Thursday, February 20, 2014

#RootsTech – Ancestry DNA Saturday

AncestryDNA's Ken ChahineSaturday, Dr. Ken Chahine introduced the final keynote speaker. Chahine is a senior vice president at and general manager of AncestryDNA. Chahine made some brief remarks with some cool information for DNA fans.

He mentioned that AncestryDNA’s database is up to 300,000 profiles. That’s pretty impressive when one compares it to the National Geographic’s GenoGraphic project, which has just over 660,000 samples after the many years it has been in operation. Coincidentally, Family Tree DNA also has just over 660,000. (Both figures coming from their respective home pages.) To be fair, I should point out that AncestryDNA’s number does not include those acquired from Sorenson Molecular Genealogy Foundation (over which I am still sorely displeased that Sorenson could sell my DNA and accompanying pedigree).

Chahine told us that AncestryDNA has additional advancements in ethnicity estimates in the works for later this year. You’ll recall they just barely upgraded their ethnicity regions in October 2013 from 22 to 26. (See “AncestryDNA Gets Big Update.”)

Chahine said that one day advancements might allow placement of ethnicity down to a particular town of ancestry. They one day may be able to show you the migration patterns of your ancestors. They might be able to tell you the slave trading market in South Carolina where an ancestor might likely have been sold. They may be able to identify physical characteristics of ancestors like hair and eye color.

That’s pretty amazing stuff. also sponsored a luncheon Saturday where AncestryDNA had a small panel of people. Kenny Freestone I know from my days at He is the DNA product manager. Ken Chahine chaired the panel. My apologies to the other guy on the panel; I didn’t catch his name.

In introductory slides we learned that when you submit your DNA sample, it and the derived information travel all over the country. AncestryDNA sends your sample to the Illumina DNA lab in San Diego. The lab is almost fully automated, which minimizes handling errors. They extract the DNA from your saliva sample and put it on a processing chip. The chip can do 12 people simultaneously. It reads 700,000 markers.

The data is then sent to San Francisco for analysis. There they compare your DNA to reference samples that allow placement of your ethnicity to 26 global regions. For the upgrade that will occur this year, they don’t know yet how many regions they will be able to further break down.

They are also expecting a dramatic increase in accuracy of their cousin estimates.

They then took questions from the audience. For lack of a wireless microphone, they utilized Crista Cowan. :-) She’d move around the floor, listen to a question, and then repeat it so everyone in the large room could easily hear. Good job, Crista! Can I use Cristaphone the next time I speak in a large hall?

I only captured a few questions before I had to leave early. (I discovered I had left my cell phone someplace. And I had to prep for a lab right after lunch.) Here’s what I’ve got:

  • When you upgrade your system, will I have to submit a new sample? No. We will rerun your results automatically, for free.
  • Is there a way to narrow down the list of possible relatives? To filter your list, there is a search field at the top right of the match page where you can enter a surname and location.
  • Is there anything that can be done in the cases where matches have private trees? will create tools to make people feel more comfortable with sharing some information with related persons. Maybe they’ll let you reveal part of a private tree.
  • Why do I get different ethnicities from different companies? We can't address that but we can tell you that we use a lot of rigor. The results depend on the samples used for deep reference. Europe is much more difficult because of recent interactions. That’s where the error bars become important. There’s very little difference between Great Britain and Europe. That's why the error bars are so important. The assets of Sorenson’s DNA sampling program give lots of confidence to AncestryDNA’s results and gives them power to figure things out.
  • Why spitting and not swabbing? Saliva gives higher quality DNA samples. They would have liked to stay with swabs, but went with the higher quality.


UPDATE: AncestryDNA's sample count does not include the samples obtained from Sorenson.


  1. When National Geo came out with their program - I bought level 1 and 2 for DNA results. That was in 2007... Was all of that for naught? The number of DNA samples in the various databases - you mean they are not being shared? The numbers are low enough when you consider the population of the United States, but not have them available for cross reference? Please tell me I am wrong and set me straight...

  2. Will the Ancestry DNA test be available in Canada soon?
    Can you explain the delay, please?
    Thanks, Sue

  3. You can choose to remove your DNA and genealogy information from the SMGF database so that your information is not used for profit. Details of how to do this are provided in the SMGF consent form:

    I think Ancestry should be getting fresh consent to re-test those Sorenson samples. I have not had satisfactory responses to my questions on this matter from Ken Chahine. See the comments section on this blog post by Ricki Lewis:


Note: Only a member of this blog may post a comment.