An example of why Ancestry DNA shared ancestor hints need to be verified

To better illustrate why the Ancestry DNA shared ancestor hints should be used only as a suggestion of where the possible connection may be with your DNA match, rather than something taken at face value as confirmation of where the connection is with your DNA match, I wanted to create an example to show how a shared ancestor hint can be completely wrong and misleading. As you can see, in the list of matches for my maternal grandmother’s brother (R E H), he has a shared ancestor hint with both myself and my mother, as indicated by the leaf next to the number of people in our respective trees.

John Q Sample Example 2

Upon clicking on my name to look at the details of the match with me, it shows the shared ancestor hint, and the specifics of how we are related. It is showing we share John Q. Sample as our most recent common ancestor.

John Q Sample Example 1

Looking at the details of John Q. Sample, you can see how I purposely used dates which are impossible, in addition to the completely fabricated name.

John Q Sample Example 3

How did I get this result? I first removed my great grandparents as the parents of my maternal grandmother and my maternal grandmother’s brother (R E H). Next, I created my fictitious John Q. Sample, and then attached him as the father of my grandmother and her brother. I waited a few hours to allow the Ancestry DNA system to process the change, and the shared ancestor hint then appeared. This would be the same result as if someone had attached an incorrect person to their Ancestry member tree and taken an Ancestry DNA test. If your DNA happens to match the test taker, and the incorrect person they have attached to their tree as one of their direct ancestors happens to match one of your direct ancestors in your tree, a shared ancestor hint will be shown in each of your match lists.

As I hope this example clearly shows, the shared ancestor hint is only indicating that you and the other person have the same person in each of your trees as a direct ancestor. It is not an indication of the accuracy of the person’s tree, nor is it in any way a verification of where in your tree the DNA connection is. The only way you can properly verify the accuracy of the match, and with which ancestor(s) the match exists, is to use a chromosome browser (such as the one at the GEDmatch site) to make sure the same chromosome segments are shared by at least three people who share the same direct ancestor(s). This is called triangulation. How to go about doing triangulation will be the subject of my upcoming posts on how to use the GEDmatch site and tools.

For those who reached this post from the example link, return to Chromosome Mapping And GEDmatch: An Overview Of What They Are And What The Benefits Are.

This article last updated 16 Dec 2014

Advertisements

5 thoughts on “An example of why Ancestry DNA shared ancestor hints need to be verified

  1. Pingback: Chromosome Mapping And GEDmatch: An Overview Of What They Are And What The Benefits Are | Adventures In Genealogy Research: No Stone Unturned/The Wright Stuff

  2. I see your point, and I think most of us already know this. But, if I repeatedly find different DNA matches whose approximate degree of relationship matches the “shared ancestor”, and those are back around 1700, then I think it would be an odd coincidence that those other people accidentally traced incorrect lineage over 300+ years to reach the same incorrect person that I incorrectly reached.

    Even in your example (only two descendants making up a fake name a generation or two back), the odds would be slim that you both would come up with the same fake name, unless you were to plan to do so beforehand.

    So, if I repeatedly find DNA Matches to the same ancestors along the same line, within a three generation period, then it seems to me that they are likely truly my ancestors.

    Example: I have had one DNA Match that leads me to my 9th great grandpa, born in 1685, as Most Recent Common Ancestor. That means that both of us would have to incorrectly trace our family trees for nine generations over 329 years, and accidentally reach the same wrong person. I’ve had several matches to his son, my 8th great grandpa…all traced up from different descendent lines.

    I’ve also had DNA Matches with MRCA links to his great grandson, 2nd great grandson and 3rd great grandson.

    Does triangulation take into account the fact that if three people do triangulate, then that is proof. But because DNA is not always passed on over so many generations, a lack of triangulation might not be proof of a non-relation.

    Am I understanding correctly?

    • Hi Albert,

      I appreciate your feedback. Thank you for your thoughtful comments and great questions.

      If you are repeatedly finding DNA matches to the same ancestral couple, then by all means look into it further. I was not trying to say that all Ancestry DNA shared ancestor hints are incorrect and should be ignored. By the same token, however, one should not assume all Ancestry DNA shared ancestor hints are correct and just accept them at face value. The key point I was trying to make is that the shared ancestors hints are suggestions, and need to be looked at further via a chromosome browser and triangulation to verify which ancestors the segment has been passed down from. Without looking at the segment and doing triangulation, one is merely guessing as to the accuracy of the match, as others have also pointed out (see the Ancestry – How Great Thou Aren’t section about halfway through this post: http://dna-explained.com/2013/12/28/2013s-dynamic-dozen-top-genetic-genealogy-happenings/).

      While the odds are probably small that several people will attach the same incorrect ancestor to their trees, I’ve seen it happen. In my ancestral line, there are at least two cases of this I am aware of (https://stonefamilytree.wordpress.com/2013/09/24/col-jonathan-jones/ and https://stonefamilytree.wordpress.com/2013/09/25/john-trumbull/). In addition to verifying segments using a chromosome browser and triangulation, traditional genealogy is also necessary to examine the lineages of the matches back to the common ancestor and make sure there are not errors. The good thing is that as more and more DNA comparisons are done, any errors in the lineages will start to stand out, and will likely be caught and corrected upon further examination.

      In regards to whether “a lack of triangulation might not be proof of a non-relation,” you are correct. A lack of a DNA match does not ‘disprove’ a relationship – it only means the relationship cannot be supported via the DNA comparison with that tester. Chimerism is an example of the extremely rare cases where a mother and child will not show as a DNA match to each other (http://en.wikipedia.org/wiki/Lydia_Fairchild). In addition, a DNA match cannot be considered ‘absolute proof’ of a relationship, only that the relationship is likely beyond a certain amount of doubt (usually expressed in percentage of confidence in the match). As scientists often point out, accumulated evidence can never provide ‘absolute proof’ of anything – it can only ever provide support for the likelihood of something.

      Best regards,

      Dan

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s