Wilson - DNA

A Wilson Family Tree


DNA testing is an exciting new method of genealogical research. It can supplement traditional records-based research by suggesting directions for the research or by confirming uncertain links where definitive records have not been found. It can even identify connections for which documentation might never be found; though, at least with current technology, DNA evidence can only identify that there is a connection, but not exactly what that connection is.

There are three kinds of DNA tests: Y chromosome tests, which give information only on the direct paternal ancestry (my Wilson ancestors, in my case); mitochondrial DNA (mtDNA) tests, which give information only on the direct maternal ancestry (i.e., mother, mother's mother, mother's mother's mother, etc.); and what are called autosomal DNA (atDNA) tests, which look at all the other chromosomes (i.e., not including Y and mt DNA). I have done DNA tests through four companies now: Y and autosomal tests through Family Tree DNA; the National Geographic Genographic 2.0 test and 23andMe, which look at all three types of DNA; and Ancestry DNA, which just looks at atDNA. Ancestry is by far the biggest in terms of the number of people they have tested, so there is a much greater chance of finding matches with them. As an extra added bonus, Genographic also told me that I have 1.5% neanderthal DNA! (23andMe also has some information about neanderthal DNA.)


My Y chromosome results were unexpected. R1b is by far the most common Y haplogroup in Western Europe and the British Isles, and therefore is by far the most common Y haplogroup for Wilsons (see, for example, the results for the Wilson DNA project at Family Tree DNA), so that is what I expected. When my results came back as haplogroup E3b (notation since changed to E1b1b1), I was very surprised. Only a few percent of the population of the British Isles are in this haplogroup. Further testing revealed that I am in a small subgroup even of that haplogroup, a subgroup designated E-V22, and another test showed me to be in a subgroup of E-V22 called E-CTS6080. Though unexpected, this is pretty interesting. Being in such a small group means that there is a good chance I am related to any other E-V22 Wilsons, though possibly pretty far back. Of further interest is that I have more Ramsey/Ramsay Y DNA matches on FTDNA than Wilson matches. For matches with 37 markers, I have nine Ramsey matches, four Ramsay matches, four Wilson matches, two Ferguson matches, and one Barnett match. Furthermore, I have Ramsey, Wilson, and Barnett matches who are confirmed to be E-CTS6080. It's likely that the other matches are E-CTS6080 also, but have not been tested for it. This suggests that there were so-called "non-paternity events" (adoption or adultery, so that the genetic father is not the same as the father in name) among those families, perhaps in the 1700s or 1600s.

Here is a map showing the distribution of haplogroup E1b1b1 (obtained from The Genetic Atlas):

E1b1b1 distribution

As shown in the map, E1b1b1 occurs with greatest frequency in Eastern and Northern Africa, but it does also occur with modest frequency in the parts of Europe that border the Mediterranean Sea. There is more detail on Europe in this map from Eupedia:

E1b1b in Europe

One theory for how E1b1b1 came to the British Isles is that it might have been introduced by Roman soldiers from the Balkans during the period when Britain was part of the Roman Empire. There is an interesting paper by Steven C. Bird giving evidence for this theory [Journal of Genetic Genealogy 3(2), pp. 2646, 2007; available at http://www.jogg.info/32/bird.pdf]. The discussion in the paper is specific to E-V13, but E-V22 could have followed a similar route. The Eupedia page referenced above says that E-V22 in Europe is associated with Phoenicians and Jews (see also the discussion at http://e-v22.net/descendants/). E-V22 is relatively common in the Middle East, where both Phoenicians and Jews originated. The Phoenicians had many colonies along the north African coast and across the Mediterranean in Sicily, Sardinia, and southern Spain, while Jewish people spread throughout the Mediterranean area during the Jewish diaspora. More information is needed on how E-V22 got from the Mediterranean to Britain. Information about the prehistoric migration of E1b1b within Africa and up to the Mediterranean area can also be found in the Wikipedia articles on haplogroups E-M215 and E-V68. Here is a map showing the distribution of E-V22 (obtained from the Genographic Project):

E-V22 Heatmap

Mitochondrial DNA

Unlike for Y DNA, my mtDNA results fall into a mitochondrial haplogroup, H, which is very common in Europe. It is believed to have originated in southwest Asia and to have reached Europe before the last glacial maximum (ice age). Most of Europe was depopulated during the glacial maximum because it was covered by a large ice sheet. Then, as the ice sheet retreated (around 15,000 years ago), people moved back into the depopulated area, many of them with mtDNA haplogroup H. Today, about 40% of all maternal lineages in Europe belong to haplogroup H. Some additional information is in the Wikipedia article on Haplogroup H.

Here is a map showing the distribution of mtDNA haplogroup H (obtained from the Genographic Project):

H heatmap

Autosomal DNA

In contrast with Y and mt DNA, atDNA results reflect all ancestors, not just the direct paternal or maternal line. However, the results only reflect the most recent generations rather than thousands of years in the past. A popular product of the atDNA tests is information about the ethnic mixture indicated by your DNA. Be careful not to put too much credence in the reported results, though! From my four atDNA tests, I have received eight ethnicity reports. FTDNA and 23andMe have redone their results, so I have two reports from each of them, and I have three sets of results from Ancestry. The results are different enough that it really makes you wonder how meaningful any of it is. At a very high level, they are consistent in reporting that I am mostly European. However, the breakdown of regions within Europe is not very consistent. Below is a simplified comparison of the eight ethnicity results. Note that Genographic didn't have the British Isles broken out separately, but had all of northwestern Europe lumped together. Scandinavia is not listed separately in the second and third sets of Ancestry results; it must be included in "Germanic Europe", which is part of West-Central Europe in this figure, or "Northwestern Europe", which is lumped together with England and Wales. Perhaps I should mention that "other" is so much higher for 23andMe because of a category they call "Broadly European".


Some of the results are fairly odd. The second FTDNA results reported me as 51%  East Europe, which does not seem reasonable. They also reported a Middle Eastern component from the East Middle East, shown as being around the Persian Gulf, and Genographic mentioned the Persian Gulf as well. Once again, that seems very unlikely. I think it most likely that my Middle East components are actually ancestors who were in southeastern Europe, with possibly a little infusion from the near Middle East or northern Africa. My paternal grandmother was a full-blooded Croatian, so I am 25% Croatian. The first Ancestry results are the only ones that show a large South Europe component – but they show quite a bit of that as being from the Iberian Peninsula, which seems unlikely. Probably most of the South Europe and East Europe components are from Croatia (23andMe actually includes a Balkan component, which I included in South Europe, but it is only around 5% for me). Anyway, more work is definitely needed on separating out the various European sub-populations. The 23andMe results have been more stable than the others, not changing so much in their update. Their breakdown seems fairly reasonable, too. In contrast, FTDNA and Ancestry results have changed a lot each time, and seem to me to be getting worse.

23andMe has another interesting figure, below. I'm not quite sure what to make of it, but it's interesting. It shows the Eastern European / Balkan ancestors relatively recent, which makes sense.

Ancestry Timeline

Autosomal matches to other people who have taken the test are of great interest because they can offer an independent confirmation of parts of the family tree. FTDNA, Ancestry, and 23andMe offer this sort of matching, but Genographic does not. The matches I have obtained are heavily concentrated on my mother's side of the family, at least in part because the families are bigger so I have a lot more relatives on that side of the family. Plus, some of the lines on my father's side of the family are relatively recent immigrants, so many of the potential matches are over in Europe rather than in the U.S. So far, matches have been identified for the following common ancestors (just showing the farthest back for each surname).

Great grandparents:

Great-great grandparents:

Third-great grandparents:

Fourth-great grandparents:

Fifth-great grandparents:

According to FTDNA, their Family Finder test is able to identify relatives out to about the fourth cousin level, possibly a little further. Ancestry has some interesting statements: "Our research shows that third, fourth, and fifth cousins in DNA Circles are all likely to share DNA with each other. But relatives who have a common ancestor further than six generations back are less likely to share any DNA. For example, the randomness of genetic inheritance means that there’s only an 11 percent chance that two sixth cousins will share any DNA ... If we look beyond six generations of ancestors, we also find that while two individuals might share DNA and have a common ancestor in their trees, it’s often not because they inherited their shared DNA from that ancestor. Because of the way modern populations have grown, individuals who share DNA are often related through multiple ancestors.... Our research shows that descendants of an ancestor who lived more than six generations ago have more DNA in common with other descendants of that ancestor than they’d be expected to. This discrepancy increases the more generations you go back in time and suggests that descendants are actually related through multiple ancestors". Taking that into account, I have truncated the above list at fifth-great grandparents (and even then, some of the matches listed might not be meaningful).
Note: Some of the information in these pages is uncertain. Please let me know of errors or omissions using the email link above.    ...Mike Wilson

Page updated March 2020