Biogeographical Ancestry

   Earlier this year I had my Autosomal DNA scanned by Family Tree DNA. They use this data for various products they offer as well as providing a copy that you can download for yourself. So now that I have a copy of my data that I've been submitting  to numerous biogeographical studies. These studies use the 500,000 pairs of locations (SNPs) to look for ancestral markers. Basically we all carry little markers that give hints to what people or tribe we come from. The old tests were pretty lame using 13-100 CODIS markers to identify your origin. The new tests available today utilize 500,000 + markers within our DNA , a substantial difference.

   Of course you can have your entire Genome scanned by a few companies, but I don't have the extra $20,000 at the moment to do this. The average person is getting their data from FTDNA or 23andMe with the price ranging from $250-$500 US for these tests.

   So we are dealing with 500,000 markers and researchers believe that only 100,000 are necessary to get a good idea "what" you are. My data has been run four times by different researchers. Here are the results:







Test # 1
First I ran my data through Dienekes'
  EURO-DNA-CALC

This program gives you a percentage estimate of your Northwest Euro, Southeast Euro, or Ashkenazi Jewish heritage. I came back 100% Northwest Euro.


[1] "NORTHWEST EURO: Maximum Likelihood Estimate=100% Interval=[71, 100]"
[1] "SOUTHEAST EURO: Maximum Likelihood Estimate=0% Interval=[0, 29]"
[1] "ASHKENAZI JEWISH: Maximum Likelihood Estimate=0% Interval=[0, 22]"






 

Test # 2

Next I submitted my data to the Eurogenes Project.  This project is one of the most exciting ones happening right now along with Dr. McDonald's project. David sends out updates almost weekly as new people join the project and he gives you a percentage breakdown along with numerous maps.

I'm listed as F-US9 and sit between the Orcadian and French samples.

Here's an example of what David provides:







My percentage breakdown from Eurogenes:

Western European 39.23%
Anatolian/Caucasian 4.35%
Middle Eastern 0.5485%
Northeast European 35.72%
Mediterranean 20.13%
North African 0.001%


A run with Native American reference populations for comparison.  I have no Native American DNA according to this test.

Amerindian 0.001%
Anatolian/Caucasian 23.19%
Northeast European 75.63%
North Eurasian 1.14%
East Asian 0.001%








Test # 3
I sent my data to Professor J.Doug McDonald,
 Professor of Chemistry at the University of Illinois at Urbana-Champaign. His current research interests include studies of the folding of single protein molecules. He is the Assistant Administrator and Data Custodian of the Clan Donald DNA Project.


Dr. McDonald has been accepting data and running reports for DNA hobbyists like myself for the last few years. Dr. McDonald's analyses of my DNA was as follows:



I can't tell specific European mixtures, only if you fit a pure comparison panel I have, or else I can just give what the "average" is, unless the Euro is one of the oddball ones like Sardinia.

My "spot on the map" puts you in the English Channel between London and Belgium. This implies, of course, that British/French works, and it does. You should take anything other than the average with a very large grain of salt. The average is usually reliable.

I see no Native American using any of the tests I have, which includes one like 23andMe's "Ancestry Painting" but I have real Native American comparisons. This test can IN SOME CASES reliably detect or exclude Native American on the 0.8% scale, and is very reliable at 3%. Dr. McDonald says "In your case, I'd simply say that while I see teeny bit of it purportedly here and there, they are WELL within the noise and are not real."

No test reliably gets individual country fractions accurately inside Europe, mine, or others (except those special cases ...if somebody is 50% Basque 50% Romanian, for example, it will get the Basque quite correct but may say Russian or Lithuanian, not Romanian, etc.)

Doug






Test # 4
Family Tree DNA Population Finder

FTDNA provided the raw data I used in the tests above. Their Population Finder test is still in beta and they tend to be more conservative than the other tests.

My results were:
Europe (Western European) 93.26% ±6.82%
Europe 6.74% ±6.82%


The Population Finder Program determines your biogeographical ancestry — the story of your personal genetic history — by comparing your autosomal DNA to that of our world DNA population database. Your Population Finder results consist of up to four out of seven continental groups.* For each, the percentage of your genome that matches is shown. You may view your results as either a stacked bar or pie chart.


So around 90% of my DNA matches the Western European reference populations which are:

Basque
French
Orcadian
Spanish
 
The other 10% or so of my DNA match the other European populations but not close enough to give a subgroup, so it's put into the generic Europe category.

These are listed below:

Northeast European
Finnish
Russian


Caucasus
Adygei

Southeast European
Romanian

Southern European
Italian
Sardinian
Tuscan
 
Hopefully when the Population Finder product comes out of beta, there will be further refinement.
 
 
So what do I think about my ancestry now that I've had my DNA analysed four different ways? I think I'm mostly Irish, English, and French. Traditional genealogy has pointed to these populations and the DNA tests seem to point in this direction as well. 
 
My family has a strong history of Native American ancestry through my mom's side of the family as well. I believe that this does not exist since I've found no proof genealogically or within my Genome.
 
I'm very grateful for the researchers above that have given their time with these projects. They are filling a needed void in the Biogeographical Ancestry testing field.




2 comments:

Jim Owston said...

Nice summation of the various analyses that you received. I'm going to have to do the Eurogenes project myself and I am waiting on FTDNA results.

Salabencher said...

Thanks Jim.

I would love to be part of Dienekes' project as well if he opens it up again.