Not yet…
AncestryDNA launched its new feature, New Ancestor Discoveries, this week, and the pixels had barely started showing up on our screens before the outcry began.
The theory here is that AncestryDNA can take your DNA and compare it to the DNA of other people in the database and, based on the DNA alone, identify at least some of your ancestors.
AncestryDNA is actually promising just that:
Now with the easy-to-use AncestryDNA test, customers will have the unique ability to find their ancestors, who lived hundreds of years ago, using just their DNA. Only possible through the groundbreaking work of the AncestryDNA science team, New Ancestor Discoveries is a technical innovation that combines the latest in genetic science, new patent-pending algorithms, and access to AncestryDNA’s extensive database to push the boundaries of human genetics, and help people find ancestors from their past using just a DNA test, no genealogy research required.
“This is the biggest advancement in family history since we introduced our Hint feature, the Ancestry shaky leaf, which scours billions of historical records to automatically find new information about your family,” said Tim Sullivan CEO of Ancestry. “Now, through a simple DNA test, AncestryDNA is fundamentally revolutionizing the way to discover your family history, transforming the experience by making it faster and easier to go further into your family’s past, and instantly discover new ancestors you never knew you had.”
New Ancestor Discoveries are revealed through a unique combination of AncestryDNA results and the millions of family trees shared by Ancestry members. First, living cousins of each AncestryDNA member are found and organized into family networks, called DNA Circles, which bring together groups of people who are genetically related to the same ancestor. When a new AncestryDNA customer is connected into that DNA Circle, it’s likely they also share that same ancestor. As a result, it is now possible to simply take the AncestryDNA test and see the name of an ancestor from your family’s past appear in your DNA results.
“It is effectively a shortcut through time — you take the test today and we tell you who your ancestors were, for example, in the 1700s. You don’t need to research records or build a family tree — AncestryDNA now transports you to the past,” said Dr. Ken Chahine, SVP and GM of AncestryDNA. “It’s a combination of three things that allowed us to achieve this breakthrough innovation: 1) millions of family trees created by Ancestry members, 2) the fastest growing genetic database in the world, currently with more than 800,000 genotyped members and 3) a dedicated team of scientists who are pushing the boundaries of genetics and statistics to help people make family history discoveries in ways never before possible.”1
Um… let’s just say not so fast here.
Because what’s really happening here is that you’re being connected to specific “ancestors” by being linked to people who’ve tested with AncestryDNA and who’ve put their family trees online. The end result is that you’re being linked to folks who may or may not be descended from the same people you’re descended from at all. They may be linked to some collateral relatives of yours; they may be related to you in a different line of descent altogether that has nothing to do with the new “ancestor” who’s been “discovered.”
Case after case has already been posted where the new ancestor being discovered isn’t — cannot possibly be — an ancestor at all.2 Some collateral relative, maybe. Ancestor, no.
Don’t get me wrong here. I’m not jumping on the “go after AncestryDNA with a hatchet” bandwagon here. What AncestryDNA is trying to do is go after what is absolutely the Holy Grail of genetic genealogy: it’s trying to figure out how to identify ancestral lines using using combined DNA and family tree data on a macro scale using the computational power of modern technology.
If we ever could come up with a combination of carefully-analyzed DNA data and carefully-documented paper-trail genealogical evidence, we would have the best of all possible genealogical tools right at our fingertips. At that point, someone who tested and who matched in a scientifically and genealogically validated way to that documented DNA tree might very well discover new ancestors.
As any genealogist who’s ever tried to use DNA in family research can attest, even trying to do this matching of DNA data and genealogical evidence on the micro scale of a single family is not an easy task. It may turn out that, even with the best computers and scientific minds of the 21st century, doing it on the bigger scale isn’t possible.
But we can’t find that out if we don’t start trying.
So I applaud AncestryDNA for trying.
Where I part ways with AncestryDNA is over the hype — those promises made in the advertising quoted above.
Because we don’t yet have a databank of carefully-documented paper-trail genealogical evidence to link DNA results to, the notion that “people (can) find ancestors from their past using just a DNA test, no genealogy research required” or that you “don’t need to research records or build a family tree” is — to put it mildly — simply patently absurd.
Until we have that databank of carefully-documented paper-trail genealogical evidence, even if we were 100% sure of the genetic link between people who’ve tested, we’d still need to confirm the specific line shared with any match by a combination of traditional paper-trail research and DNA triangulation.
Unless we have that databank of carefully-documented paper-trail genealogical evidence, call me still waiting.
SOURCES
- “AncestryDNA Launches Revolutionary New Technology to Power New Ancestor Discoveries: Latest Breakthrough in Consumer Genetics Connects People to Ancestors Dating Back to the 1700s Using Just Their DNA,” News Room, MarketWired.com, posted 2 Apr 2015 (http://www.marketwired.com/ : accessed 4 Apr 2015). ↩
- See e.g. Roberta Estes, “Ancestry Gave Me A New DNA Ancestor – And It’s Wrong,” DNAeXplained, posted 3 Apr 2015 (http://dna-explained.com/ : accessed 4 Apr 2015). Also, Elizabeth Wilson Ballard, “AncestryDNA Has Now Thoroughly Lost Its Mind,” Diggin’ Up Graves, posted 2 Apr 2015 (https://digginupgraves.wordpress.com/ : accessed 4 Apr 2015). ↩
Face it, most people don’t have the desire to actually WORK at genealogy.
Ancestry is reeling those folks in with this.
You’re quite right about that, darn it.
ancestry is too concerned with adding new bells and whistles instead of fixing problems they already know about. Some of us wait forever for the email allowing us to download raw DNA data. I don’t take other people’s family trees on faith so why would I take ancestry’s DNA links?
And, oh, those the shaky leaves – my mother is related to the four children tested on ancestry’s DNA, but according to ancestry they are not all related to her nor to each other.
I’m not convinced they understand what the problems truly are, not for genealogists.
I received exactly one “New Ancestor Discovery” and have 2 dna 5-8th cousin moderate matches to people who do have this ancestor. However, I no more descend from this new discovery or any one in the trees of my 2 matches than a dog descends from a flea on its back.
Changing focus just a bit…in DNA Circles, and I have 22 Circles, Ancestry has now resumed the awful biographies, and not just the biographies, but a version of the new web design that is currently in Beta Trial. Ancestry has now given me freak show ancestors with 6 wives, 4 husbands, female ancestors with 25 children, bigamists, trigamists (polygamists)with children born before the parent, marriage to a spouse at age 7…you name it.
Unless you absolutely recognize the names of your “New Ancestry Discoveries” or someone in the trees provided or have a great dna match to someone with that ancestor, please tread very carefully. Even a great dna match can be very misleading as to the actual shared ancestor. This has the makings of a nightmare for adoptees and those who know little about their ancestors.
“Tread carefully” is what we’re all going to have to do here, for sure.
This the thing I fear the most about having spent that money for the test. I’ve found SO MANY horribly documented trees, and I’ve lost count how many times I’ve clicked “ignore” on those cursed shaking leaves, that I’m sure that my “New Ancestor Discovery” links will be just as horrific.
I take some comfort in having documented so many of my ancestor siblings (gr-aunts and gr-uncles) and know the general geographic area where my ancestors settled when they came to the US, but my adoptee grandmother is going to be a real fright fest.
I don’t think we’ll see the Tool of Our Dreams come to be until several generations of DNA can be documented and studied, and then backed up with a nearly flawless paper trail. I do have hope that (for example) with the 1950 US census records become available (in 2022), in combination with advances in technology, we may see better overall documented trees (because documentation requirements have become more stringent over time).
One could only hope.
It would be nice, but the reality is that autosomal DNA is always going to mean a lot of hard slogging through the paper trail trenches — and a lot of “why can’t I figure this out?” frustration. What’s being promised is a long way away if it can be done at all.
I agree completely. There are also various “intangibles” that tie into what was discussed in comments above – unearthing a family skeleton related to an “NPE” scenario.
This is why I have tested at 23andme and Ancestry (waiting on the Ancestry results). Based on some of the data I got from 23andme, and some “gut” instinct, I suspect that my adoptee grandmother was somehow biologically related to her adoptive father, but without the FACTS to back that up all I have is family mythology and “gut instinct”.
This is one of the most maddening things about Ancestry. Some of the badly documented trees I’ve encountered are on that elusive line that I’m hoping to unravel. For example, I know that my grandmother had (x) “sisters” and (y) “brothers”, and I know roughly when they were born (they were children of her adoptive father and his second wife). One of those faulty trees has the entirely wrong person listed as her sister (same name, just the wrong person entirely, wrong spelling, etc.
My only hope in being able to sort the “cruft” from the facts are tracing these shaky leaf potential “ancestors” myself to see if the timelines and locations match up (based on what I’ve been able to gather about that line from the tests at 23andme).
This is CRAZY making, but I enjoy it. It’s part History, part “Drama”, and a whole lot of detective work.
I don’t think many people have the patience to do that type of cross-checking on locations, surnames, dates, etc.
As long as the desire for instant gratification prevails, this may very well be a pursuit only for the true genealogy “nerds”.
It certainly is frustrating when a promising lead doesn’t have the info online and won’t respond and… I have a first cousin match at 23andMe who hasn’t responded… Sigh…
I am frustrated because Ancestry has given me this “potential ancestor” but they have not given me any tools to evaluate the premise. While I know this particular individual (female born in 1854) cannot be the actual ancestor, I do have a great great grandmother who was illegitimate and her father’s identity is unknown. Conceivably, “female potential ancestor X” could be related to this unknown ggg grandfather.
Or not … I suspect this potential ancestor’s husband was excluded as a “potential ancestor” not because of any DNA but because the two matches who pointed at the female as a potential ancestor spelled the husband’s last name differently (Robson vs Robinson) and one included a middle name. I suspect, but I don’t know for certain, because there is no one to ask. The computer that used the algorithm to spit this “clue” out is not taking questions. Grrrr!!!!!!
This from Dr. Steven Mount: “In summary, it is impossible to know the relationship one has to relatives who are discovered by virtue of their sharing a single autosomal segment of DNA. The “predicted relationship” is uncertain, and even the range is hard to be sure of. The extensive information provided by 23andMe is a very useful tool for genealogy, but it cannot tell you about relatives with whom you do not share any genetic material by descent. On the other hand, relatives with whom you do share genetic material by descent can be quite distant”. Ancestry seems to be offering the “ultimate DNA test” and I hope it does answer questions for some, but for most its just another tool in the DNA toolbox.
“This is the biggest advancement in family history since we introduced our Hint feature, the Ancestry shaky leaf,……..”
And we all know how accurate those are.
🙂
Judy,
Even if we had databases of the most “carefully-analyzed DNA data and carefully-documented paper-trail genealogical evidence”, the system would only work if the genetic ancestors were the same as the genealogical ancestors.
But they are not. The “official” family is not always the biological family. Anything could have happened. One spouse cheating on another. A criminal violation. An unreported adoption. Or maybe a child switched at birth.
It’s a nice idea, but it may ultimately prove to be its best at uncovering family embarrassments and secrets that the family wanted swept under the rug.
That’s not what people expect from it. But that’s likely what they’ll get.
Louis
Misattributed parentage (the “NPE” that Emily Aulicino defines as “Not the Parent Expected”) has always been an issue, Louis, and families have been discovering these issues as long as there have been families.
Thank you for pointing out Emily’s definition of NPE. I like hers better than the other interpretations of NPE such as Non-Paternity Event: http://www.isogg.org/wiki/Non-paternity_event
I do too. Fact is, if there’s a baby, it’s a pretty sure bet there was a paternity event. Maybe just not the one everybody thought there was…
We are just at the beginning of what I hope will turn out to be very helpful. As an old IT person, I can tell you that beta testing is usually fairly messy. You just can’t really test something out until you use it in the real world. So, let’s give feedback and be patient. My own four circles are merely variations on one line, the Farnsworths, which include several polygamist Mormons, in other words, a lot of ancestors, cousins, etc. My “New Ancestors” are cousin Farnsworths plus non-blood spouses. Darn, I’m still hoping for something new! I really do have more lines than just the Farnsworth! 🙂
I’d rather have them called something other than “ancestors” when they’re not “ancestors” at all, Mary. Beta or not.
Yes, no question about that! Surely, they must correct that false title! Also not sure why the non-blood spouse of a cousin appears.
Judy, you say “Until we have that databank of carefully-documented paper-trail genealogical evidence, even if we were 100% sure of the genetic link between people who’ve tested, we’d still need to confirm the specific line shared with any match by a combination of traditional paper-trail research and DNA triangulation. Unless we have that databank of carefully-documented paper-trail genealogical evidence, call me still waiting.”
Isn’t that the point…that these new hints always NEED CONFIRMATION? I’m thrilled to have more POSSIBLE clues. Would you really prefer that Ancestry not give new users this information? Or that they NOT try to get more people in their database…which makes for more connections for everyone?
For the life of me, I can’t understand all the complaining and whining. Your anti-Ancestry feelings are really showing (as always)!
I thought I’d pretty much made it clear that I give AncestryDNA credit for trying to do something we all know is very hard. I’m not sure how else you can read the statement that “I applaud AncestryDNA for trying.” But it’s important that people be told, and understand, that these things need confirmation. That’s exactly my point — don’t oversell, don’t overpromise.
False advertising to the max! Ancestry is obviously never going to admit that much of the info contained in their member pedigrees is inaccurate, having been copied off other peoples’ inaccurate pedigrees and put into perpetual circulation on the net through their site. To add insult to injury by advertising an ability to perform the impossible is truly unbelievable!
Wasn’t it P.T. Barnum who said “There is a sucker born every minute?” Only those who are uninterested in performing the true, difficult, expensive, and time-consuming research found in source documents would fall for this fairytale.
But I’ve found that many of our matches are hoping that someone will hand them their pedigrees with no exertion on their part. And to be honest, I’ve had my fill of spending weeks helping these people, who don’t even say “thanks”, at the cost of my own research. And, I’m not interested in handing off my hard-won research to the cousins who make a beeline to give it to Ancestry.
As my family group has found through the Family Tree Family Finder test, most of the matches there have very limited pedigrees that make it impossible to connect with our lines that we have documented back to the 1400 – 1600s. And the fact that we have a few brickwalls from the early 1800s means that we may never be able to connect with anyone who does have information that could help us with those families.
I plan to learn to use the scientific portion of the Family Finder testing to hopefully identify families in common; but when those matches lack the paper trails necessary to locate the common ancestors, it’s very frustrating. The families with whom we have linked have our same level of documentation and there has been no advantage for anyone to making the links. It has become an exercise in futility, and I have no plans to test any more relatives.
I understand your frustration, Sharon — but am never going to turn my back on a resource, and a database of this size is a resource I don’t want to ignore. I just don’t want folks to fall for the hype: DNA testing is a tool, not an answer by itself.
HI, Thank you for this information. I asked my brother to submit a DNA sample because I have spent the last 6 years trying to find my GG Grandfather’s parents. And what we found out is that it is most likely he was illegitimate. I have been struggling trying to find out how to even begin research to find his mother and father. We did both Ancestry and Family Tree testing. We found the connection to our “C” line through Ancestry, we assume that it is his mother and the Family Tree testing showed us the “B” line – assuming that is the father. I had hoped to find some names with the new information. But will look closely at the choices.
I really enjoy your daily blog and have learned so much. Thank you!!
Sincerely,
Ann
When I was a paying member of Ancerstry, I had 259 pages of matches on my DNA, how that I’m not a paid member I only have 95 pages matches. Has any one else noticed this with there results. If so do you know by any chance why this happened. Thank you
See When less is more, posted last October.
I’m afraid Ancestry is shooting themselves in the foot by overselling this technology. Let’s hope that people are too busy finding new ancestors to write about their successes. All I am reading about are the absurd results the technology is producing. I’ve no horror stories to report as I am 6-8 weeks away from getting my lab results.
Given that so many of the trees at Ancestry cite absolutely no sources for their information, how do you extract true facts from them? Look at the experience over at FamilySearch and trying to make sense of Ancestral File family trees. I think they may be onto something with their latest version of Family Tree using a collaborative model, reducing the number of trees, and simplifying the documentation process.
What Ancestry is doing with their family trees is akin to alchemy, turning lead into gold with the help of lots of computing power. I use Ancestry trees all the time in my research but always with a healthy dose of skepticism on my part. I only add individuals after I have verified the facts myself. It is surprising (or maybe not) how often you find basic errors in Ancestry trees that could have been found and corrected by PAF software in 1990. Then clone those errors 10,000 times and that is what you get with Ancestry Family Trees.
Hopefully the initial feedback from customers will force Ancestry to re-evaluate their product or at least tone down the hype.
I’d really like to see people told, flat out, that these are not ancestors, more often than not.
Regarding the “scientific” part – I am a proponent of exhuming American colonial founders to do DNA tests (not unlike the recent Richard III work), in order to provide the controls necessary to do well constructed tests, as expected by modern science methods.
Simply finding shared half-identical (or in the case of phased data, identical) chromosome regions among living people does not tell us who the actual MRCA was. Shared chromosome regions simply tell us that two people have a common ancestor not too long ago.
Many records are of dubious nature themselves, and there is no end of errors that have been discovered in 17th century documents. This is just not a problem of NPEs or undocumented adoptions, but just erroneous information, whether on purpose or by accident.
My own experience with AncestryDNA’s New Ancestor Discoveries is telling me that we are going to see lots of collateral lines passed off as “ancestors”.
Using Jeff Snavely’s AncestryDNA Helper Chrome extension I had already identified family groups, that I’ve used successfully in my research. The latest AncestryDNA NAD innovation (ahem) is still lagging behind my work with that method.
Exhuming colonial era bodies wouldn’t work really well. (a) Identification of any particular set of bones is problematic at best. (b) Only mtDNA would be a sure thing.
Huzzah! Huzzah! Keep up the good work.
The thought occurs to me that the usefulness of DNA testing for genealogical purposes is about to be reduced as a result of modern advances in the treatment of infertility. What with artificial insemination, surrogacy, etc. it will soon be remarkable if anyone tests out as a relative of either of their parents, or any of their own siblings.
It certainly complicates things for the future… but I suspect that, even with scientific advances, most babies will still come into the world in the usual way. 🙂
It’s their “science team” and we know we mustn’t argue with science.
This is, I suspect, less science and more math. And when you apply “the odds are” reasoning to insufficient facts, well…
I’m still waiting too, Judy!
Speaking of the math…last week I discovered what I expected to see with Ancestry using family trees in their algorithms–a “high” match for a person I didn’t match with at all at FTDNA. Of course, this “cousin” and I have the same ancestors named in one branch of our Ancestry trees. With collaboration, the DNA mystery will have to be unraveled at FTDNA and/or GEDmatch. No thanks to AncestryDNA’s lack of a chromosome browser and their sometimes false match results–who knows when and at what level of “certainty.”
I don’t have a background in genetics, but I do have a science degree and worked professionally in my field. I well know the importance of cross checking the data referenced in a scientific report before accepting the conclusions as stated. I do not understand how Ancestry continues to get away with taking our money, handing us a list of “matches” with predicted confidence levels, providing no supporting information as to how we are matching those individuals, and retaining any credibility. Minimally each match should include a description of the shared DNA (which segment(s) of which chromosome(s)) along with a synopsis of how Ancestry determines the various confidence levels for a match. In response to our repeated requests for chromosome matching information we were given DNA circles, which also lack any basic information to support if and how the folks in the circles are matching. Now we are being given “new ancestors” (since qualified as “possible new ancestors”) based on DNA circles based on DNA matches that still lack any basic supporting information. Unsupported matches plus unsupported family trees equals faulty DNA circles, small chance of the prediction of a true “new ancestor” for those of us stuck at brick walls, and many, many hours of wasted time. If from the beginning Ancestry had simply given us the basic information that supports each match, and a tool to identify which matches we had in common with our matches, think how much time could have been saved and how much progress may have been made. Instead of being provided a mechanism to participate in the process of identifying our ancestors, we have been put into a position of waiting for ancestry to identify those ancestors for us. Much as we all want to get through those brick walls, isn’t it the detective work that makes what we do so rewarding? Nothing can beat the enjoyment of assembling facts/clues, developing a theory, figuring out where to look for the next fact/clue to test that theory, then finally finding the document that proves the theory. In cases where that document cannot be found and likely does not exist, we are turning to DNA to test our theories, in hopes it may eventually provide strong support if not definitive proof (or tell us if we are on the wrong track). If Ancestry continues to refuse to provide the supporting data for matches determined between individuals, and those individuals do not upload their raw data elsewhere for comparison, then any conclusions regarding actual relationships based on those matches will remain unsupported. “Because Ancestry said so” is not sound scientific support.
We have some small grounds for hope that more information will be made available by AncestryDNA as we go on, Cheryl, but remember: you can always download your own raw data from AncestryDNA and use it to get more information at third-party sites such as GedMatch.
Thanks, Judy. Long ago I downloaded my raw data and that of close family members whose tests I administer and uploaded it to FTDNA and GEDmatch. I have also encouraged known cousins and suspected cousins to do the same. I have had success with both agreement to upload, and the corresponding results. Convincing others to upload is the uphill battle I continue to fight. This battle might become easier if more folks understood the importance of verifying the science behind the match. (Have to admit that some natural skepticism may have been passed down to me through my ancestors from the “Show Me” state.)
I couldn’t agree more: the more people understand what makes a match, the more they understand why we need to know.
The very fact that the DNA results are linked to “millions of family trees” that are available at their site is enough to boggle the mind. I am a long time ancestry.com subscriber, do not get me wrong–they do have wonderful resources BUT the trees I take with a grain of salt. Before incorporating any info in another’s tree I look at their sources–in many cases the sources are trees available in their database, all those LDS mini-databases still accepted as gospel [they are nothing more than a person’s contributed tree without any documentation], etc. If those sources are present I completely disavow the information. I have 2d-4th “cousins” with whom I cannot find a connection at all; also have some listed who have not submitted any family tree at all or else their trees are private. Although I do check all the “new” information I do not put a lot of reliance in the majority of it.
Verification is the name of the game for sure. I do use the trees, for hints, for tips. But not by themselves as proof. Not ever.
Judith Knight- You are absolutely right. I spoke to an agent from Ancestry and he told me the only DNA I will get on my husband is when another person from his so called family does their DNA at Ancestry. So I said ” So in other words we swab our mouths, send it to you and you enter my husbands name and info from his family tree and whoever has a name the same as his in his tree will be a match.
When you get your kit one of the first things they tell you on their instructions is to enter as much info as possible. Why is that if their matching your DNA? I feel they take that swab and throw it in the waste basket and enter your name into their data base and pull out straws.
We got 3rd and 4th cousins. Not one name of Bangs in the lot, Not a Bryant, not a Bisbee.
Then I sent his raw DNA to FTDNA and was to be free Ya right ,give them your DNA so they can use it and then you have to pay $39.00 to open anything that is locked. As with Ancestry anything that might be close is either, nothing posted to tree or locked. I have written to many so called 3rd and 4th cousins and they say I have noting for you or we’re not related even when it says you are a 98% match. Disappointing…..
I’m afraid that too many genealogists have the idea that (a) DNA research is easy (it’s not) or that (b) DNA research is free or should be free (it isn’t and it can’t be).
I received my AncestryDNA test results two days ago, exciting moment. The ethnicity section was interesting reading, but I got the feel a lot of it is based on mathematics and statistical probability theory. Should I be surprised at being 38% West European and 21% Irish but only 30% GB when in my tree 99% of my ancestors are GB ?
On the matches section, for the six DNA tree hints, whilst these correctly identified common ancestors, the DNA match with the other person was never more than moderate.
Conversely for the 34 cousin matches with closely match DNAs (high or good confidence), none of their trees that I could view had any ancestors remotely close or shared with mine. Again I think this is just a statistical exercise.
So I am left a bit underwhelmed by the whole process at the moment. Comments anyone ?