Field of Science

Doug McDonald's BioGeographical Ancestry Test

In my quest to explore various aftermarket sources of information on my 23andMe raw data, I emailed my data to Doug McDonald for his BGA testing service. I first found out about his service on the 23++ Chrome extension website, although the email they have listed for McDonald is wrong. I finally got the correct email (mcdonald at scs dot uiuc dot edu) through the 23andMe community forums.

Here's what he sent back:

Michelle: The program says you are English, 100%. However, on the chromosomes one sees a small <.5% American block that is fairly strong and likely real, but not 100% sure. The other blocks on the chromosomes are likely noise.

Here's a look at the chromosome painting you get from McDonald (similar to 23andMe's "ancestry painting"):

Click to enlarge.

The red sections are European, the brown sections are no data/not enough data, the two small blue sections are African, the small grey section is Mideastern South Asian, and the green section is Amerindian. I would like to know which side the Amerindian segment is on so I can explore this further. My mother's 23andMe results are due any week now, so I will likely email her data to him as soon as I get my hands on it.

You also get a number of what I think are PCA plots, showing where you map out with respect to his reference groups. He only sent one for me, but he sent multiple for my boyfriend. Here's mine:

Click to enlarge.

The crosshairs show where I fall, in the middle of the English cluster, close to the French side. I thought I had more of an Irish minority than a French minority, but hey, whatever. I notice he doesn't have a sample German population, either.

Most of this is not new information to me, but does provide corroborating evidence for the ~1% Amerindian that shows up for me in Razib's admixture analysis. The results for my boyfriend go along more or less with what came up for him at the Harappa Ancestry Project.

3 comments:

  1. Fascinating!

    I was one of the early participants in this, and his first Irish sample a few years ago. I must ask for an updated analysis, and have my parents analysed too.

    In terms of veering towards the French, this should be expected, if your Irish or English ancestry is from a Norman source, or from South East Ireland, or Southern England, areas where the Belgae settled - and probably other continentals before them.

    Did you do participate in the Dodecad IBS? My mother had many French reference population matches. When I transferred my data to DeCodeMe, the population I matchede most was French - out of the set (Orkneyar, Icelandic, French, Basque, Russian) - and second was Icelandic, third Orkneyar, though the latter is supposed to represent people from Britain and Ireland.

    ReplyDelete
  2. Yep. I'm pretty sure you commented on that post, Paul. Getting forgetful? ;)

    ReplyDelete
  3. Didn't have my 2nd cup of coffee - so brain still booting up ;)

    ReplyDelete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS

Anonymous comments will NOT be approved. You must include a name at the very least.