Sunday, February 8, 2015

AncestryDNA Sells 100,000 Kits In December 2014 And Comparing Cousins Results

Ancestry.com reported their 4th quarter earnings, and year end earning results, on February 3rd ( you can listen to the full report here).  I guess the AncestryDNA boycott hasn't caught on yet because they sold 100,000 kits in December of 2014 alone; that's double what they sold in the same month last year which was around 50,000 kits. Most of these kits, I suspect, were purchased by the Millennial generation who are interested in the Ethnicity results. This seems to be reflected in the Ancestry subscription numbers which are declining instead of growing. Ancestry lost 25,000 subscribers in 2014. I'm not sure whether adding a subscription requirement to see your matches' surnames will increase subscriptions? The young people testing for ethnicity don't seem interested in looking at match information. I had planned on subscribing to the trees using the "Ancestral Insights" subscription. That was before they instituted the new subscription requirement for new accounts. I don't like to be coerced into buying a subscription so I'm thinking that purchase over now.

Revenues are up for Ancestry.com in the 4th quarter. Losses are down compared to a year ago. There were some great sales last year apparently due to slowing sales. With such incredibly high kit sales recently I don't expect to see another sale, at least until they process all the kits purchased in late 2014. AncestryDNA is driving revenue growth at Ancestry. AncestryDNA will be expanding into Germany and Mexico this year.

Sounds like more novelty features will be added to the AncestryDNA product. Maybe a Neanderthal prediction? As I understand it these new features will not include the sharing of segment information. An interesting blog post was published this week about the DNA Circles and their weakness as a tool for determining how you are related to matches. The post  Anatomy of a DNA Circle explains the problems faced when drawing conclusions based on Circles. When I contact matches at AncestryDNA, sadly, most don't understand the benefits of triangulation with DNA matches. AncestryDNA misleads their testers into believing they don't need to do anything except attach a tree.

Like so many genetic genealogists I've been trying to figure out possible cousin relationships based on the amount of shared DNA. My Mom was born in Granada, Nicaragua to Charles Lynn Forgey, a US Marine based in Nicaragua, and his Nicaraguan wife Graciela Del Castillo. My mother was 4 years old when she came to the US. I have very little information about my Nicaraguan family. Luckily the civil registration records for Managua, Nicaragua are online. Unfortunately the records for Granada, Nicaragua aren't, and that's where most of my ancestors lived. I've been examining my Mom's Nicaraguan matches at AncestryDNA and Family Tree DNA. I had been contacted, a couple weeks ago, by a distant cousin whose Aunt matched my Mom at AncestryDNA. She is a "Very High" confidence match for my Mom, and my family was acquainted with her great-grandfather Francisco who lived not far away from my family here in Southern California. AncestryDNA predicts they are 3rd to 6th cousins. Her 2nd cousin, on her only Nicaraguan line, also matches my Mom at FTDNA. Her 2nd cousin doesn't match my Mom as closely however. He is a 5th cousin remote sharing an 11 cM segment. Looking at both of these predictions, plus the fact our families were acquainted with each other, I'm guessing 4th cousin? Don't know if I will ever be able to confirm that? If the Granada records still exist, and go back far enough, I may be able to discover our exact relationship..

I've been analyzing my cousin Darryl's results. These comparisons, again, highlight the fact it's nearly impossible to predict relationships past the 2nd cousin degree of relationship. The first chromosome chart compares him with his Aunt Loretta. He shares 1704 cM's total with her. The longest segment is 162 cMs. My mother and her niece share 1769 cMs. Longest segment is 156 cMs. Generally those who share this degree of relationship share between 1500-2000 cMs. (got most of the cM averages from this ISOGG page here)


Aunt Loretta and Nephew Darryl

 
 
A comparison of Darryl and I. We are first cousins. We share 1074 cMs. Our longest segment is 156 cMs. I share 1041.81, total cMs, with another first cousin. And the longest segment is 75 cMs. First cousins generally share 548-1139 cMs. 
First Cousins Darryl and Annette
 
 
Moving on to a third Cousin to Darryl. He shares only 27 cM's with our  third cousin; the longest is 16 cMs. Vastly smaller share than his Aunt Loretta and I. I share a total of 149 cMs with our third cousin. The largest segment is 69 cMs. Aunt Loretta share 182 cMs with the largest segment being 88 cMs. Third cousins generally share 16-111 cMs,  I believe my Aunt would be a 2nd cousin 1x removed from this cousin. Second cousins once removed generally share 19-197 cMs.
 
Comparing between 2nd cousin 1x removed and 3rd cousins


Darryl's 3rd cousin 1x removed shares 35 cMs and the largest segment is 14.9.  I share 91.5 cMs and the longest segment is 32 cMs. Our Aunt Loretta shares 65 cMs the largest segment being 37.9. I share more DNA than my Aunt and cousin with this match. The average 3rd cousin 1x removed shares 0-99 cMs.
 
3rd cousins 1x removed and 3rd cousin



Darryl shares DNA in the same place on chromosome 1 as a 5th cousin on our Melvin line. We have great triangulation with this Melvin match. Again Darryl's segment is smaller at 7 cMs but is in the same location. The rest of us share about 15 cM segments in the same place.

4th and 5th cousins to Melvin match


 
Apparently Darryl's father, Thomas Kapple, inherited some segments from his father that my father, Robert Kapple, did not get. We don't know which line or lines of our Burgenland family this DNA was inherited from? This match shares the surnames Jost, Kurta, and Koppel. Since my Aunt inherited half of her DNA from her father I would have thought she would have inherited the longer segment. Instead her nephew Darryl shares the large 32.5 cM segment, and Aunt Loretta shares an 11.8 cM segment in the same place.
 
Unknown relationship with several possibilities
 




4 comments:

M. Dawn said...

Very interesting blog post.

M. Dawn said...

Very interesting blog post.

Jason Lee said...

Thanks for the link.

AncestryDNA's biggest supporters have emphasized the great potential in combining Ancestry's trees with Ancestry's DNA test. But Ancestry's tests sell quite well as ethnicity reports. Consequently, Ancestry isn't deeply interested in combining the power of their trees with the power of genetic genealogy, and they've erected a big paywall between the DNA test results and the trees.

Customers who ARE interested in genetic genealogy are being led to believe that AncestryDNA's "shaky leaf hints" and "DNA Circles" offer "new genetic evidence" for ancestral relationships. Unfortunately, the DNA "evidence" at Ancestry cannot be verified or checked for errors without orchestrating a comparison using third party resources.
Most customers undoubtedly have no idea what they're missing.

Ancestry apparently wants to give the impression that DNA confirmation of ancestral relationships is simply a matter of linking test results with a family tree. The reality of genetic genealogy is not nearly so simple.

Annette said...

Thanks for the comments! I find that using AncestryDNA has become such a time consuming proposition. Not having any hints whatsoever on how many segments we share with matches and their size makes it impossible to determine the quality of our matches. Matches have to be contacted, if they don't answer they have to be re-contacted. They often will not answer or agree to compare elsewhere. Someone asked me to recommend a company for testing the other day. It's nearly impossible to recommend AncestryDNA other than based on their price on sale.