In my previous blog AncestryDNA Autosomal Test Results, I outlined my research question in determining the relevancy of Scandinavian ancestry in my research. However, AncestryDNA does not provide any raw data so there are no single nucleotide polymorphisms (SNPs). What this means is that the actual scientific test results are not included with the admixture mapping results which are percentages of a population group provided. In my case the admixture results showed 49% British Isles, 49% Scandinavian, and 2% unknown.
Apparently AncestryDNA uses modern population groups when providing admixture results. This can be great to pinpoint a country, or regional area, to research if you are looking at a time period when records existed. However, modern population groups alone do not add context to the family history recorded in the DNA.
Again, 49% British Isles was no surprise. My father was born in Hindley, Lancashire, the son of an American GI and a true Lanky Lass. In fact the paper trail has indicated that many of the lineages for both my parents is Scots-Irish. Not knowing if the 49% Scandinavian was from the paternal or maternal side, or both, I knew I had to gather more information. After calling AncestryDNA and talking with a knowledgeable and very helpful customer service representative, I determined that the AncestryDNA product as it is structured now would not be very helpful in determining my research question due to certain limitations.
To begin with, there is a matching service that allows those that tested to contact others that have been compared through the DNA and have a scientifcally proven match. AncestryDNA users are allowed to connect a single family tree they have created to the DNA results. This can be a big problem as sometimes Ancestry subscribers copy family trees without fact checking which can lead to an erroneous family tree. If DNA results are attached to an erroneous family tree it can make it seem that the DNA proves the lineage when in fact it belongs to a completely different family. In looking through my matches I could not find a common ancestor with any of them, even the close matches within the past few generations. Even with the DNA scientifically proving a genetic connection to an individual, the results may not be helpful due to wrong information association with a family tree.
Descriptions about the matches to the modern populations are provided. Part of the description for Scandinavia was that it seemed that I may have some Viking ancestry. Since there was no raw data, it was unclear where the information came from.
The representative with AncestryDNA explained that the professional team is looking at providing raw data in the future, but have no plans to make it available soon. The test is certainly cheaper than most other labs at $99, but not much information is provided and there is good potential for not being able to use the matches due to erroneous family trees and the descriptions do not have the proof of SNPs or raw data.
Another Autosomal Test
In order to provide a comparison to my autosomal test at AncestryDNA, I chose to have my father's DNA tested at another lab, Family Tree DNA. When the autosomal results were returned, not only was he 100% Western European (Ireland and Great Britain), but he was 100% Orcadian. Now I was getting somewhere! The Orkney Islands are some of the most northern islands of Scotland. The islands were originally inhabited by Iron Age Picts, the Vikings moved in and mixed with the Picts, and finally they became part of Scotland. The Viking inhabitants in the Orkneys could account for some of the 49% Scandinavian as shown with the AncestryDNA test.
With raw data results I was able to upload an excel file file to Gedmatch.com, which has additional admixture tests for researchers to use. When using the admixture tests, be sure to read how they are structured and what population groups are used. Reading the descriptions about them will help you make sense of the results. If you are European it is best to used a test slanted for Europeans as there is more comparative data. Also helpful is the oracle selection to define specific populations, such as Dutch, Orcadian, and Assyrian. I found it helpful to use all the admixture tests and compare results. Some admixture tests were more helpful to my specific lineage and research question than others.
Gedmatch: Dodecad V3 Oracle
One of the most helpful admixture tests was the Dodecad V3 Oracle which seems to be more for Western Europeans. Results included British, Orcadian, Dutch, Cornish, German, French, Norwegian and others. This certainly supports the results with Family Tree DNA. There is a Dutch line on my father's side of the family with the last full Dutch ancestor being born in 1749 in New York. There has also been some circumstantial evidence linking our Isbell ancestors to Cornwall, England. In reading the definitions of groups used by Dodecad, Norwegian seems to be more like Scandinavian as used by AncestryDNA. In my paper research I have not uncovered anything to indicate French ancestry, but there has been family traditions linking our Isbell ancestors to French-Huguenots. I have been somewhat dismissive of the tradition in the past as nothing much has been found to support it. What a clue for future research!
Gedmatch: Other tests
Using the other tests, I eventually found some other remnants of lineages from my research. Northern Italian showed up on a few of the MDLP Oracles and I do have one documented Italian line from Torrino, Italy in the early 1700s. Another interesting group from the test is Swedish. Although I do not have a documented lineage from Sweden, I do have one from Switzerland in the early 1700s. It is quite possible the family originally came from Sweden or that the Swiss population is included in Sweden for the test. Welsh was also another result which makes sense given most of my father's ancestry is in the British Isles. Other MDLP tests showed the following groups frequently and I have no idea where the lineage comes from: Hungarian, Finn, Slovak, Croat, and Bosnian.
Most of the results from the admixture tests I could explain with documented genealogy research. The paper trail is just as important in establishing kinship as it is in establishing the relevancy of DNA test results. DNA and traditional research rely upon each other so DNA is not a stand alone substitute for historical records.
What was surprising is that many of the population groups that are associated with my father's paternal lineage entered what we now know as the United States in the 1700s. This was quite helpful in determining the reach of the admixture tests and how they compare in regards to time and scope.
Did the autosomal tests answer my research question?
Yes and no. In order to determine the authenticity of my own AncestryDNA autosomal test I had to also test the autosomal DNA of my father with Family Tree DNA. This was needed for a comparison of admixture results and to receive a set of raw data for further admixture comparisons. In using different admixture tests, there was a necessity to research each test and what populations were used since each admixture test used different names for different populations.
So where does the Scandinavian come from in my AncestryDNA admixture? There is a good probability that some of it comes from the Orcadian group that my father descends from as the Orcadians do have some Viking and Pict history. For more accurate predictions, I would need to have my mother submit an autosomal DNA test and compare the raw data. Since we all receive half of the chromosomes of each parent, it is reasonable to suppose that some of the Scandinavian may also have come from her.
What I did not expect was to have some of lineages seemingly confirmed through the admixture tests with Gedmatch. And to gather more clues to prove or disprove traditions was an added bonus!