Friday, April 3, 2015

Ancestry's New Ancestor Discoveries Feature - Grade "F" - Update

Update #1:

Well I'm not the only one blasting this new feature. Roberta over at the DNAeXplained - Genetic Genealogy blog did an excellent job presenting her case. You can read her comments and analysis at http://dna-explained.com/2015/04/03/ancestry-gave-me-a-new-dna-ancestor-and-its-wrong/.

Original Post:

With a bit of fanfare yesterday Ancestry rolled out it latest DNA tool/feature "New Ancestor Discoveries" and I will admit it is in Beta.


I first saw this new feature early yesterday afternoon and on first glance, my initial impression was that maybe this would be the genetic genealogy tool replace the tedious researched match digging I have been doing for the last 3 years.

Is this truly a ground breaking DNA research tool that could simplify finding long lost ancestors and knock down lineage brick walls?  Will this be Ancestry's answer and replacement to the long sought after chromosome browser that they refuse to make available to their DNA customers?



In short, no on both accounts. In fact, to be quite honest, I think this is one of the worse new features I have ever seen Ancestry roll out in the nearly 15 years I have been a member/subscriber of the site, and I have seen more than a few of their stinkers.

I manage six Ancestry DNA tests and work with a seventh test (a cousin who was adopted). I have two Ancestry "ancestor only" trees involved with these six tests and they are both very mature and well documented. The ancestral lines have been well proven using paper trails and various DNA testing techniques (autosomal/Y-DNA at Ancestry and FTDNA). All my results have been uploaded to GEDMatch and I use that site to verify and supplement the Ancestry/FTDNA testing results.

Of the six testers that I closely manage, four testers had these "New Ancestor Discoveries," and two in particular, my father and spouse," had hits using this new network genetic feature. The other two are cousins off one side of the family tree and since we have not proven all lines, their matches cannot be verified for accuracy.

I carefully studied the all of the "genetically networked" circle ancestors presented by Ancestry and did so in-depth since I am extremely familiar with my dad and spouse's lineage (35 years experience). Closely comparing the supposed new ancestor matches in trees of those in the circles to our trees and came to the conclusion that none of the "new ancestors—just by looking at your DNA" that Ancestry found for my testers can even remotely be associated with anyone in our trees during the time frames implied by the timelines of those ancestors and the relationships described to the matches in these networked Ancestry DNA circles.

I agree with several individuals and in particular Shannon Christmas who posted comments on the Ancestry Blog in regard to this rollout. Shannon's comments were particularly pointed and since she says it much more eloquently than I ever could, I have quoted her comments below from the Ancestry blog.

"...the gimmicky features AncestryDNA heralds as an alternative to a chromosome browser fail to meet the needs of genetic genealogists.

"From my experience and the testimony of many other customers, AncestryDNA’s attempts to automate discoveries do not clarify connections, but instead confuse them.
The problem at the core of these beautifully rendered but often dysfunctional features is the underlying methodology: genetic network theory.

"Genetic network theory has arguably less explanatory power than the segment triangulation methodology that genetic genealogists have employed for years. Genetic network theory claims that if a ring of people share DNA with at least one other person in the group and they claim descent from a common ancestral couple, then the known common ancestors are the source of the shared DNA. The flaw in the genetic network theory is that it fails to account for endogamy, pedigree collapse, and the possibility that the various members of a DNA Circle group share – and likely inherited their common DNA from – multiple family lines other than the one illustrated in the circle/discovery ecosystem. Segment triangulation, on the other hand, is a more precise methodology that alleges that in many cases a group of people all sharing the same DNA segment inherited that DNA segment from a common ancestor. Segment triangulation, while not flawless (some shared DNA segments are artifacts of ancient population bottlenecks and evolutionary changes in humans), seems far more logical than the genetic network theory AncestryDNA espouses.

"Even AncestryDNA staffer Dr. Julie Granka has conceded “The reality is that if you share DNA with members of a DNA Circle, it does not necessarily mean that you also share the DNA Circle ancestor. You could instead have another ancestor in common with the Circle members – for example, if the Circle ancestor is the sister of your great-grandmother. You could also share several different common ancestors with multiple members of the Circle – even if none of them are actually the ancestor of the Circle.” Segment triangulation with an onsite chromosome browser would permit customers in many cases to sidestep some of the imprecision that the genetic network theory-powered Circles and Discoveries engender. This would also provide, as customers of Family Tree DNA and 23andMe know, an extra layer of quality control, quality control that AncestryDNA desperately needs.

"However, despite Dr. Granka’s admission, AncestryDNA continues to deny customers the convenience of an onsite chromosome browser or even matching DNA segment data, preventing customers from employing more illuminating analysis on their site and almost forcing them to accept the questionable, if not erroneous claims their flawed systems all too often generate. This is not a breakthrough, but instead a clear backward step."

Shannon has it absolutely correct. What probably ticked me off more than anything was the fact that even though many of the posters commented about the lack of a chromosome browser tool, Anna Swayne a DNA spokesperson at Ancestry who was answering the comments on their blog for them deliberately ignored them all and answered the fluff comments instead. This is not the first time she has done this and to be honest she once removed some simple suggestive comments I made on the same blog regarding this same subject. I'm surprised that she hasn't purged any of these latest comments that do not put their product in a favorable light.



I'm not sure what hard headed individual is driving these decisions at Ancestry but this whole thing is starting to get a bit more frustrating than it has in the past. Ancestry's refusal to even acknowledge the obvious raises a lot of important questions regarding their service and commitment to their customers to provide a quality service. Glitz doesn't excitement me, accurate comprehensive genealogical research does.

My advice to my students is to continue to document your results as I have indicated in past classes and it would be best to ignore this new feature for the time being as I find it is a waste of time and energy. As I see it right now it has NO value or impact on your ongoing genetic genealogy research and I see no one in the future threatening Ancestry's patent on something that is this big of a big bust.

The Bottom line

From my Dad's DNA page ----

New Ancestor Discoveries BETA        
These are people who are not already in your family tree

    

We found you new ancestors—just by looking at your DNA.
Our latest scientific innovations make it possible to discover ancestors you never knew you had–just through your DNA. It's an entirely new path to finding your family story.


 
The reason they aren't in my dad's tree Ancestry is because . . . wait for it . . . they aren't my relatives and they never will be. I will give you an A- for the nice graphics (innovative programming for sure) but an "F" on everything else.
 
If you would have just poured your efforts into developing an eloquent and genetically effective DNA chromosome browser tool as you did with this genetic networking circle mess, you might have truly made a significant breakthrough that needed to be patented. I truly believe you really need to rethink this network circle mess you have just doubled down on and get some adult supervision in your DNA division of the company.