello again,

I have another question for ya! I heard that the data from gedmatch is for "deeper" ancestry and not modern populations. This is true?

thanks much

RSoza said...

Ahhhhhh HaH! Thank you for replying to my questions Davidski, but i get it now. I read the HarappaWorld and Dodecad Project blogs and everything is clear now. None of the results should be taken literally or to be used for ancestry. this is all just population comparison and at best can only be regional. And those with real mixed ancestry are going to have a hard time comparing populations. I do have to say that both Dodecad and Harappaworld calculators actually got both of our results in sync with what FTDNA has, plus FTDNA has the largest user database to compare to. According to the other blogs, even the slightest bit of difference can skew the results. So while JTEST did was not compatible with the Jewish markers we have, the other calculators were. Thanks for suggesting me to run ALL the tools further.

Davidski said...

The Jtest has an Ashkenazi cluster, but membership in that cluster doesn't imply Jewish ancestry.

However, those with Ashkenazi ancestry will see large Ashkenazi segments on their Jtest chromosome paintings, and also AJ in their oracle results. Other tests won't pick up Ashkenazi segments because they don't have Ashkenazi clusters.

I explained this in the write up above.

Davidski said...

The Ad-mix proportions reflect ancient and recent gene flows, and then the Oracles attempt to pinpoint recent ethnic affinities.

RSoza said...

oh yes! I read your posts, just sometimes i don't pick up on things, and the other blogs were blatantly clear to me. Yes, the chromosome paintings are EXTREMELY helpful to me. The tools you have posted are invaluable to me. Thanks again for all the help and info you provide.

Davidski said...

Well I do occasionally ramble on a bit when trying to explain DNA stuff. But it's actually very simple in this case:

The Jtest is essentially for finding Ashkenazi-specific chromosome segments (via the Jtest chromosome painting) and signals of recent Ashkenazi ancestry via the Jtest Oracles.

On the other hand, it's probably best to ignore the initial Jtest Ad-mix proportions, because the Ashkenazi cluster shows up in almost everyone's results.

Taylor said...

Hello, I am very new to all this so please excuse me if I'm going about this wrong, but I was hoping someone could tell me if I'm Jewish. I have a "Cherokee" ancestor on my father's side that seems to come up as a Jew (probably Sephardi--from what I can tell as a newbie) in my jtest results. 23andme nor any other test I've tried has given me any Native American figures so I'm starting to really believe it was a Jew and not a Cherokee. Before I proceed I should also note that I have my mom's jtest results pasted below and they are 2 points lower than mine, both of my parents coming from loosely similar populations according to 23andme, so I believe that my dad's jtest results would be fairly high if I had access.

My Jtest:

Population
SOUTH_BALTIC 13.01%
EAST_EURO 10.40%
NORTH-CENTRAL_EURO 26.95%
ATLANTIC 26.37%
WEST_MED 12.07%
ASHKENAZI 5.85%
EAST_MED 2.86%
WEST_ASIAN 2.05%
MIDDLE_EASTERN -
SOUTH_ASIAN 0.25%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN 0.18%
WEST_AFRICAN -


My Eurogenes K9:

Population
South Asian 1.24%
Caucasus 5.08%
Southwest Asian 0.95%
North Amerindian + Arctic -
Siberian 0.82%
Mediterranean 28.03%
East Asian -
West African -
North European 63.88%

My mom's less Jewish Jtest:

Population
SOUTH_BALTIC 12.67%
EAST_EURO 9.67%
NORTH-CENTRAL_EURO 30.07%
ATLANTIC 26.60%
WEST_MED 12.00%
ASHKENAZI 3.98%
EAST_MED 1.66%
WEST_ASIAN 3.35%
MIDDLE_EASTERN -
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -

Any help is greatly appreciated!!! Thanks in advance!

Taylor said...

I should have included my relevant Oracle results with my last post probably. Pardon me kind internet people.

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 94.8% Cornish + 5.2% AJ @ 2.58
2 94.6% English + 5.4% AJ @ 2.64
3 61.5% DK + 38.5% FR @ 2.66
4 80.5% English + 19.5% FR @ 2.7
5 57% Cornish + 43% NL @ 2.77
6 89.3% English + 10.7% ES @ 2.79
7 89.5% English + 10.5% PT @ 2.84
8 52.8% Cornish + 47.2% English @ 2.84

AJ shows up in spots one and two, otherwise it's absent.

Davidski said...

The Jtest looks for Ashkenazi ancestry, and you need to read carefully the post above to learn how to get the most out of it.

Also, I'm really not sure why Cherokee ancestry would be similar to Sephardic ancestry? The former are Native Americans, while the latter a Mediterranean population.

Davidski said...

Did you check the chromosome painting for AJ segments, and if they're there, do you find that you see matches with Ashkenazi Jews in those areas when using various genealogical tools?

Taylor said...

I think the Cherokee thing was false, I don't think there is a similarity as such. I'm trying to identify what they meant by "Cherokee". Pardon me.

Taylor said...

I suppose that will be the next thing I learn about. I'm so new to this I don't even know what I don't know and I really appreciate you tolerating me.

Taylor said...

Here is my chromosome painting from jtest on gedmatch. If anyone has a sec and can tell me what I'm looking for that'd be Awesome. Also, thanks very much for your help so far David. http://tayloroliphant.blogspot.com/2013/09/jtest-admixture-proportions.html

Davidski said...

I can take a look at it, but can you upload a smaller version?

Taylor said...

Sure, I think I have it setup here: http://tayloroliphant.blogspot.com/2013/09/jtest-admixture-proportions_16.html

Thanks again!

Davidski said...

OK, I doubt you've got anything but minor Ashkenazi ancestry based on that chromosome painting. I can only see one AJ spot that qualifies as perhaps more than noise, and that's sitting on chromosome one (near the top there). But you do show some large eastern Mediterranean tracts, and these could be the result of Sephardic or other Jewish ancestry. On the other hand they could mean South Italian, Greek of Lebanese ancestry.

Taylor said...

Fascinating! Are you aware of any tests that would be good for determining if I'm Sephardi? Also, my best friend is Jewish, can I tell him I might be a very distant member of his people or is it still a bit too early? Thanks again for your help. I will call it a wrap on my questions after these two as you have been more than kind!

Davidski said...

I don't think there are any DIY tests that can confirm minor Sephardic ancestry and differentiate it from other eastern Mediterranean ancestry. You definitely have something minor from that part of the Mediterranean, but the results don't say anything about the ethnic origins and age of this admixture. It might be something Jewish, but that's not certain, so you can't really say you're part Jewish at this point.

Taylor said...

David, I am honestly amazed at how helpful you have been. I said I'd drop the questioning because I'm sure you have important things to do, so don't feel the need to answer this, but I found an additional tool on 23andme that gave me my Ashkenazi percentage. I posted the results at: http://tayloroliphant.blogspot.com/2013/09/additional-ashkenazi-data.html If you have time: Do you think this is accurate?

Davidski said...

Well that seems to be in line with your Jtest chromosome painting. In other words, if you do have any Ashkenazi ancestry, then it's very minor and distant. Therefore, it's pretty clear that your eastern Mediterranean admixture comes from a different source, which may or may not be Jewish.

Taylor said...

You are the man! I wish every good thing in life upon you good sir.

David Joyce said...

Hello again,

I was wondering,are chromosomes related to a distance of time? For example, does your first chromosomes tell you these are your parents chromosmes?

Thanks much :}

Davidski said...

Sorry, I don't really understand the question. In any case, it seems like your query pertains to genetic genealogy rather than biogeographic ancestry, so I think you'll find better resources online than me to clear that up for you.

David Joyce said...

No Worries! I will check out other sources as suggested, thanks for your time.

David Joyce said...

I apologize for so many questions! This is all new to me especially the oracle-4 feature. Am I correct in that this program judges distance over time? And that the lower the number the higher the chance you came from there? For example:

Cornish @ 4.150
2 NL @ 5.050
3 West_&_Central_German @ 5.589
4 English @ 6.148
5 Orcadian @ 6.294
6 IE @ 6.527
7 Scottish @ 8.215
8 DK @ 8.880
9 FR @ 9.207
10 NO @ 10.658

I am closer to Cornish than NL?

Thanks for your time :}

g and r said...

Hi David,
I think this got lost on the wrong posting so here it is again;

I curious as to why my Ashkenazi on Jtest to Oracle X jumped from 3.42 to 8.16? Is this enough to substantiate a closer generation? Is it because of the Eastern/Western Mediteranean numbers? The Mixes seem about right with what we know, mothers both sides Swedish to at least 1600 with one branch originating in Finland. One unknown maternal great grandfather. Fathers side eastern European/Russian/northern Poland another unknown maternal great grandfather(always the rumour of Jewish ancestor. Also a small bits of Scandinavian on fathers side.(marauding Vikings??)

Kit Number: M045540

Admix Results: Jtest and Oracle X
# Population Percent
1 SOUTH_BALTIC 17.85
2 EAST_EURO 19.92
3 NORTH-CENTRAL_EURO 23.71
4 ATLANTIC 21.80
5 WEST_MED 7.70
6 ASHKENAZI 3.42
7 EAST_MED 3.22
8 WEST_ASIAN 2.36
9 MIDDLE_EASTERN 0.00
10 SOUTH_ASIAN 0.00
11 EAST_AFRICAN 0.00
12 EAST_ASIAN 0.00
13 SIBERIAN 0.00
14 WEST_AFRICAN 0.00


Pct. Calc. Option 2

0 Unable to determine 0.02%
1 North_Swedish 45.73%
2 AJ 8.16%
3 West_Russian 6.97%
4 RO 6.89%
5 IE 6.51%
6 Serbian 6.01%
7 Ukrainian-Russian 5.60%
8 South_Finnish 5.46%
9 NO 5.28%
10 French_Basque

EU Test and Oracle X

Admix Results:

# Population Percent
1 SOUTH_BALTIC 17.88
2 EAST_EURO 20.20
3 NORTH-CENTRAL_EURO 23.92
4 ATLANTIC 22.23
5 WEST_MED 8.13
6 EAST_MED 4.37
7 WEST_ASIAN 2.91
8 MIDDLE_EASTERN 0.36
9 SOUTH_ASIAN 0.00
10 EAST_AFRICAN 0.00
11 EAST_ASIAN 0.00
12 SIBERIAN 0.00
13 WEST_AFRICAN 0.00


Pct. Calc. Option 2

1 North_Swedish 62.72%
2 RO 9.91%
3 AT 8.61%
4 West_Russian 7.76%
5 HU 5.44%
6 French_Basque 3.02%
7 Cornish 1.13%
8 North_Russian 1.09%
9 Serbian 0.31%
10 Samaritan

Here's my 23andMe breakdown

From father

8.8% French and German
4.2% Scandinavian
1.0% Finnish
0.7% British and Irish
12.4% Nonspecific Northern European
13.1%
Eastern European
4.3% Ashkenazi

Southern European
0.2% Iberian
0.2% Nonspecific Southern European
5.1% Nonspecific European

From mother

24.0% Scandinavian
5.4% Finnish
2.7% British and Irish
0.6% French and German
16.3% Nonspecific Northern European
0.3% Eastern European
0.1% Ashkenazi
0.6% Nonspecific European
< 0.1% Sub-Saharan African
< 0.1% Unassigned

Any Ideas or logical explanations?
Rhonda

Davidski said...

Generally speaking 3.42% AJ isn't high, and the jump to over 8% in one of the oracles might be some sort of glitch. If you really have Ashkenazi ancestry then you'll be seeing AJ regularly in a couple of the oracles.

In any case, you can check the Jtest chromosome painting for AJ segments and then see whether these correlate with Ashkenazi matches. If they do, then that's a sure sign of Ashkenazi ancestry, even if only minor.

g and r said...

I did the Chromosome painting and there are some small segments and some large segments. 13 of my chromosomes have an AJ percentage. Low of .2% high of 17.1% (3 at 10% or higher, 4 at 5-9% ) and the paint seems to show this. my gedmatch number M045540

Ryan Dell'Erba said...

Davidski,

My overall results for the JTest:

SOUTH_BALTIC 8.68%
EAST_EURO 8.04%
NORTH-CENTRAL_EURO 23.88%
ATLANTIC 22.63%
WEST_MED 14.99%
ASHKENAZI 2.09%
EAST_MED 7.59%
WEST_ASIAN 9.49%
MIDDLE_EASTERN -
SOUTH_ASIAN 1.85%
EAST_AFRICAN 0.43%
EAST_ASIAN -
SIBERIAN 0.34%
WEST_AFRICAN -

My reduced sized JTest chromosome painting: http://ww2.gedmatch.com:8006/autosomal/gifs/A671417_DB1478.png

And then my Ashkenazi percentages by chromosome:

01.) 1.8%
02.) 4.0%
03.) -
04.) -
05.) -
06.) 5.0%
07.) -
08.) -
09.) 0.1%
10.) 2.1%
11.) 3.9%
12.) 0.8%
13.) 21.4% (second largest percentage on this chromosome)
14.) -
15.) 6.9%
16.) 5.6%
17.) -
18.) 8.1%
19.) 9.8%
20.) 13.9% (third largest percentage on this chromosome)
21.) 8.9%
22.) -

Thoughts? If I did have a Jewish relative, it would have to be through my father because my mom is like 97%+ Irish and maybe a little bit of Scottish, and I think if it is true it would be through the Italian side of my father's family, because many of the towns with high numbers of Jews are in that area, including the town of Turi (apparently one of the most Jewish towns in Italy), which is quite close to Alberobello (15.6 miles between them according to Google), where (most of) my dad's family came from.

Davidski said...

The chromosome painting doesn't show up.

In any case, 2% Ashkenazi is too low to indicate any recent Ashkenazi ancestry. But if you do have some distant minor Ashkenazi ancestry, then you should be able to see that on your chromosome painting. Look for fairly large Ashkenazi tracts that correlate with Ashkenazi hits from various genealogical (IBD) tools.

BeccaLT said...

I recently got my 23andme results back and was surprised at a 20% AJ result, plus some surprise Italian. There are some skeletons in the closet on my dad's side, all we knew about his mom was that her mother was a Hungarian emigrant who died young and her father was not known.

Running the J-test here seems to show the same thing in the 3 and 4 population results with AJ in each result:

Admix Results (sorted):

# Population Percent
1 NORTH-CENTRAL_EURO 21.69
2 ATLANTIC 19.92
3 WEST_MED 14.53
4 SOUTH_BALTIC 12.22
5 EAST_EURO 10.56
6 ASHKENAZI 8.20
7 WEST_ASIAN 6.18
8 EAST_MED 5.45
9 MIDDLE_EASTERN 1.25


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


Using 2 populations approximation:
1 50% AT +50% FR @ 6.038
2 50% NO +50% North_Italian @ 6.199
3 50% North_Italian +50% South_&_Central_Swedish @ 6.390
4 50% North_Italian +50% North_Swedish @ 6.599
5 50% DK +50% North_Italian @ 6.634
6 50% FR +50% HU @ 6.664
7 50% English +50% RO @ 6.865
8 50% Cornish +50% RO @ 7.199
9 50% Cornish +50% Serbian @ 7.241
10 50% North_Swedish +50% Tuscan @ 7.367
3081 iterations.



Using 3 populations approximation:
1 50% NO +25% AJ +25% FR @ 4.773
2 50% South_&_Central_Swedish +25% AJ +25% FR @ 4.894
3 50% NL +25% AJ +25% NL @ 5.034
4 50% NL +25% AJ +25% IE @ 5.068
5 50% NL +25% AJ +25% Cornish @ 5.101
6 50% South_&_Central_Swedish +25% AJ +25% ES @ 5.181
7 50% NL +25% AJ +25% Orcadian @ 5.185
8 50% NO +25% AJ +25% ES @ 5.220
9 50% NL +25% AJ +25% West_&_Central_German @ 5.240
10 50% IE +25% AJ +25% AT @ 5.299
89877 iterations.



Using 4 populations approximation:
1 AJ + FR + IE + South_&_Central_Swedish @ 4.693
2 AJ + FR + IE + NO @ 4.772
3 AJ + FR + NO + NO @ 4.773
4 AJ + FR + NO + South_&_Central_Swedish @ 4.808
5 AJ + FR + NL + South_&_Central_Swedish @ 4.820
6 AJ + FR + NL + NO @ 4.821
7 AJ + FR + Orcadian + South_&_Central_Swedish @ 4.827
8 AJ + FR + IE + North_Swedish @ 4.868
9 AJ + FR + NO + Orcadian @ 4.876
10 AJ + FR + South_&_Central_Swedish + South_&_Central_Swedish @ 4.894
11 AJ + Cornish + FR + South_&_Central_Swedish @ 4.966
12 AJ + FR + NL + North_Swedish @ 5.014
13 AJ + FR + North_Swedish + Orcadian @ 5.015
14 AJ + FR + NO + West_&_Central_German @ 5.020
15 AJ + NL + NL + NL @ 5.034
16 AJ + Cornish + FR + NO @ 5.035
17 AJ + Cornish + FR + North_Swedish @ 5.042
18 AJ + FR + South_&_Central_Swedish + West_&_Central_German @ 5.048
19 AJ + English + FR + South_&_Central_Swedish @ 5.050
20 AJ + IE + NL + NL @ 5.068

Davidski said...

Yes, you clearly have Ashkenazi ancestry, and it should also show up very well on the Jtest chromosome painting.

LadyBelka1 said...

Hi David,my kit nr F294441 on Gedmatch:)
I have strange results on both EUTest and JTest, it shows me mostly Ukrainian or Ukrainian-Russian instead of Russian- my main ancestry. All my ancestors to 5 generations back had pure and classical Russian surnames and lived in Russia, maybe with some Lithianian influence on my father's side. But definitely no Ukrainians, However we always suspected some southern admixture on my mother's side, because my grandmother had very dark curly hair and very Med look and her father looked even more Mediterranean. By the way we found his last name in some Jewish lists but still cannot be sure he was Jewish.
Other tests and oracles which show big Russian part, also show 10-15% Italy or Spain or 7-10% Jews (mostly Ashkenazi but also from Morocco). My Jtest results are:
1. SOUTH_BALTIC 28.89
2 EAST_EURO 26.28
3 NORTH-CENTRAL_EURO 18.36
4 ATLANTIC 9
5 WEST_MED 5.95
6 WEST_ASIAN 4.6
7 ASHKENAZI 3.31
8 SIBERIAN 2.6
9 EAST_MED 0.73
10 WEST_AFRICAN 0.28
Now I am a bit confused...
Do you think it can be Italian/Iberian, Jewish or some other southern ancestry?

Many thanks :)

Davidski said...

Russians are often difficult to tell apart from Ukrainians genetically, and that's because the difference between many Russian and Ukrainian populations is ethnic rather than genetic.

Also, with an Ashkenazi score of 3.3%, you probably don't have any significant Ashkenazi ancestry. But you might want to check your Jtest chromosome painting for any large Ashkenazi segments. These could indicate minor Ashkenazi ancestry.

MixedRoots said...

Your DNA work and blog are amazing. I uploaded my DNA...I have a heavy percentage of SW Asian. Who knew?

Davidski said...

OK, but make sure you compare your results to those of other people of similar ancestry to confirm that your SW Asian percentage is unusually high.

Anyway, I've got a new test on the way. So try that when it's up.

Mike Shake said...

Hi again,

I've been reading your advice on checking specific fragments on chromosomes that would paint as Ashkenazim to see if they fit a known Ashkenazi matches. A couple of questions:
1) How big and clear do you think a fragment should be to be considered as "Ashkenazi" painted? I get some chromosomes between 10% and 22% Ashkenazi (overall percentage) and if I look closer, there are some chunks of about 5 Mbp that are more or less pure Ashkenazi but also some with Ashkenazi colors that stretch over 10 Mbp but also contain other admixtures.
2) There is a very intriguing chunk on my chromosome 15 of about 12 cM or more that matches only Jews, more than a dozen of them in FTDNA, and they don't seem to be closely related between each other. Gedmatch and FTDNA algorithms predict they should be my 4-5th cousins (or more remote). However this chunk in MY chromosome does not paint Ashkenazi at all, it's rather East- or Central European. If I look at the color it takes in THEIR chromosomes, then it's Ashkenazi or Middle East, depending on the person I check. I am very intrigued.. Could you please explain this case genetically?
Thanks again,
Mike.

Davidski said...

Anything more than a slither might be real Ashkenazi admixture on a chromosome painting. It depends whether the result can be backed up with those from other tools.

Also, yes, a segment inherited from an Ashkenazi ancestor might well be Eastern or Central European, because most Ashkenazi Jews do have minor admixture like that.

Another thing to keep in mind is that the chromosome painting doesn't split the chromosomes, so the results might be skewed by very strong signals from one chromosome. In other words, even though you might share a segment with someone, that doesn't necessarily mean that you're looking at that segment when you check out their chromosome painting. You might be looking at a very strong signal from the chromosome on which you don't share anything with that person.

Mike Shake said...

If I get you right, it's exactly the point I was curious about:
The sequencing data cannot tell from which one of the homologous chromosomes the nucleotide comes, and I can reasonably share only one of the two homologues with a given match (sharing both with a distant match is too improbable), right?
So I guess that's why the segment must be very long to make sure that the match is real: otherwise it may be a coincidence that we share some letters from one homologue and some others from the other homologue, right?

Therefore - if I got right what you just said - I can well be that I share a fragment on one homolog with Ashkenazim but it's the other homologue that shows up in my painting; whereas it's the Ashkenazim homologues that shows up in their painting, right?

Also, does it add any weight to the Ashkenazim origin of that funny fragment on one of the homologues of my 15th chromosome that all those who match it are Jews? If we shared it through some slavic admixture, it shouldn't appear in so many Jews and should also come up in some non Jews, right?


Davidski said...

John at GEDmatch can probably explain in more detail how the algorithm he uses interprets the data from both chromosomes. But I'd say that on some occasions it does appear as if the signal from one chromosome drowns out the signal from the other.

And yes, it's very possible that the Ashkenazi segment you inherited is ultimately of Slavic origin. In most cases, Ashkenazi segments tend to look Ashkenazi or East Med on the Jtest, but sometimes they do come through as Eastern Euro or Baltic.

Mike Shake said...

It's getting more interesting. Here is my feedback on your new Eurogenes K15 calculator. That's what I get on gedmatch with Eurogenes EUtest V2 K15 4-Ancestors Oracle:

# Population Percent
1 Baltic 25.14
2 Eastern_Euro 19.59
3 North_Sea 18.09
4 Atlantic 12.73
5 West_Asian 6.82
6 East_Med 6.52
7 West_Med 6.46
8 South_Asian 1.93
9 Siberian 1.28
10 Sub-Saharan 0.8
11 Red_Sea 0.63

Single Population Sharing:

# Population (source) Distance
1 Ukrainian_West 6.2
2 Ukrainian 6.33
3 Russian 6.49
4 Polish 6.61
5 Ukrainian_Northeast 6.7
6 Belorussian 9.91
7 Kargopol_Russian 10.29
8 Hungarian 10.36
9 Erzya 11.63
10 Austrian 12.58
11 Lithuanian 13.02
12 East_Finnish 13.25
13 Estonian 13.34
14 Serbian 13.45
15 Finnish 14.29
16 Romanian 15.26
17 Southwest_Finnish 15.35
18 Bulgarian 17.26
19 North_Swedish 20.87
20 German 21.66

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 88.4% Russian + 11.6% Assyrian @ 2.32
2 88% Russian + 12% Lebanese_Muslim @ 2.44
3 88.6% Russian + 11.4% Armenian @ 2.46
4 89.1% Russian + 10.9% Lebanese_Druze @ 2.47
5 87.8% Russian + 12.2% Cyprian @ 2.53
6 88% Russian + 12% Syrian @ 2.54
7 89.2% Russian + 10.8% Lebanese_Christian @ 2.55
8 83.3% Russian + 16.7% Ashkenazi @ 2.56
9 86.6% Russian + 13.4% Turkish @ 2.59
10 89.1% Russian + 10.9% Samaritan @ 2.73
11 88.1% Russian + 11.9% Kurdish @ 2.74
12 88.1% Russian + 11.9% Iranian @ 2.78
13 88.7% Russian + 11.3% Jordanian @ 2.87
14 86.9% Russian + 13.1% Sephardic_Jewish @ 2.93
15 89.1% Russian + 10.9% Palestinian @ 2.93
16 72.2% Russian + 27.8% Romanian @ 3
17 83.3% Ukrainian_Northeast + 16.7% Ashkenazi @ 3.08

It seems to make sense as I think I am about 75% Slavic + 12.5% Armenian + some 12.5% Jewish.

Can it be that this approximate 25% of non Slavic makes Oracle chose between Armenian and Ashkenaze(or Sephardic) but it cannot show a mixture of these two?

Funny enough though, if I use Oracle-4, totally impossible populations come up (like Erzya + Estonian + Greek) along with Russian, but they are just off given my history.

Thanks for your work and answers again.

Davidski said...

If your Slavic (I assume Russian) ancestors were from near Mordovia then it's likely they were very similar to Finno-Ugrian populations, like Erzya and Moksha. There are only two Russian reference groups in the K15; one is from Kargopol and the other a general western Russian population. So Russian users from near the Volga will probably see Erzya in their results, and those from southern Russia are likely to see Ukrainian.

Mike Shake said...

I see your point. My slavic origins are both from (mother) Ukraine and mixed Russia or from (father) Southern Russia (like Cossacks from Kuban). Volga region is not excluded but I don't know anyone from there in my history... If something is there, it would explain Hungarian, Estonian and Erzya coming up depending on the test, then, right?....

The closest distribution (according to my parents' recent update) should be something like:
Southern Russian 50% Other Russian 12.5% Ukrainian 12.5% Armenian (+Ossetian) about 12.5% and Jewish 12.5%.

What's not clear is how Jewish my "Ukrainian" part is, as they might have been Ashkenazi, too.

In the end, I hope I am using your calculators correctly to affirm that what my parents think we are is totally plausible :)



Davidski said...

I'd say your fairly high Eastern Euro score is making you look partly Erzya in the oracles. But that's probably from your Kuban Cossack ancestry (ie. the further east you go in European Russia, and even within Ukraine, the higher the Eastern Euro component). Also, yeah, I've tested some Russian/Ukrainian Ashkenazi Jews who were far from 100% Jewish.

Aloha79 said...

Help! From my results I gathered that I am 1/2 English/Scottish/Welsh, 1/8 Lithuania, 1/8 Ashkenazi 1/8 Hungarian 1/8 German/French but its hard to tell...
Admix Results (sorted):

# Population Percent
1 ATLANTIC 17.38
2 SOUTH_BALTIC 16.47
3 NORTH-CENTRAL_EURO 15.53
4 WEST_MED 12.08
5 ASHKENAZI 11.68
6 EAST_EURO 11.65
7 WEST_ASIAN 8.23
8 EAST_MED 5.11
9 MIDDLE_EASTERN 1.89


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

Least-squares method.

Using 1 population approximation:
1 AT @ 10.189
2 HU @ 10.485
3 Serbian @ 13.176
4 RO @ 13.693
5 West_&_Central_German @ 16.655
6 FR @ 17.330
7 NL @ 17.781
8 Cornish @ 19.288
9 North_Italian @ 19.354
10 English @ 19.868
78 iterations.



Using 2 populations approximation:
1 50% North_Italian +50% West_Russian @ 9.549
2 50% AT +50% HU @ 9.764
3 50% Tuscan +50% West_Russian @ 9.898
4 50% North_Italian +50% Ukrainian-Russian @ 9.981
5 50% AT +50% RO @ 10.038
6 50% North_Italian +50% PL @ 10.101
7 50% AT +50% AT @ 10.189
8 50% PT +50% West_Russian @ 10.270
9 50% PT +50% Ukrainian-Russian @ 10.288
10 50% ES +50% Ukrainian-Russian @ 10.415
3081 iterations.



Using 3 populations approximation:
1 50% HU +25% AJ +25% IE @ 5.813
2 50% HU +25% AJ +25% Orcadian @ 5.900
3 50% HU +25% AJ +25% Scottish @ 5.905
4 50% West_Russian +25% AJ +25% French_Basque @ 5.919
5 50% Ukrainian-Russian +25% AJ +25% French_Basque @ 6.025
6 50% PL +25% AJ +25% French_Basque @ 6.159
7 50% UA +25% AJ +25% French_Basque @ 6.192
8 50% HU +25% AJ +25% Cornish @ 6.251
9 50% HU +25% AJ +25% NO @ 6.349
10 50% HU +25% AJ +25% South_&_Central_Swedish @ 6.448
104215 iterations.



Using 4 populations approximation:
1 AJ + HU + IE + West_Russian @ 5.592
2 AJ + Cornish + HU + West_Russian @ 5.698
3 AJ + HU + Scottish + West_Russian @ 5.728
4 AJ + HU + Orcadian + West_Russian @ 5.770
5 AJ + HU + IE + PL @ 5.776
6 AJ + HU + HU + IE @ 5.813
7 AJ + AT + IE + West_Russian @ 5.819
8 AJ + HU + IE + UA @ 5.851
9 AJ + HU + IE + Ukrainian-Russian @ 5.879
10 AJ + AT + Cornish + West_Russian @ 5.898
11 AJ + HU + HU + Orcadian @ 5.900
12 AJ + HU + HU + Scottish @ 5.905
13 AJ + French_Basque + Ukrainian-Russian + West_Russian @ 5.917
14 AJ + French_Basque + West_Russian + West_Russian @ 5.919
15 AJ + French_Basque + PL + Ukrainian-Russian @ 5.926
16 AJ + French_Basque + PL + West_Russian @ 5.929
17 AJ + French_Basque + UA + West_Russian @ 5.932
18 AJ + HU + PL + Scottish @ 5.944
19 AJ + HU + Scottish + UA @ 5.949
20 AJ + AT + Orcadian + West_Russian @ 5.952

295573 iterations.

Brian L said...

Quick question....When looking at these results and seeing the @ (number) is the lower the number the better the chance that that particular admixture is more likely the case? Also are these test assuming closer or more distant ancestry?


Admix Results (sorted):

# Population Percent
1 ATLANTIC 27.57
2 NORTH-CENTRAL_EURO 23.52
3 WEST_MED 14.04
4 SOUTH_BALTIC 13.07
5 EAST_EURO 10.44
6 EAST_MED 5.34
7 ASHKENAZI 3.91
8 WEST_ASIAN 2.11


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

Least-squares method.

Using 1 population approximation:
1 Cornish @ 4.845
2 NL @ 6.460
3 West_&_Central_German @ 6.684
4 FR @ 7.254
5 English @ 7.260
6 Orcadian @ 7.993
7 IE @ 8.508
8 Scottish @ 9.334
9 DK @ 10.253
10 AT @ 11.064
78 iterations.



Using 2 populations approximation:
1 50% FR +50% Orcadian @ 2.929
2 50% Cornish +50% FR @ 3.077
3 50% FR +50% Scottish @ 3.301
4 50% FR +50% IE @ 3.373
5 50% FR +50% NL @ 3.597
6 50% FR +50% West_&_Central_German @ 3.818
7 50% DK +50% FR @ 4.003
8 50% English +50% FR @ 4.044
9 50% ES +50% South_&_Central_Swedish @ 4.066
10 50% ES +50% NO @ 4.168
3081 iterations.



Using 3 populations approximation:
1 50% West_&_Central_German +25% AT +25% French_Basque @ 1.650
2 50% NL +25% AT +25% French_Basque @ 1.673
3 50% English +25% French_Basque +25% HU @ 1.827
4 50% DK +25% French_Basque +25% HU @ 2.153
5 50% DK +25% AT +25% French_Basque @ 2.198
6 50% DK +25% French_Basque +25% Serbian @ 2.252
7 50% NL +25% French_Basque +25% HU @ 2.279
8 50% AT +25% DK +25% French_Basque @ 2.308
9 50% Scottish +25% AT +25% ES @ 2.345
10 50% Cornish +25% AT +25% FR @ 2.356
57633 iterations.



Using 4 populations approximation:
1 AT + DK + French_Basque + West_&_Central_German @ 1.164
2 AT + French_Basque + NO + West_&_Central_German @ 1.419
3 AT + French_Basque + South_&_Central_Swedish + West_&_Central_German @ 1.513
4 AT + DK + French_Basque + NL @ 1.525
5 DK + English + French_Basque + HU @ 1.548
6 AT + French_Basque + NL + West_&_Central_German @ 1.564
7 AT + French_Basque + West_&_Central_German + West_&_Central_German @ 1.650
8 AT + French_Basque + NL + NL @ 1.673
9 AT + English + French_Basque + West_&_Central_German @ 1.678
10 AT + French_Basque + NL + South_&_Central_Swedish @ 1.691
11 AT + French_Basque + NL + NO @ 1.696
12 AT + English + French_Basque + South_&_Central_Swedish @ 1.784
13 English + English + French_Basque + HU @ 1.827
14 DK + French_Basque + HU + NL @ 1.851
15 DK + French_Basque + HU + West_&_Central_German @ 1.892
16 AT + English + French_Basque + NO @ 1.910
17 English + French_Basque + HU + NL @ 1.920
18 AT + English + French_Basque + NL @ 1.923
19 AT + DK + English + French_Basque @ 1.924
20 English + French_Basque + Serbian + South_&_Central_Swedish @ 1.929

160964 iterations.

Catchwords said...

Hi David,

I took the FTDNA Family Finder test in hopes of determining whether I have any Sephardic or North African ancestry. (I suspect that I am 10% or less non-Ashkenazi.) My results through Family Finder: 100% Jewish (Middle Eastern). I was rather disappointed at the lack of specificity. I ran the JTest and EUTest and got the results below, but I'm not sure whether they really tell me anything. I know I may be barking up the wrong tree because I'm using the tests for a purpose other than the one you designed them for, but I'm wondering whether the Sephardic and other non-Ashkenazi results are of any real significance.

JTest
1 ASHKENAZI 29.63
2 EAST_MED 17.01
3 WEST_MED 12.34
4 MIDDLE_EASTERN 11.65
5 WEST_ASIAN 7.70
6 ATLANTIC 6.45
7 SOUTH_BALTIC 5.93
8 NORTHCENTRAL_
EURO 4.84
9 EAST_EURO 3.10
10 SIBERIAN 1.33
--------------------------------
Least-squares method.
Using 1 population approximation:
1 AJ @ 5.425
2 South_Italian_&_Sicilian @ 26.067
3 GR @ 28.328
4 Tuscan @ 29.952
5 RO @ 31.893
6 North_Italian @ 31.899
7 Serbian @ 34.338
8 Moroccan @ 34.730
9 Algerian @ 35.148
10 PT @ 35.717
Using 4 populations approximation:
1 AJ + AJ + AJ + AJ @ 5.425
2 AJ + AJ + AJ + Moroccan @ 8.357
3 AJ + AJ + AJ + Algerian @ 8.834
4 AJ + AJ + AJ + South_Italian_&_Sicilian @ 9.008
5 AJ + AJ + AJ + Mozabite_Berber @ 9.420
6 AJ + AJ + AJ + GR @ 9.700
7 AJ + AJ + AJ + North_Italian @ 9.837
8 AJ + AJ + AJ + Tuscan @ 9.883
9 AJ + AJ + AJ + RO @ 9.912
10 AJ + AJ + AJ + PT @ 10.464
11 AJ + AJ + AJ + Serbian @ 10.601
12 AJ + AJ + AJ + ES @ 10.774
13 AJ + AJ + AJ + HU @ 10.933
14 AJ + AJ + AJ + TR @ 11.229
15 AJ + AJ + AJ + IQ @ 11.250
16 AJ + AJ + AJ + AT @ 11.262
17 AJ + AJ + AJ + FR @ 11.525
18 AJ + AJ + AJ + Samaritan @ 11.836
19 AJ + AJ + AJ + Mandean @ 12.147
20 AJ + AJ + AJ + Armenian @ 12.185


EUTest
1 East_Med 25.18
2 West_Med 15.43
3 Atlantic 14.92
4 West_Asian 14.62
5 Red_Sea 12.00
6 Baltic 6.47
7 Eastern_Euro 5.79
8 North_Sea 4.45
9 Siberian 1.14
Using 1 population approximation:
1 Ashkenazi @ 8.558
2 Sephardic_Jewish @ 11.151
3 Greek @ 12.042
4 Tuscan @ 15.669
5 Cyprian @ 18.936
6 Bulgarian @ 19.239
7 Turkish @ 20.818
8 Syrian @ 20.954
9 Lebanese_Muslim @ 21.216
10 Romanian @ 21.793
165 iterations.
1 Greek + Saudi + Southwest_French + Turkish @ 5.117
2 Greek + Saudi + Spanish_Andalucia + Turkish @ 5.427
3 Greek + Saudi + Spanish_Aragon + Turkish @ 5.532
4 Greek + Saudi + Spanish_Castilla_La_Mancha + Turkish @ 5.582
5 Greek + Saudi + Spanish_Cantabria + Turkish @ 5.627
6 Austrian + Kurdish + Sardinian + Saudi @ 5.650
7 Armenian + French_Basque + Greek + Saudi @ 5.657
8 Greek + Saudi + Spanish_Valencia + Turkish @ 5.673
9 French_Basque + Greek + Saudi + Turkish @ 5.777
10 Armenian + Greek + Saudi + Southwest_French @ 5.813
11 Bulgarian + Saudi + Spanish_Andalucia + Turkish @ 5.819
12 Assyrian + French_Basque + Greek + Saudi @ 5.929
13 French_Basque + Greek + Kurdish + Saudi @ 5.940
14 Assyrian + Bulgarian + French_Basque + Saudi @ 5.963
15 Armenian + Bulgarian + French_Basque + Saudi @ 5.996
16 Austrian + Sardinian + Saudi + Turkish @ 6.002
17 Greek + Kurdish + Saudi + Southwest_French @ 6.023
18 Austrian + Iranian + Sardinian + Saudi @ 6.067
19 Armenian + Austrian + Sardinian + Saudi @ 6.093
20 Armenian + Bulgarian + Saudi + Southwest_French @ 6.182

Davidski said...

Yes, it seems you do have a bit of North African ancestry, but it's less than 10%.

2 AJ + AJ + AJ + Moroccan @ 8.357
3 AJ + AJ + AJ + Algerian @ 8.834
5 AJ + AJ + AJ + Mozabite_Berber @ 9.420

Perhaps part Sephardic or Moroccan Jewish? It seems fairly recent, so you should be able to catch it in some of the genealogical tools.

Catchwords said...

Thanks, David. I think I know which line it is (based *solely* on the appearance of one of my maternal great-grandfathers). I imagine a Y-DNA test would be the best tool, but I am female and my mother had no brothers. Anyhow, thank you for providing the tests. These are the most helpful instruments I've found so far for researching the specifics of my family's ethnic background.

Steven said...

Is there any difference between Ashkenazi and Eastern Med? Isn't Israel an Eastern Mediterranean country?

Aloha79 said...

David,

I posted my results on October 20th and I am still having trouble understanding the %. I recently discovered that I had a great-grandmother from Lithuania. I have always been under the impression that the rest of my ancestry was from England. I am really confused with the South Baltic, Eastern Europe, and Ashkenazi percentages. Are they intertwined? 23andme has me at the conservative for Ashkenazi at 25%, does this seem accurate?

Thanks

Davidski said...

It depends what you mean by intertwined. Ashkenazi Jews from Eastern Europe aren't Eastern Europeans genetically, even though they might have some Eastern European admixture. So it's likely you get most of your Baltic and Eastern Euro from your English side, due to shared ancient ancestry across Northern Europe.

And you're definitely around 25% Ashkenazi, because that's what both 23andMe and the Oracles say. So your Lithuanian great-grandparent wasn't ethnically Lithuanian, but Jewish. In fact, it seems you have two Jewish great-grandparents.

Aloha79 said...

I think I figured it out. I did the Eurotest, I know 4 of my great-grandparent were from England, I know one was from Sweden,I think the Lithuanian great-grandparent is Lithuanian, and the last 2 were probably Jewish. Using 4 populations approximation:
1 Armenian + French_Basque + LIT + North_Swedish @ 3.964
2 Armenian + French_Basque + LIT + South_&_Central_Swedish @ 4.160
3 Armenian + French_Basque + LIT + South_Finnish @ 4.240
4 Armenian + French_Basque + LIT + NO @ 4.291
5 French_Basque + LIT + South_&_Central_Swedish + TR @ 4.354
6 ES + ES + Lezgin + LIT @ 4.373
7 French_Basque + Northwest_Russian + South_&_Central_Swedish + TR @ 4.446
8 Armenian + French_Basque + North_Swedish + Northwest_Russian @ 4.454
9 French_Basque + LIT + NO + TR @ 4.460
10 Armenian + French_Basque + Northwest_Russian + South_&_Central_Swedish @ 4.477
11 French_Basque + LIT + North_Swedish + TR @ 4.479
12 Armenian + EE + EE + French_Basque @ 4.501
13 ES + FR + Lezgin + LIT @ 4.516
14 Armenian + EE + French_Basque + PL @ 4.534
15 French_Basque + PL + PL + TR @ 4.550
16 EE + French_Basque + TR + UA @ 4.559
17 EE + French_Basque + PL + TR @ 4.561
18 ES + IE + LIT + TR @ 4.576
19 Armenian + EE + French_Basque + UA @ 4.632
20 French_Basque + NO + Northwest_Russian + TR @ 4.637

Davidski said...

Try the new K13...

http://bga101.blogspot.com.au/2013/11/updated-eurogenes-k13-at-gedmatch.html

Aloha79 said...

Here are my results: No Lithuanian, interesting.... 1 Austrian + Greek + Southeast_English + Ukrainian_West @ 2.149
2 Austrian + Greek + Southwest_English + Ukrainian_West @ 2.194
3 Austrian + Greek + Southwest_English + Ukrainian @ 2.323
4 Greek + Irish + Polish + Serbian @ 2.335
5 Greek + Hungarian + Southwest_English + Ukrainian_West @ 2.335
6 French + German + Greek + Ukrainian_Northeast @ 2.364
7 Greek + Hungarian + Southeast_English + Ukrainian_West @ 2.372
8 Greek + Polish + Serbian + West_Scottish @ 2.385
9 Austrian + Greek + Southeast_English + Ukrainian @ 2.387
10 Austrian + Danish + Greek + Ukrainian_West @ 2.389
11 Greek + Serbian + Ukrainian + West_Scottish @ 2.400
12 Greek + Irish + Serbian + Ukrainian @ 2.446
13 Austrian + Greek + Orcadian + Ukrainian_West @ 2.447
14 Irish + Romanian + Tuscan + Ukrainian_Northeast @ 2.453
15 German + North_Italian + Romanian + Ukrainian_Northeast @ 2.457
16 Danish + Greek + Polish + Serbian @ 2.460
17 Austrian + Dutch + Romanian + Romanian @ 2.466
18 Austrian + Greek + Irish + Ukrainian_West @ 2.467
19 Greek + Hungarian + Southwest_English + Ukrainian @ 2.492
20 Austrian + Dutch + Greek + Ukrainian_West @ 2.493

JoeyGee1000 said...

Hi. I have a question regarding possible AJ ancestry. 23 and Me told me I'm 0.5% Ashkenazi, which I dismissed as either noise or not enough to count. This Jtest, however, gives me about 6.1% AJ, but none of the Oracle calculators speculate any. Now, my mother is of Calabrese Italian ancestry, and my father English/German (probably Irish and Scottish in there somewhere) and I don't know of any Jewish ancestry. The Oracle calculators for this test generally say 1/2 Greek, 1/2 English which is essentially correct. Is it likely the +/-6% AJ is just Southern Euro?

1 EAST_MED 18.65
2 NORTH-CENTRAL_EURO 17.85
3 ATLANTIC 16.32
4 WEST_MED 14.79
5 SOUTH_BALTIC 9.48
6 WEST_ASIAN 7.94
7 ASHKENAZI 6.06
8 EAST_EURO 5.09
9 MIDDLE_EASTERN 3.03
10 SOUTH_ASIAN 0.67
11 EAST_AFRICAN 0.11

Single Population Sharing:

# Population (source) Distance
1 Tuscan 7.65
2 North_Italian 7.71
3 RO 11.44
4 Serbian 11.55
5 PT 13.02
6 FR 14.34
7 AT 14.49
8 South_Italian_&_Sicilian 14.91
9 ES 15.17
10 GR 16.54
11 HU 16.56
12 West_&_Central_German 18.09
13 NL 19.37
14 Cornish 20.5
15 English 20.53
16 Orcadian 22.48
17 DK 22.75
18 IE 22.82
19 Scottish 23.77
20 NO 24.2

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 55.4% GR + 44.6% English @ 1.81
2 54% GR + 46% NL @ 2.22
3 52.3% GR + 47.7% West_&_Central_German @ 2.58
4 55.5% GR + 44.5% Cornish @ 2.61
5 55% South_Italian_&_Sicilian 45% West_&_Central_German @ 2.81
6 58.1% GR + 41.9% DK @ 2.82
7 61.2% Tuscan + 38.8% Serbian @ 2.89
8 56.7% South_Italian_&_Sicilian + 43.3% NL @ 2.96
9 82.7% Tuscan + 17.3% EE @ 3.11
10 57.8% GR + 42.2% Orcadian @ 3.12
11 58.2% South_Italian_&_Sicilian + 41.8% English @ 3.24
12 80% Tuscan + 20% PL @ 3.33
13 67.3% Tuscan + 32.7% AT @ 3.38
14 82.7% Tuscan + 17.3% East_Finnish @ 3.4
15 82.4% Tuscan + 17.6% Belorussian @ 3.41
16 81.5% Tuscan + 18.5% South_Finnish @ 3.43
17 81.3% Tuscan + 18.7% West_Russian @ 3.44
18 60.8% South_Italian_&_Sicilian + 39.2% DK @ 3.5
19 59.7% GR + 40.3% NO @ 3.5
20 58.3% GR + 41.7% IE @ 3.51

Davidski said...

Scoring some Ashkenazi in the overall Ad-mix results doesn't mean you have Jewish ancestry. It's all explained in the blog entry above.

JoeyGee1000 said...
This comment has been removed by the author.
Mr West said...

SOUTH_BALTIC 12.14%
EAST_EURO 9.97%
NORTH-CENTRAL_EURO 34.01%
ATLANTIC 23.70%
WEST_MED 9.85%
ASHKENAZI 3.03%
EAST_MED 0.10%
WEST_ASIAN 6.25%
MIDDLE_EASTERN 0.52%
SOUTH_ASIAN 0.30%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN 0.14%
WEST_AFRICAN -


Hi, does this mean I either have very distant Ashkenazi ancestry, or none at all? considering it's 3.03%? If very distant, how far back are we talking?

Thanks

Jordan Christopher said...

Hello,

Going off my family tree, I previously estimated that I was about half German, 1/5 Irish, 1/10 Welsh, 1/10 English and the rest being a melange of French, Ulster Scots and possibly Italian with more distant Dutch and Swiss ancestry. My family tree very filled out, so there wasn't much of a surprise to my results. The only surprise was the admixture percentages. My ethnicity estimate on AncestryDNA is the following:

Great Britain 43%
Europe West 23%
Ireland 22%
Iberian Peninsula 4%
Finland/Northwest Russia 3%
Scandinavia 3%
Italy/Greece 1%
Caucasus <1%

At first I was surprised to see the trace regions, but figured that those were most likely distant ancestry like the Iberian being from Irish/Welsh ancestry and the Scandinavian being from English/Irish ancestors from Viking colonization.

It's to my understanding that British/Continental European genes are commonly mixed up, so I took the Eutest to see if there was a change and got these results.

Admix Results (sorted):

# Population Percent
1 NORTH-CENTRAL_EURO 28.76
2 ATLANTIC 27.58
3 WEST_MED 12.75
4 SOUTH_BALTIC 11.93
5 EAST_EURO 11.28
6 WEST_ASIAN 4.54
7 EAST_MED 2.99
8 EAST_AFRICAN 0.16

Single Population Sharing:

# Population (source) Distance
1 English 2.08
2 Cornish 2.8
3 NL 3.29
4 Orcadian 3.75
5 IE 4.15
6 West_&_Central_German 4.9
7 DK 5.32
8 Scottish 5.74
9 NO 7.3
10 South_&_Central_Swedish 8.41
11 FR 10.17
12 North_Swedish 12.37
13 AT 12.53
14 PT 17.74
15 HU 17.94
16 ES 17.96
17 Serbian 20.02
18 South_Finnish 21.18
19 North_Italian 21.57
20 RO 22.97

Using 4 populations approximation:
1 English + English + English + IE @ 1.796
2 Cornish + English + English + IE @ 1.839
3 English + English + IE + NL @ 1.908
4 Cornish + English + English + English @ 1.913
5 Cornish + Cornish + English + English @ 1.944
6 Cornish + English + English + Orcadian @ 1.977
7 English + FR + IE + NO @ 1.985
8 Cornish + English + IE + NL @ 1.992
9 English + English + English + Orcadian @ 2.054
10 Cornish + English + English + NL @ 2.073
11 English + FR + NO + Scottish @ 2.119
12 DK + English + FR + IE @ 2.126
13 Cornish + Cornish + English + NL @ 2.143
14 Cornish + English + NL + Orcadian @ 2.164
15 English + English + NL + Orcadian @ 2.196
16 Cornish + Cornish + DK + English @ 2.241
17 English + FR + NO + Orcadian @ 2.247
18 English + English + NL + Scottish @ 2.270
19 English + IE + NL + NL @ 2.273
20 English + English + English + Scottish @ 2.273

These approximations are nothing close to what I should be finding based off my family tree...

From the both data sets, English (and Great Britain in general) was dominant. doesn't it seem like my English ancestry which should only be 10-15% is overrepresented? My theories are that either the German is being picked up in the English percentages or that just by chance I ended up with a ton of English genes. Let me know what you think.

Thanks!

Meredith said...

I'm sure this is fascinating, but I HAVE NO EARTHLY IDEA WHAT I'M LOOKING AT when viewing Admixture or anything else. Isn't there a way to put all of this in layman's terms? I don't see anything clearly explained.

Guy Jacks said...

Does eastern Mediterranean coastal areas include Cyprus, Greece, Crete etc or just southwest Asia?

Unknown said...
This comment has been removed by the author.
Unknown said...

Hello, I had the Family Finder test, and the results were 50% Western European (French) and 50% Jewish (I am Bulgarian, and we are not aware of any Jewish or foreign ancestry on either side). Here are the results from your Jtest -- could you please help with an interpretation/opinion? Many thanks, Tzenka

Admix Results (sorted):

# Population Percent
1 EAST_MED 17.47
2 NORTH-CENTRAL_EURO 16.96
3 WEST_MED 14.38
4 SOUTH_BALTIC 13.31
5 ATLANTIC 10.97
6 EAST_EURO 10.04
7 WEST_ASIAN 7.81
8 ASHKENAZI 4.67
9 MIDDLE_EASTERN 4.2
10 SOUTH_ASIAN 0.19

Single Population Sharing:

# Population (source) Distance
1 Serbian 6.96
2 RO 7.66
3 North_Italian 12.37
4 HU 12.89
5 AT 13.14
6 Tuscan 13.14
7 South_Italian_&_Sicilian 17.01
8 PT 17.12
9 FR 17.57
10 GR 17.69
11 West_&_Central_German 18.77
12 ES 19.44
13 NL 20.24
14 English 22.07
15 Cornish 22.38
16 DK 23.15
17 UA 23.16
18 PL 23.36
19 South_&_Central_Swedish 23.66
20 NO 23.87

Mixed Mode Population Sharing:

# Primary Population (source) Secondary Population (source) Distance
1 73.6% Serbian + 26.4% South_Italian_&_Sicilian @ 3.6
2 68.6% Serbian + 31.4% Tuscan @ 3.95
3 75.5% Serbian + 24.5% GR @ 4.17
4 57.6% HU + 42.4% South_Italian_&_Sicilian @ 4.55
5 66% Tuscan + 34% Ukrainian-Russian @ 4.66
6 58.3% South_Italian_&_Sicilian + 41.7% PL @ 4.66
7 58.2% AT + 41.8% GR @ 4.68
8 58.1% South_Italian_&_Sicilian + 41.9% UA @ 4.7
9 58.8% GR + 41.2% North_Swedish @ 4.71
10 64% South_Italian_&_Sicilian + 36% EE @ 4.72
11 58.8% HU + 41.2% GR @ 4.85
12 88.1% Serbian + 11.9% Samaritan @ 4.85
13 60.9% GR + 39.1% South_Finnish @ 4.91
14 62.2% South_Italian_&_Sicilian + 37.8% Belorussian @ 4.92
15 61.9% South_Italian_&_Sicilian + 38.1% South_Finnish @ 4.93
16 57.7% GR + 42.3% South_&_Central_Swedish @ 4.99
17 65.1% Tuscan + 34.9% UA @ 5.09
18 68.9% Tuscan + 31.1% Belorussian @ 5.21
19 59.4% South_Italian_&_Sicilian + 40.6% Ukrainian-Russian @ 5.21
20 57.3% AT + 42.7% South_Italian_&_Sicilian @ 5.23

K Dub said...

My ancestry is French, Scottish, Irish, English and German. Does this seem right to you?

Eu test

Population
SOUTH_BALTIC 16.83%
EAST_EURO -
NORTH-CENTRAL_EURO 23.78%
ATLANTIC 31.26%
WEST_MED 16.72%
EAST_MED 6.57%
WEST_ASIAN 2.81%
MIDDLE_EASTERN 1.03%
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN 0.99%

JTest

Population
SOUTH_BALTIC 16.60%
EAST_EURO -
NORTH-CENTRAL_EURO 23.29%
ATLANTIC 30.62%
WEST_MED 16.14%
ASHKENAZI 4.96%
EAST_MED 5.23%
WEST_ASIAN 2.05%
MIDDLE_EASTERN 0.15%
SOUTH_ASIAN -
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN 0.96%

Daniel Tinney said...

@ Davidski

I am still waiting on my 23andme results, but I did run my Ancestry raw data. According to your blog above, I think its picking up on my rumored Ashkenazi ancestry. AJ showed up in my top 20 mixed mode Oracle results, and it was on both options of Oracle X mode. I should be about 1/16 Ashkenazi, and I believe this is what the test is indicating. I think the test worked out good for my 1/16 results . I looked at the difference between my Jtest an Eutest and most of my Ashkenazi was hiding in East Med. This sounds constant with the 1/ 16 rumor to me. Please tell me what you think, and thank you for the test.


Pct. Calc. Option 2


0 Unable to determine 0.02%
1 NL 28.48%
2 Cornish 27.68%
3 English 15.56%
4 South_&_Central_Swedish 7.32%
5 DK 4.86%
6 Balochi 4.57%
7 NO 4.46%
8 AJ 4.32%
9 East_Finnish 2.73%
10 Burusho 0.01%


Pct. Calc. Option 1


1 NL 39.94%
2 English 32.90%
3 DK 17.13%
4 AJ 5.10%
5 Kalash 2.92%
6 AT 1.21%
7 RO 0.39%
8 Balochi 0.37%
9 Bangladeshi 0.04%
10 Serbian 0.00%



# Primary Population (source) Secondary Population (source) Distance
1 94.3% NL + 5.7% AJ @ 1.76
2 96% NL + 4% Balochi @ 1.77
3 96% NL + 4% Brahui @ 1.78
4 95.6% NL + 4.4% IR @ 1.92
5 96.3% NL + 3.7% Burusho @ 1.93
6 96.2% NL + 3.8% Kalash @ 1.93
7 96.6% NL + 3.4% Sindhi @ 1.94
8 95.8% NL + 4.2% Kurdish @ 1.96
9 96.4% NL + 3.6% Mandean @ 2.08
10 97.3% NL + 2.7% IN @ 2.09
11 95.6% NL + 4.4% TR @ 2.1
12 97.4% NL + 2.6% Gujarati @ 2.11
13 96.3% NL + 3.7% Armenian @ 2.12
14 96.4% NL + 3.6% Assyrian @ 2.14
15 97.4% NL + 2.6% Bangladeshi @ 2.15
16 95.9% NL + 4.1% Lezgin @ 2.15
17 96.7% NL + 3.3% GE @ 2.17
18 96.4% NL + 3.6% IQ @ 2.17
19 93.6% English + 6.4% Balochi @ 2.21
20 93.6% English + 6.4% Brahui @ 2.21


JTest

Population
SOUTH_BALTIC 12.47%
EAST_EURO 11.67%
NORTH-CENTRAL_EURO 26.81%
ATLANTIC 23.65%
WEST_MED 10.34%
ASHKENAZI 4.67%
EAST_MED 4.21%
WEST_ASIAN 4.43%
MIDDLE_EASTERN -
SOUTH_ASIAN 1.74%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -

EuTest

Population
SOUTH_BALTIC 12.55%
EAST_EURO 11.94%
NORTH-CENTRAL_EURO 27.13%
ATLANTIC 24.34%
WEST_MED 11.05%
EAST_MED 5.89%
WEST_ASIAN 5.24%
MIDDLE_EASTERN -
SOUTH_ASIAN 1.85%
EAST_AFRICAN -
EAST_ASIAN -
SIBERIAN -
WEST_AFRICAN -

Crixus said...

My 4 grandparents are born in the Middle-East (Turkey), but I know our ancestors left Iran 400 years ago and settled in Turkey. I know that less than 1% is noice and that I have to ignore that. But how about Atlantic and West-Mediterrean? My paternal Hablogroup is R1b1b12a* and my maternal line shows J1b3.

More interesting, can I consider myself as a descendant of Askhenazi people? See below my result with Jtest.

Population
SOUTH_BALTIC 0.56%
EAST_EURO 4.31%
NORTH-CENTRAL_EURO 3.01%
ATLANTIC 2.10%
WEST_MED 1.86%
ASHKENAZI 4.26%
EAST_MED 38.41%
WEST_ASIAN 29.04%
MIDDLE_EASTERN 7.60%
SOUTH_ASIAN 7.96%
SIBERIAN 0.90%

When using Using 1 population approximation, I have the following results:

1 Kurdish @ 5.879
2 IR @ 7.502
3 Assyrian @ 9.619
4 Mandean @ 11.412
5 Armenian @ 11.643
6 TR @ 12.367
7 IQ @ 20.099
8 Druze @ 24.282
9 GE @ 25.228
10 GR @ 26.604

Suddenly Kurdish DNA shows up at #1??? And do I have Greek DNA as well? Which country is IQ?

Jesse Alvarez said...

it is said that less than 1% result of the admix test is just a noise. is this also true at the chromosome level?