Search for content in message boards

Five Myths about Ancestry DNA

Five Myths about Ancestry DNA

Posted: 6 Jan 2013 4:59PM GMT
Classification: Query
My local paper has a popular feature called "Five Myths about . . . " in which commentators provide insights about common assumptions on a variety of topics. Having read through these message boards for a number of months, I've seen some recurring issues that seem to be related to various assumptions about how DNA analysis should work. I've provided my thoughts below. I'm far from expert, and would welcome corrections or elaborations on these and other "myths" associated with Ancestry DNA.

Jim


Five Myths about Ancestry DNA

1. My ethnicity report should match my family tree.

Not necessarily. Most family trees go back a few generations or at most a few hundred years. The ethnicity report is based on the analysis of genetic markers that may go back hundreds or even thousands of years. Moreover, the ethnicity labels do not necessarily reflect modern national borders or cultural backgrounds; they are names given to particular reference populations by the scientists who conducted the studies and some are quite arbitrary. The ethnicity analyses are fairly reliable on a continental basis (Europe, Asia, Africa) but much less precise for regions or countries, especially for those where there has been a great deal of mixing among groups over the centuries, such as Western Europe.

2. My matches should have the same ethnicity that I do.

Because recombination of DNA is random, and the shared segments diminish with each generation, it is quite possible to have a close match with a person whose ethnicity is very different. Consider, for example, a distant cousin who descends from a common relative, but whose family members in each subsequent generation intermarried with individuals from different regions and cultures. By the time you each get to the current generation the newly introduced ethnicities could very well overwhelm that of your common ancestor. Another possibility is the one or both of you have a “Non Parental Event” (NPE) in your line. This may be an adoption, orphaning, fostering, out of wedlock birth or other event that caused a mismatch between a person’s genetic background and surname.

3. If there are no shared surnames in our trees, my match is false.

The Anceststry.com surname list includes only direct ancestors’ surnames out to 10 generations. If your match is on a collateral line, those surnames will not be included. Also, the match may be before either you or the other individual has documentation, or there may have been an NPE, as discussed above. It is always important to examine the actual trees, including both individual names and locations, rather than relying only on the surname list.

4. Ancestry family trees are full of errors and useless for understanding DNA results.

It is true that many trees have incorrect entries or lack documentation. This is especially true when the authors have done “name harvesting,” indiscriminately copying entries into their trees without investigation. However, diligent research even in these trees can lead back to original sources that may be valuable in resolving mysteries. Moreover, communicating with the authors of such trees, and sharing information about DNA results, may result in a correction that will benefit both of you and other future researchers.

5. Once Ancestry.com provides our raw data these problems will be resolved.

Raw data refers to the actual values associated with each location that was tested, usually made available in a large table or spreadsheet. In itself, this array of data is not particularly useful except to people who are expert in reading the entries. The value will come if the data are provided in a form that will facilitate further research. For example, with appropriate analytical tools (some of which are available currently in various online sites) users may be able to identify the exact chromosomes and segments on which matches occur and independently evaluate the reliability of their matches and identify others who share those segments. They may also be able to use other admixture (ethnicity) calculators and compare the results with those provided by Ancestry. Finally, if the format of the data permits, it may be uploaded sites where people who have tested with other companies have posted their results, thereby expanding the pool of potential matches. Merely having the raw data, however, will not necessarily provide more information and understanding unless it is accompanied by the ability to do more analysis than is currently possible using Ancestry.com’s tool kit.

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 5:32PM GMT
Classification: Query
Where are you getting this surname comparison list you speak of in #3?

Even the one tree brought up by my "has a hint" filter does not suggest the two individuals involved.

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 6:09PM GMT
Classification: Query
The surname comparison list is to the left when you bring up your match's tree.

Here's a screen shot, showing the shared surnames and the list:
Attachments:

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 6:21PM GMT
Classification: Query
Be sure to click on those names in the shared surnames.

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 6:27PM GMT
Classification: Query
And here is a screen shot showing an Ancestry hint. Note that his match was a very low probability distant cousin who ended up being much closer than predicted.

I agree that one should click on the surname, and also look for all people of that surname in the list of names in your match's tree, since your connection may be with a sibling who is not a direct ancestor.

One further point, which I have made before, regarding going beyond shared surnames. I have a 95% 4-6 cousin match with a man named Kelly, a surname that is not in my tree, and none of my family's surnames are in his. However he has an ancestor named Driscoll, a common name in my family's home area in Ireland. His Driscoll ancestor's parents were not listed in his tree, however I found them in online Irish records and his mother was my 2xGGM's niece. This made him a 4th cousin once removed, a match which confirmed my ancestor's first name and place of origin.

Jim
Attachments:

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 6:48PM GMT
Classification: Query
Now I'm really upset. I've never seen those two screens on any of my DNA, and I have 100s of matches. That would have saved me hours of work.
I hope I'm not hijacking too much.

When I first open the matched tree I get a popup saying s_initial_beacon not defined". I just click OK and it opens,
up the tree, but never with the info you show, just the regular tree. Maybe I have an ad block that is blocking. I'll try another browser. I'm on firefox now.

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 7:19PM GMT
Classification: Query
Edited: 6 Jan 2013 7:28PM GMT
Excellent piece! I am bookmarking this one to share in other forums when trying to clear up some of the confusion surrounding AncestryDNA results, what they mean, and what to do with them.

I would add: The list of shared surnames has no significance other than the fact the surnames appear in the direct lines, in the first 10 gens, in both users trees. It is ancestry's way of saying "Hey, you share dna with this person and you both have these names in your paper trees. You might want to check that out for a *possible* connection." Your match is to the other user and was based on dna. The interface that displays your match's tree data is simply a tool that that allows you to more easily compare that data. There is nothing presented there - other than the predicted relationship and confidence level - that is based on dna, (including leaf hints to a common ancestor in both trees).

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 8:15PM GMT
Classification: Query
Just pulled up on IE and there it is!

Something must be blocking on Firefox. I hope no one else is not aware of this.

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 8:21PM GMT
Classification: Query
Your AncestryDNA experience just got a little easier :-)

Re: Five Myths about Ancestry DNA

Posted: 6 Jan 2013 10:34PM GMT
Classification: Query
This post was deleted by the author on 16 Jan 2013 3:19AM GMT
per page

Find a board about a specific topic