Is 23AndMe Wrong/Mostly Wrong About Corded's Mom's AJ Ancestry?

**~~CordedWhelp~~** · 06-06-2018, 11:01 PM

Can't believe I have to implement this, but: I have no motivation either way for or against that kind of ancestry in me. I just had a thought and wish to explore said thought more. Below are her results from 23andme, Wegene (I know, I know...), Eurogenes K13 and JTEST.

23andme has my mother at 5.5% Ashkenazi Jewish. Her current AC (on 50% spec):

Eastern European: 35.2% (four dots next to Belarus, three next to Russia, three next to Poland, two next to Hungary, one next to Ukraine, one next to Lithuania, one next to Slovakia)

Scandinavian: 9.0% (one dot next to Denmark)

French and German: 6.8% (two dots next to Germany)

Ashkenazi: 5.5%

British and Irish: 3.0% (one dot next to U.K.)

Balkan: 2.3% (one dot next to Croatia)

Italian: 0.4%

Broadly NW Euro: 24.2%

BRoadly South Euro: 3.0%

BRoadly European: 10.4%

East Asian and Native American: 0.2%

Siberian: 0.1%
Broadly: 0.1%

Importing to WEGENE (not great for Euros yet, but just gonna post it all the same):

Finnish/Russian: 50.33%
Spaniards: 29.12%
Balkan: 11.64%
British: 8.32%
Ashkenazi: 0.53% (Again, I know this site is mostly geared for East Asians, but AJ's are genetically a pretty distinct lot. Wouldn't it be at least somewhat higher here?)
Other: 0.02%

EUROGENES K13:
Admix Results (sorted):

# Population Percent
1 Baltic 37.28
2 North_Atlantic 32.84
3 West_Med 11.92
4 East_Med 6.83
5 West_Asian 5.66
6 South_Asian 1.77
7 Red_Sea 1.43
8 Siberian 1.39

Finished reading population data. 204 populations found.
13 components mode.

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

Least-squares method.

Using 1 population approximation:
1 South_Polish @ 5.615799
2 Hungarian @ 6.944128
3 Croatian @ 7.364062
4 East_German @ 7.437738
5 Ukrainian_Lviv @ 7.537293
6 Ukrainian @ 8.367209
7 Polish @ 9.347516
8 Austrian @ 9.387593
9 Moldavian @ 11.180273
10 Southwest_Finnish @ 12.414995
11 Southwest_Russian @ 12.811396
12 Russian_Smolensk @ 12.933526
13 Ukrainian_Belgorod @ 13.416420
14 Estonian_Polish @ 13.507435
15 Belorussian @ 13.618048
16 North_Swedish @ 14.169118
17 Serbian @ 15.586457
18 Finnish @ 15.714602
19 Estonian @ 16.162874
20 Kargopol_Russian @ 16.297646

Using 2 populations approximation:
1 50% East_German +50% South_Polish @ 2.123548

Using 3 populations approximation:
1 50% Estonian_Polish +25% Orcadian +25% Romanian @ 1.106264

Using 4 populations approximation:
++++++++++++++++++++++++++++++++++++++++++++++++++ +++++++++++++++++++++
1 Estonian + Estonian_Polish + Serbian + South_Dutch @ 0.994914
2 French + Hungarian + Lithuanian + Southwest_Russian @ 1.006805
3 Lithuanian + North_Italian + North_Swedish + Ukrainian_Belgorod @ 1.031910
4 East_German + Polish + Serbian + Southwest_Finnish @ 1.041756
5 Estonian + Russian_Smolensk + Serbian + South_Dutch @ 1.048694
6 Estonian_Polish + Hungarian + South_Dutch + Southwest_Russian @ 1.097012
7 Estonian_Polish + Estonian_Polish + Orcadian + Romanian @ 1.106264
8 Estonian + Lithuanian + Romanian + South_Dutch @ 1.110855
9 Estonian + French + Lithuanian + Serbian @ 1.130045
10 Estonian_Polish + Lithuanian + Romanian + Southwest_English @ 1.159342
11 Estonian + Estonian_Polish + Serbian + West_German @ 1.160213
12 Lithuanian + North_Italian + North_Swedish + Southwest_Russian @ 1.169903
13 Estonian_Polish + French + South_Polish + South_Polish @ 1.170097
14 Belorussian + Croatian + South_Dutch + South_Polish @ 1.170770
15 East_German + Russian_Smolensk + Serbian + Southwest_Finnish @ 1.178339
16 Estonian_Polish + Estonian_Polish + Hungarian + South_Dutch @ 1.185554
17 Estonian_Polish + Hungarian + Russian_Smolensk + South_Dutch @ 1.188112
18 Estonian_Polish + Estonian_Polish + Orcadian + Serbian @ 1.195627
19 Croatian + Estonian_Polish + South_Dutch + South_Polish @ 1.197117
20 Estonian_Polish + Lithuanian + Romanian + Southeast_English @ 1.197196

JTEST:

Jtest 4-Ancestors Oracle
This program is based on 4-Ancestors Oracle Version 0.96 by Alexandr Burnashev.
Questions about results should be sent to him at: Alexandr.Burnashev@gmail.com
Original concept proposed by Sergey Kozlov.
Many thanks to Alexandr for helping us get this web version developed.

Jtest Oracle population reference data revised 06 Nov 2012.

Admix Results (sorted):

# Population Percent
1 SOUTH_BALTIC 20.93
2 NORTH-CENTRAL_EURO 20.21
3 EAST_EURO 19.16
4 ATLANTIC 18.91
5 WEST_MED 8.08
6 ASHKENAZI 4.44
7 WEST_ASIAN 3.60
8 EAST_MED 3.04
9 SOUTH_ASIAN 1.26

Finished reading population data. 78 populations found.
14 components mode.

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

Least-squares method.

Using 1 population approximation:
1 AT @ 9.713467
2 HU @ 9.770246
3 PL @ 10.138852
4 South_Finnish @ 11.383050
5 North_Swedish @ 11.439263
6 West_Russian @ 12.068996
7 UA @ 12.193340
8 South_&_Central_Swedish @ 12.816602
9 Ukrainian-Russian @ 13.229326
10 West_&_Central_German @ 13.580915
11 NO @ 13.712624
12 NL @ 14.526819
13 EE @ 14.824071
14 Belorussian @ 14.869775
15 DK @ 15.004644
16 East_Finnish @ 15.252179
17 Serbian @ 15.813924
18 East_Russian @ 15.866424
19 Orcadian @ 16.811895
20 English @ 17.273344

Using 2 populations approximation:
1 50% NL +50% Ukrainian-Russian @ 2.628475

Using 3 populations approximation:
1 50% NL +25% Ukrainian-Russian +25% Ukrainian-Russian @ 2.628475

Using 4 populations approximation:
++++++++++++++++++++++++++++++++
1 AT + DK + West_Russian + West_Russian @ 2.119482
2 AT + IE + Ukrainian-Russian + West_Russian @ 2.184635
3 AT + East_Russian + Orcadian + PL @ 2.191941
4 AT + Scottish + Ukrainian-Russian + Ukrainian-Russian @ 2.197498
5 AT + Cornish + East_Russian + PL @ 2.243753
6 AT + Cornish + North_Russian + PL @ 2.251406
7 AT + Orcadian + Ukrainian-Russian + West_Russian @ 2.251611
8 AT + Scottish + Ukrainian-Russian + West_Russian @ 2.259606
9 AT + IE + Ukrainian-Russian + Ukrainian-Russian @ 2.264094
10 AT + East_Russian + IE + PL @ 2.279253
11 AT + IE + North_Russian + PL @ 2.286020
12 AT + East_Russian + Scottish + Ukrainian-Russian @ 2.308384
13 AT + East_Russian + Scottish + UA @ 2.308877
14 AT + East_Russian + PL + Scottish @ 2.316906
15 AT + North_Russian + Orcadian + PL @ 2.329704
16 Cornish + Ukrainian-Russian + Ukrainian-Russian + West_&_Central_German @ 2.334728
17 AT + DK + Ukrainian-Russian + West_Russian @ 2.358416
18 AT + Cornish + East_Russian + West_Russian @ 2.372596
19 Cornish + NL + Ukrainian-Russian + Ukrainian-Russian @ 2.390525
20 AT + Orcadian + West_Russian + West_Russian @ 2.392798

**Leto** · 06-06-2018, 11:10 PM

Using 2 populations approximation:
1 50% East_German +50% South_Polish @ 2.123548

Basically Czech or at least pretty close to them.

**~~CordedWhelp~~** · 06-06-2018, 11:52 PM

Bump

**~~CordedWhelp~~** · 06-06-2018, 11:52 PM

Originally Posted by Leto

Basically Czech or at least pretty close to them.

It must be because she's half East Slav, half Danish-German mix. A logical average would be West Slavs, I suppose.

**Annie999** · 06-06-2018, 11:59 PM

Ive heard 23andme is the best reader for AJ. Wegene is really bad, I would ignore whatever it says. Why dont you upload the data to MyHeritage?

**Peterski** · 06-07-2018, 12:14 AM

Originally Posted by CordedWhelp

but AJ's are genetically a pretty distinct lot.

In some ways yes, but in other ways - not.

What Davidski wrote about this (he wrote this few years ago, maybe his opinion has changed, IDK):

"(...) 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."

Originally Posted by Annie999

Ive heard 23andme is the best reader for AJ.

Yeah, just like Living DNA is the best reader for British.

The problem is (as Davidski wrote), that a lot of people who don't have AJ ancestry end up scoring it on 23andMe, and a lot of people who don't have British ancestry end up scoring it on Living DNA.

For example I am 1/5 British according to LivingDNA:

https://www.theapricity.com/forum/sh...ingDNA-results

I haven't tried 23andMe yet, but all the Polish 23andMe results that I've seen score some minor AJ. And in other companies (like FTDNA) Polish people usually don't score AJ. Only on 23andMe.

**~~CordedWhelp~~** · 06-07-2018, 01:29 AM

Originally Posted by Peterski

In some ways yes, but in other ways - not.

What Davidski wrote about this (he wrote this few years ago, maybe his opinion has changed, IDK):

"(...) 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."

Yeah, just like Living DNA is the best reader for British.

The problem is (as Davidski wrote), that a lot of people who don't have AJ ancestry end up scoring it on 23andMe, and a lot of people who don't have British ancestry end up scoring it on Living DNA.

For example I am 1/5 British according to LivingDNA:

https://www.theapricity.com/forum/sh...ingDNA-results

I haven't tried 23andMe yet, but all the Polish 23andMe results that I've seen score some minor AJ. And in other companies (like FTDNA) Polish people usually don't score AJ. Only on 23andMe.

How much are these Polish users scoring in the AJ department?

Although I have some logical questions, I still think at the end of the day that my mom's 5.5% is a bit too high to be noise. 23andme seldom seems to show AJ noise beyond like 0.5%. I dunno.

**~~Carlito's Way~~** · 06-07-2018, 01:30 AM

she has jewish ancestry, shes a jew, you are a jew, we all sons of god

**~~CordedWhelp~~** · 06-07-2018, 01:31 AM

Also, what you make of my mom's Balkan score with Croatia being given a dot? One could say it's just because Croats have a good amount of actual Slavic admixture and is reflecting shared ancestry between them and north slavs...although I've seen some of my moms cousin matches on there and they're seldom scoring beyond 1% Balkan. Have you seen Polish results on 23andme scoring high Balkan with specified reference populations?

Edit: That was meant as a reply to Peterski, but I'm a dolt and didn't do it properly.