Ancestry.com Continues to Be Best In Class for DNA Ancestry Ethnic Composition

I've been very kind to 23andme in the past because of it's easy-to-use interface and it's candor when it comes to disclosing the weaknesses in its algorithm.  Nothing was worse than the other testing companies representing to people that their ethnic calculators were accurate, only to discover that the science was really just a guess.  Many authors have written entire chapters in books (this one quite funny!) that discuss these concepts.

But as 23andme prepares for its exciting and certainly in-demand upcoming IPO, it needs an update.  It needs to offer X chromosome searching, for one example.  

 And it's DNA ancestry has been lapped now, twice, by Ancestry.com.  Ancestry.com, who we've been harsh on before, now features INCREDIBLY accurate DNA ancestry estimates.  To tell you how far they've come, so fast, it'd be like going from horse and buggy to the space shuttle.  Their new tool is that accurate.

One user wrote me who hired a genealogist to complete a full pedigree.  That's 64 ancestors!  That user has a complete 64 ancestor pedigree now, well-documented with church and family records.  Of her 64 ancestors, 62 come from northeast Bavaria in Germany, 1 comes from Sweden, and 1 from the Czech Republic.  In other words, she's 96.8% German, from the Bavarian forest, and she's about 1.56% Swedish and Czech.

She got her ancestry results from Ancestry.com, and would you believe it said she is 96% German, from the Bavarian forest, and 2% Swedish, 2% Eastern European?  I mean, WOW.  Impressive.  Doubly impressive because, as we've posted before and many of you know, German and French ancestry is the hardest to call.

23andme still says this woman is German, Italian, British, Northwest European, etc.  In other words, it's pretty far off.  It has a ways to go.

Kudos to Ancestry for getting best-in-class and for cracking the German ancestry code.  We give major kudos to Tim Sullivan and everyone there for their hard work to become the absolute best.

Media Starts to Gets Home DNA Testing for Ancestry Right -- Thank God!

Kristen V. Brown is back with an excellent piece for Bloomberg, on home DNA testing, that is remarkably astute for a piece in the popular media.  

It confirms much of what readers of this blog have seen posted here repeatedly.  It's so good, it's worth quoting at length:

DNA is great at identifying familial relationships like parents, grandparents, aunts, uncles and even second and third cousins. Beyond that, it gets fuzzy.

The genes that make you a superfast runner or that identify you as Irish are less well-studied. The accuracy of any one test depends on the data your DNA is being compared to. One 2009 journal article said consumer DNA tests were akin to horoscopes exploiting the human tendency to hunt for patterns in meaningless data.

So what does it mean when a test says I’m 25 percent Irish?

It’s a misconception that these tests can tell you where your DNA was in the past. 

If a test tells you that you’re 25 percent Irish, what it actually means is that you are genetically similar to other people who are a part of the reference data set of Irish DNA that the company has collected. 

Because each company uses a different algorithm and data set, your results may vary based on which company you use. 

In other words: Take all this with one very large pinch of salt!

Meanwhile, in Slate, appeared another excellent piece by John Edward Terrell.  Here's the quote for you to read for yourself:

Whatever the motivations, the current popularity of commercial genetic profiling worries me for two reasons. One is that these companies may be promising results they can’t actually deliver. 

The notion, for example, that our genes can be used to trace our personal ancestry far back into the past—say, to Genghis Khan, the Emperor Charlemagne, or one of the pharaohs of ancient Egypt—makes little statistical sense. You may disagree, but to me this comes across as selling something more akin to snake oil than science.

What worries me most, however, is that companies offering personal genetic testing customarily seem to report back to those sending along a sample of their spit that they are a mix of different “ethnicities.” This is more than simply statistical nonsense.

We are happy that the mainstream media is finally getting it right, instead of publishing starry eyed pop-sci nonsense about DNA tests.

Don't ever forget: if you come from Central Europe (France, Germany, Italy, or nearby countries), or if you come from a country where there are simply insufficient samples (much of the rest of the world), these DNA tests will wipe your heritage off the map, by telling you that you are something else.  Basically, they're most accurate at the Continental level, unless you happen to come from an island in Europe with massive amounts of people getting tested, i.e., the U.K.

Read This If You're Curious About Your MyHeritage Ethnicity Results And You're Italian

MyHeritage ethnicity estimates seem to be THE WORST of all the major testing companies.

We've received dozens of emails from people of 100% document Italian heritage where MyHeritage says they are 0% Italian.  We've received three screenshots, which we will not share due to privacy concerns.

Something is amiss.  These people showed us documented Italian heritage, 100% Italian cousins, and some were born in Italy.  They show up on MyHeritage as Sardinian, Middle Eastern, Spanish, West Asian -- anything BUT Italian!!!

MyHeritage ain't getting it done.  We would demand a refund kind readers.  It's OK to come close.  As we've noted, all ethnic calculators are pseudo-science.

But MyHeritage isn't hitting the dart board in the bar next door.

Ancestry DNA Issues Revised Ancestry Estimates, Finds that Germans Exist

Judy G. Russell, the Legal Genealogist, is out with a fantastic new post on AncestryDNA's new ethnicity estimate percentages.

As she wryly notes in the opening, she is delighted to find out that they have discovered that Germans exist.

We've wrote about this before, as have others.  The major testing sites -- some of which are run by people who seem hostile to Germans (America's biggest ethnic group) -- have written Germans off the map.  23andme is particularly bad at identifying German DNA.  They disclose it too, but they bury it in the fine print.

We have been repeatedly depressed by newbies, who know from good paper records that they are a quarter German (or Swiss, or French, or Austrian) say, "duh, gee, duh, this unscientific website tells me I am really 21.2% English wow gee duh am I adopted?"  NO!  The science isn't there yet.  As Judy Russell says, "it's not quite soup."

And it STILL isn't quite soup.  This post focuses on Germans, but the major testing services have an equal problem with Italians, another major American ethnic group.  Poor Italians who get tested often end up with anything but Italian.  (Spare me your pseudoscience on how Italy has been invaded.  EVERY country has been invaded.)  Italy is a long country with many peaks and valleys, and for much of its history was an exporter of population to surrounding areas.  The testing sites need more samples to identify all the different permutations of Italians.

Bottom line, as we've said before, and as every credible scientist says - DO NOT TRUST the ethnicity estimates of the testing services.

DNA Testing for Heritage and Ancestry Is, Simply Put, Inaccurate

You go to take a cholesterol test, and your doctor, very thorough, sends you to four different labs.  One reports your cholesterol is 200, one says it's 180, one says 150, and one says 130...

After a car accident, you get go to four different body shops for quotes.  One says your car's paint color is taupe; one says it's sea blue; one says its ocean blue; and one calls it sea green...

You whip out four different rulers to measure your foot.  One says it's 12 inches; one says its 6; one says 8; and one says 9...

In all of these scenarios, you would make two conclusions:

1.  These test results (or body shops, or labs, or measuring sticks) are not that scientific!

2.  At least three -- and likely all four -- of these results MUST be wrong.

These parables sum up the world of DNA testing for heritage or "admixture."

We've said it before, and we'll say it again.  But today, Kristen V. Brown, a writer for Gizmodo, published an excellent piece discussing the snake oil that Ancestry.com, 23andme, Gencove, FTDNA, and other labs are selling.

Simply put, these labs cannot tell your ancestry.  I repeat, they cannot tell your heritage, or racial or ethnic admixture.  The science just isn't there yet.  And it might never be.

Brown details how she got four different results from four labs.

She also alludes to, but doesn't state um, confidently enough, about the concept about being confident about your known results.

It's what I jokingly (and longwindedly) call the:

"I was born in a tiny isolated village in the Swiss Alps that has never been invaded.  I know my mom and dad, and there's a video of my birth.  I DNA tested them and they are my parents.  I DNA tested my grandparents too, and they are my grandparents.  There were no affairs and no invasions in my town.  I know my great grandparents too and I am their spitting image.  The church records state I am Swiss going back to 1400.  But HELP, this DNA testing service said I'm British.  Am I British?"  (Or Indian or French or Dutch or whatever) problem.

NO, dummy, you're Swiss...

I for one, always read the fine print.  23andme, for example, states clearly that it cannot spot German (or French) heritage 92% of the time!  Germans make up the LARGEST PLURALITY of Americans.  Americans make up the LARGEST MAJORITY of those getting DNA tests done.  Thus, and I only say this half facetiously: these companies are engaging in the virtual ethnic cleansing of ethnic groups, wiping them off the genetic map, with their statements on people's heritage percentages, that are simply inaccurate.  

And 23andme, is, as far as we can tell, the most accurate lab!

Anyway, kudos to Kristen V. Brown and the people she interviewed for explaining it in her Gizmodo piece.  We suggest reading it.

Are Ethnicity Percentages and Ancestry Calculators from DNA Tests Accurate?

The media has blasted headlines this week that show an incredible ignorance of DNA testing for ethnic percentages.  One, which could have been pulled out of a 1980s tabloid for its ridiculousness, screeched, "Neo-Nazis are taking genetic tests and are deeply upset by the results!"

Neither of the two writers delved too deeply into the subject, and that is because shorthand reporting is easier.  As we've posted, again and again, most of the three major DNA testing sites disclose quite openly that their science is far from perfect.  For example, that if one is German, French, Dutch, Belgian, Austrian, or Swiss, that they cannot discern your ancestry 92% of the time.  (With the US having more people of German ancestry than even English or Irish -- that's a big deal).

In fact, it's quite common for someone taking a test from three different websites to receive three different results!  And as the post beneath this one shows, trying the 40 or so other "ethnicity calculators" available for free on Gedmatch produced...40 different results.

As I often say: if 5 different scales produced 5 (vastly different) weights, you would know that at least four of 'em don't work!  :-)

Anyway, for those looking for perspective, we shouldn't highlight the bad, so we've decided to highlight the good -- or the excellent, rather.

One of the best posts we have seen on the topic comes courtesy of a blog called The Legal Genealogist.  It's called, "Those Percentages If You Must" -- and is a Must Read for people curious about whether ancestry calculations from DNA tests are accurate.

It first, rather hilariously, goes into the various myths and misperceptions about DNA and human history.  Concepts like, "black Irish" or "I have some Native American in me."  Concepts that plague the world of pop-DNA-testing.

After it goes through the science (in easy to understand terms), it reveals what I have posted here time and time again:

DNA testing IS GREAT and REMARKABLY PRECISE for finding you cousins.  It CAN tell you if you are a third-cousin, once-removed (with that kind of precision).  If you don't know your heritage, and that cousin is, for example, 100% Native American -- then it follows that you too have a pair of Native American great-great-great grandparents.

DNA testing IS NOT good at ethnicity percentages and ethnic calculation.  As if anything could be so precise as to tell you that you are "4.2% Jewish."  The science is still just not ready for prime time, and many underrepresented populations, even in Europe, still confound the tests.

(As an aside, ancestry calculators should all produce nice and even results when people get back to the pre-travel era."  In other words, if you had 64 ancestors that were alive in 1500 AD, you should only see multiples of 1.56% chunks, right?  Since no one is half a human!)

The article succinctly concludes with:

DNA testing is a wonderful tool. It can connect us with cousins we’d have never found otherwise to help us reconstruct our family histories.

But in terms of “am I Native American?” “what tribe did I come from in Africa?” “am I 25% Irish?” No. No, no, no.  That’s the absolute weakest aspect of DNA testing. 

Indeed.  Well said.

In Praise of Roberta Estes and DNAeXplained.com

In a world of pseudo-science and echo chambers, a few blogs stick out for being mostly in touch with reality.  In the world of Ancient DNA, Dienekes, although less active than before, has pioneered much in the field of DNA, and still has many serious scientists who comment there.

In the world of DNA for Genealogy, one blog sticks out.  It is Roberta Estes' DNAeXplained.com.  Of all the blogs and websites dedicated to disseminating information about DNA, hers is consistently factual, science-based, and yet easy to understand. 

This scientist came across a few of her posts, and I daresay they are mandatory reading for anyone seeking a better understanding of their DNA.  Below are links and highlights:

"Determining Ethnicity Percentages"

Step 1:  Creation of the underlying population data base.
Don’t we wish this was as simple as it sounds.  It isn’t.  In fact, this step is the underpinnings of the accuracy of the ethnicity predictions.  The old GIGO (garbage in, garbage out) concept applies here. . . .

The third way to obtain this type of information is by inference.  Both Ancestry.com and 23andMe do some of this.  Ancestry released its V2 ethnicity updates this week, and as a part of that update, they included a white paper available to DNA participants.  In that paper, Ancestry discusses their process for utilizing contributed pedigree charts and states that, aside from immigrant locations, such as the United States and Canada, a common location for 4 grandparents is sufficient information to include that individuals DNA as “native” to that location.  Ancestry used 3000 samples in their new ethnicity predictions to cover 26 geographic locations.  That’s only 115 samples, on average, per location to represent all of that population.  That’s pretty slim pickins.  Their most highly represented area is Eastern Europe with 432 samples and the least represented is Mali with 16.  The regions they cover are shown below. . .

No matter which calculations you use relative to acceptable Margin of Error and Confidence Level, Ancestry’s sample size is extremely light. . . .
 

"Are You Native American?"

"having Haplogroup Origins and Ancestral Origins indicating Native American ancestry does not necessarily mean you are Native American or have Native American heritage. This is a very pervasive myth that needs to be dispelled. . . .

The good news is that more and more people are DNA testing.  The bad news is that errors in the system are tending to become more problematic, or said another way, GIGO – Garbage in, Garbage Out.

....

There are a very limited number of major haplogroups that include Native American results.  For mitochondrial DNA, they are A, B, C, D, X and possibly M.  I maintain a research list of the subgroups which are Native.  Each of these base haplogroups also have subgroups which are European and/or Asian.  The same holds true for Native American Y haplogroups Q and C.
In the Haplogroup Origins and Ancestral Origins, there are many examples where Non-Native haplogroups are assigned as Native American, such as haplogroup H1a below.  Haplogroup H is European...

One of the problems we have today is that because there are so many people who carry the oral history of grandmother being “Cherokee,” it has become common to “self-assign” oneself as Native.  That’s all fine and good, until one begins to “self-assign” those haplogroups as Native as well – by virtue of that “Native” assignment in the Family Tree DNA data base.  That’s a horse of a different color.