PDA

View Full Version : My wife, die Grossmutter's 23andMe results



Grossvater
02-04-2013, 10:40 PM
My wife's results are dribbling in a little bit at a time. We got the health section last night and were relieved to find out that she didn't inherit any breast cancer genes from her Converso Jewish Northern Mexican ancestorss. We also found out she is 2.7% Neanderthal which is rather odd to ponder. They must not do a full mitochondrial test as her results came back as just an A2. We already know through FamilyTree DNA she is an A2r.

What really has me puzzled is they are telling us she is 69.7% European, 17.9% East Asian and Native American, 1.0% Subsaharan African and 11.3% unassigned. What is up with "unassigned" section? It seems like an awful lot of genome to just go unassigned. We don't know of any recorded East Asian ancestors so can we assume that all of the 17.9% is Native American? Can anybody shed light on this for me? Are the results just incomplete? They haven't told us any relatives and we haven't been able to find any actual raw data. By the way, this is the Standard Estimate. When I click on the Speculative Estimate, only 4.1% is unassigned...most of the rest of it goes into the Native section and a little bit into the European.

I was also a little surprised at how little Subsaharan African ancestry she carries as I have found around half a dozen "mulato libre" folks in her family tree from 18th Century Mexico. They must have been more "libre" than "mulato." I thought more African ancestry would surface.

Also, when we do get the raw data, what can I do with it to further clarify her ancestry. I have read of a Dr. McDonald and something called Dodecad and Harappa but am not sure what these things actually are. Are they do-it-yourself online DNA readers or something?

AJL
02-05-2013, 12:43 AM
The Speculative view seems most consistent with the paper trail for most people, so you're probably right to stick with that.

"McDonald's BGA" is a reading by Dr. Doug McDonald, a chemistry professor at the U. of Illinois who has developed this means of analyzing autosomes.

The others are projects you could formerly participate in but which have now given way to the form of tests you can run yourself after uploading a genome to Gedmatch.com, or using offline programs like the free statistical analysis program R.

1.0% anything still seems substantial, if not epic. I have my most recent fully Sephardi ancestors in the 19th century, and get Iberian reflecting this (1.8% on Speculative). My guess, given this, is that "1%" corresponds roughly to one full ancestor in the mid-18th century. This also fits closely to the proportion of your ancestors at that stage: 1 out of the 64 of your fourth great-grandparents = 1.5625%. Of course, you won't inherit DNA from every ancestor at this point, so siblings can show fairly different components.

Grossvater
02-05-2013, 03:59 AM
AJ...thanks for the reply. Did I understand you correctly that your Sephardi Jewish ancestry appeared as Iberian? My wife has lots of Converso Jewish ancestry in the paper trail but has absolutely no Middle Eastern or North African in the testing which I found surprising. She does have 16.4% Southern European of which 6.1% is listed as Iberian, 1.4% as Italian and the rest unspecified Southern European. I found the Italian particularly surprising.

AJL
02-05-2013, 04:01 PM
Yes, it appears as Iberian for me though I also have some Italian, which is apparently both from Sephardi, and from my mother's French ancestry, some of which probably traces to the Milan area. In many cases, though, Sephardi may appear as Middle Eastern or North African, or Nonspecific (Southern) European, as well. I suspect there may be a current bug that overreports some ancestries and underreports others, but unless Sephardi is made an opt-in category, Iberian and Italian in combination may suggest Sephardi ancestry, though it could mean just that as well.

Grossvater
02-24-2013, 08:19 PM
I sent Dr. McDonald my wife's raw DNA data yesterday and had it back today! Now that's service! And the price was right, too! According to him, my wife is 25% Native American, 1.4% Sub-Saharan African and rest European and Middle Eastern. I was glad to see the Middle Eastern plainly marked...I'm guessing that will be from her Converso Sephardic Jewish Northern Mexican Conquistador ancestors.

AJL
02-24-2013, 11:18 PM
I sent Dr. McDonald my wife's raw DNA data yesterday and had it back today! Now that's service! And the price was right, too! According to him, my wife is 25% Native American, 1.4% Sub-Saharan African and rest European and Middle Eastern. I was glad to see the Middle Eastern plainly marked...I'm guessing that will be from her Converso Sephardic Jewish Northern Mexican Conquistador ancestors.

Great! Yes, Dr. McDonald's BGA is very good, especially with the chromosome paintings.

geebee
02-25-2013, 01:55 PM
Yes, it appears as Iberian for me though I also have some Italian, which is apparently both from Sephardi, and from my mother's French ancestry, some of which probably traces to the Milan area. In many cases, though, Sephardi may appear as Middle Eastern or North African, or Nonspecific (Southern) European, as well. I suspect there may be a current bug that overreports some ancestries and underreports others, but unless Sephardi is made an opt-in category, Iberian and Italian in combination may suggest Sephardi ancestry, though it could mean just that as well.

I, too, was somewhat surprised by "Italian". AC (speculative estimate) shows 2.2%. It also shows 1.1% Balkan, 1.0% Iberian, and 4.1% nonspecific Southern European. On paper, I have 12.5% Spanish -- all from the island of Menorca.

All of the "southern European" put together only totals 8.6%, which I suppose could just reflect the randomness of recombination. But there's also another 3.4% of nonspecific European, of which 2.9% is maternal, as my Spanish ancestry is. If it's all from my Spanish 2nd great grandfathers, then the total would be 11.5%, which seems somewhat more reasonable.

I didn't necessarily expect my Menorcan ancestry to necessarily show the same as other Spanish ancestry might, given the history of the island. I also wonder if some Sephardi ancestry might be possible. I seem to have virtually no Ashkenazi connections, or at least ones that can't be more readily explained by non-Ashkenazi ancestry on their part instead of Ashkenazi on mine.

Funny thing -- Dr. McDonald's most recent analysis actually suggested "Jewish 0.267". He showed no Jewish ancestry for my father, which would mean this was all from my mother. And it would be equivalent to saying that one of her parents was probably close to 100%. Only I know the ancestry of both of them pretty well, and this doesn't seem at all plausible. Plus there's the fact that even at the minimum Ancestry Finder settings, I only show 0.2% to 2.0% coverage for "Declared Ashkenazi Jewish". This seems to involve about 18 different segments, all small, and all but two of the associated relatives have only partial Ashkenazi ancestry. (In fact, fourteen of those with only partial Ashkenazi ancestry have just one Ashkenazi grandparent, which the other two have two each.)

Does it seem plausible that for me, too, the Iberian/Italian combination could reflect some Sephardi ancestry? (With the Balkan thrown in just to add a bit of mystery. I suppose "Balkan" could be "Greek", which might possibly fit into Menorcan ancestry along with everything else.)

Grossvater
02-26-2013, 01:35 AM
So...I'm teaching my students today about the the 1st, 2nd, and 3rd Punic Wars and the takeover of Spain by Rome from Carthage and it suddenly hits me...could some of the middle eastern DNA in my wife's McDonald results be from Phoenician Carthaginians rather than Converso Jews? It seems I saw an article regarding someone's research on Phoenician DNA in a magazine a few years ago. Since some of my wife's ancestors came from Huelva, not far from Cadiz and also from Malaga along the Southern coast of Spain which was historically ruled by Carthage...could it be within the realm of possibility? How different would Phoenician DNA be from Sephardic Jewish DNA?

AJL
02-26-2013, 01:51 AM
Does it seem plausible that for me, too, the Iberian/Italian combination could reflect some Sephardi ancestry? (With the Balkan thrown in just to add a bit of mystery. I suppose "Balkan" could be "Greek", which might possibly fit into Menorcan ancestry along with everything else.)

Yes, certainly! Though with Spain controlling southern Italy for a while, there are many possiblities -- including something like Corsican ancestry, which could also end up splitting as Iberian/Italian.

AJL
02-26-2013, 01:54 AM
could some of the middle eastern DNA in my wife's McDonald results be from Phoenician Carthaginians rather than Converso Jews? It seems I saw an article regarding someone's research on Phoenician DNA in a magazine a few years ago. Since some of my wife's ancestors came from Huelva, not far from Cadiz and also from Malaga along the Southern coast of Spain which was historically ruled by Carthage...could it be within the realm of possibility?

Yes, I think it's quite possible. There are still some yDNA traces of possibly Near Eastern influence in Spain and Portugal, not all of which necessarily reflect more recent Jewish or Moorish origins.

geebee
02-26-2013, 03:34 PM
Yes, I think it's quite possible. There are still some yDNA traces of possibly Near Eastern influence in Spain and Portugal, not all of which necessarily reflect more recent Jewish or Moorish origins.

That was your reply to Grossvater's question, "... Grossvater asked, "...So...I'm teaching my students today about the the 1st, 2nd, and 3rd Punic Wars and the takeover of Spain by Rome from Carthage and it suddenly hits me...could some of the middle eastern DNA in my wife's McDonald results be from Phoenician Carthaginians rather than Converso Jews?"

I found your answer interesting, because one of the many influences on Menorca was Phoenicia -- meaning that I could potentially see the same effect. However, in my case it could not account for the full "22.7% Jewish" that McDonald's program was seeing, since the total of my Spanish ancestry is just over half that much.

I wish I'd gotten results from Dr. McDonald for the two of my siblings who have 23andMe data. (They'd have to be run separately from each other or me, of course.)

I do have various calculator results for their data, and usually we have fairly similar results. FTDNA's Family Finder is kind of an exception, since it declares both my brother and me to be "100% European" -- but with more Finnish and Russian than we are likely to have -- but it says our sister also has "2.67% East Asian, 0.26%".

Of course, that's a small enough percentage that she could conceivably have it even though we missed it. However looking at 23andMe and lots of calculators (and McDonald's results for me), it doesn't actually look as if we missed anything. For example, here are our World9 results:

my brother Curt - Amerindian 2.71%
East_Asian -
African 0.21%
Atlantic_Baltic 69.72%
Australasian -
Siberian 0.53%
Caucasus_Gedrosia 11.26%
Southern 15.58%
South_Asian -

Our sister Luci

Amerindian 1.85%
East_Asian 0.91%
African 0.12%
Atlantic_Baltic 65.00%
Australasian -
Siberian 0.24%
Caucasus_Gedrosia 13.24%
Southern 18.12%
South_Asian 0.51%

Me --

Amerindian 2.58%
East_Asian 0.56%
African 0.09%
Atlantic_Baltic 65.11%
Australasian -
Siberian 0.74%
Caucasus_Gedrosia 14.68%
Southern 15.95%
South_Asian 0.30%

Curt has no "East Asian" here, but then again, his East Asian is not too far from the sum of our sister's "East Asian" plus her Amerindian. It could certainly be that she has both; but it could also be that the same component is being seen both as Amerindian and as East Asian, depending on which SNPs are involved.

And that is one of the problems with any of the biogeographical analyses. I guess one lesson may be that it helps to know what populations can give similar results. For example, Italian versus Jewish, possibly -- but I'm sure the list could be quite long. So with a given calculator -- or with Ancestry DNA, FTDNA, or 23andMe -- you have to know what the reference populations are. It would also be very useful to see what kinds of results you get for people with known ancestry from given locations -- to see how much "sense" the analysis is really able to make.

AJL
02-26-2013, 08:17 PM
I found your answer interesting, because one of the many influences on Menorca was Phoenicia -- meaning that I could potentially see the same effect. However, in my case it could not account for the full "22.7% Jewish" that McDonald's program was seeing, since the total of my Spanish ancestry is just over half that much.

Thanks! It would be interesting to see McDonald's results for all your family members. Testing families is great because sometimes full siblings have quite divergent results.

Grossvater
03-04-2013, 04:39 PM
Things are getting curiouser and curioser...

Amerinese quite generously agreed to run my wife's raw DNA date through the Dodecad software. Here are the results:

'DIYDodecadWin globe4.par'

66.61% European
1.39% Asian
1.12% African
30.88% Amerindian

'DIYDodecadWin globe10.par'

25.38% Amerindian
9.42% West_Asian
0.14% Australasian
0.11% Palaeo_African
0.94% Neo_African
2.20% Siberian
10.73% Southern
0.28% East_Asian
49.80% Atlantic_Baltic
1.00% South_Asian

'DIYDodecadWin globe13.par'

1.19% Siberian
24.63% Amerindian
1.04% West_African
0.05% Palaeo_African
2.60% Southwest_Asian
0.07% East_Asian
26.44% Mediterranean
0.11% Australasian
1.84% Arctic
8.54% West_Asian
32.70% North_European
0.80% South_Asian
0.00% East_African

Below what number does one consider something "noise?" What does Paleo-African mean? I'm thinking that the Arctic/Siberian/East Asian etc. is actually Native American...I still wonder if it isn't a reflection of some Lipan Apache (i.e ancient Athabaskan) ancestry since she has no known Asian ancestry...just lots of Native American from Northern Mexico. Thanks again, Amerinese for the help.

Grossvater
12-08-2013, 02:09 AM
I checked to see my wife's results after hearing that 23andMe was making some changes and there are some VERY interesting ones:

1. The new results (in speculative mode) say my wife carries .2% Finnish ancestry through her mother. Since her mother paternally comes from Native Americans who lived in a small village in Southern Nuevo Leon in Mexico and maternally from a Mestizo family 40 miles away, this is rather hard to swallow. Could this be ancient quasi-European DNA dating back to Mal'ta Boy's time and carried to America over the Bering Strait by my wife's Native ancestors?

2. It appears they have lowered the amount of SSA my wife carries to 1% but they say its all West African.

3. They upped her Asian/Native ancestry to 25.3% and say she is .1% Mongolian and .1% Yakut. Once again, I wonder if this is actually Athabaskan ancestry?