The abuse of ancestry testing is bad for personal genomics

By Razib Khan | February 25, 2013 2:20 pm

I have very little with which I can disagree with in this Mark Thomas piece, To claim someone has ‘Viking ancestors’ is no better than astrology. His conclusion:

Exaggerated claims from the consumer ancestry industry can also undermine the results of serious research about human genetic history, which is cautiously and slowly building up a clearer picture of the human past for all of us.

Many of the commercial companies plant stories in the media that sound exciting and seem scientific. But very often they are trivial or wrong, are not published in peer-reviewed scientific journals, and just serve as disguised PR for the company.

The only caveat I would offer is that the sort of confusions and misrepresentations that occur with Y and mtDNA phylogeography are dampened when you are looking at a million markers throughout the whole genome. This does not mean there are still no confusions and misrepresentations (e.g., the reference populations matter a great deal when you present someone as a linear combination of X populations, and that summary is still not reality as such, but an informative model). One alarming aspect of the trade in Y and mtDNA is that I’ve met several people who somehow believe that only these lineages are ancestrally informative. That is probably a function of the ease with which you can say someone is “descended from Niall of the Nine Hostages.”

Addendum: I actually asked Jim Wilson on Twitter if I could get a look at the raw results (not even raw data) for the claims made. One major problem when scientists have a go-to-media-first strategy is that things get out of hand very quickly.

CATEGORIZED UNDER: Genomics, Personal Genomics
MORE ABOUT: Personal genomics

Comments (16)

  1. Part of issue I have from the Y-Chromosome point of view is how certain people will take fairly old/broad SNP’s and try to shoehorn them into one particular category. The good example of course is the claims that U152 is Roman based on there been higher percentage in Northern Italy then in say England.

    U152 is probably on the order of at least 4,000 years old (2000-1700BC) in which case it’s a Bronze Age lineage that probably dates before the Italic and Celtic families seperated from each other (if you believe the concept of a common Proto-Italo-Celtic stage). Busby report on M269 has samples of over 20% for it in parts of France.

    Such an old SNP is going to have a mass of subclades by itself. If they had come out and said that based on large scale testing across Europe that a specific subclade was dominant in Italian U152 (highest diversity of haplotypes etc.) an had a TMRCA of around the 1st century BC/AD then you might take notice.

    There does seem to be a “courting the press” attitude. A good example of a “strange headline” was the one that claimed that “Bonnie Prince Charlie was really English!!!”

    The logic went as following:

    Stewarts (based on two illegimate ducal lines) are L744+/L746+ (subclade of DF41), their ancestral “myth” stated that they were originally Bretons -> Bretons originally came from Cornwall -> Cornwall is now part of England -> ergo Bonnie Prince Charlie == English


    As for Niall he does fit into that quasi-historical twilight after all it’s during the time of his sons that St. Patrick had his mission. Of course what’s ironic about the use of him as an example by the testing companies is that the SNP M222+ (and resulting STR haplotype) is probably a good bit older the Niall.

    What’s interesting though is discovery over last 18months of SNP’s upstream of M222, first DF23 and then DF49. We now know for example that there exists clusters that are:

    DF49+ -> DF23+ -> M222- (for example Wales/South-West England STR cluster)

    DF49+ -> DF23- -> M222-

    Hopefully the academics running massive studies across several populations will include some of these snp’s which will produce some interesting distribution maps.

    End of day Y-Chromosome is less then 2% of a man’s genome.



  2. Patrick Wyman

    This is a significant issue. As it happens, I’m teaching genetic history right now, and my students are reading Bryan Sykes’ Saxons, Vikings, and Celts. Sykes founded Oxford Ancestors, one of the many companies Thomas is targeting in this piece, and one of my students raised the issue of the extent to which people like Sykes are incentivized to easily equate cultural identity and genetic ancestry. Genetics is a powerful new tool for understanding the past, but it’s easy to go wrong even when you’re not powerfully incentivized to give people the imaginary past they desire.

  3. Sandgroper

    “This means we don’t have to look back much more than around 3,500 years before somebody lived who is the common ancestor of everybody alive today.”

    This is a self-evidently false statement.

    • villandra

      That idea actually comes from a group of responsible citizen genetic genealogists, including Ken Nordtvedt, and it’s based on sophisticated probablistic calculations. I myself don’t accept it. Certainly it is true that over many thousands of years, the number of male and female lineages people are descended from shrinks, ultimately to one line. Everyone now living is descended from one man and one woman who lived in Africa 65,000 to 85,000 years ago. Those two people were not the only people or the only mitochondrial and Y DNA lineages that existed in their time. Beyond that however I think that sophisticated probalistic calculations, done with actual calculus when Ken Nordtvedt does them, characteristically miss reality by a long mile.

      • Sandgroper

        I get what you’re saying.

        I’ve concluded the reason some of the commercial companies are doing this is that they are just giving people what they think they want, like tabloid journalists – ‘ya want rubbish, ya got rubbish.’ Either that or they have very weird ideas themselves.

        On the other hand, I don’t think Thomas is helping much. He needed to make his statements a lot more careful and precise than he did.

        • Sandgroper

          I digress, but I’m reminded of a funny story I was told by a Chinese friend – she’s very bright, successful and rather ‘worldly’, having lived in a lot of different parts of the world while growing up and having gone to university in the UK. She said in England people would always get together in the pub and get onto talking about genealogy, and everyone would be saying “I’m descended from Lord This and Lady That” and she’d say “I’m descended from a bus driver and a shipping clerk” which would always induce instant ostracism. A lot of Brits are still really locked into the “droit de seigneur” thing, which I think governs a lot of the (wishful) thinking and creates a ready market for bullshit.

    • villandra

      One of the strongest arguments for thinking that long enough ago we all
      had common ancestors, is that more than 400 years ago most people and
      their ancestors lived in small isolated rural communities, and within
      those communities, back a few generations everyone had the same
      ancestors. The best example of this I can think of is that my brother
      in law’s French Canadian grandmothers, from different Massachusetts
      towns and ultimately opposite sides of the St. Lawrence River, share
      over a third of their 17th century emigrant ancestry. I’m descended
      from small groups of families that married each other for centuries in
      small Massachusetts villages, and my father is descended from a single
      family nine times.

      Over 5,000 years, however, this would
      hardly lead to everyone on the planet having the same ancestors, because
      even though they moved over time, those isolated inbred groups trace to
      different groups of ancestors 5,000 years ago.

      It’s hard to
      argue that 5000 years ago everyone had the same ancestors, when we have a
      multitude of haplogroups that are far older than that.

      the other hand, one can see populations that lived in particular areas,
      such as those that carried U152, coalescing into large proportions of
      the population who clearly belong to a single Y DNA lineage, and that
      particular lineage probably actually dates to the Neolithic. The
      beginning of the Neolithic spread through Europe was a very good time
      for founder effects.

  4. Sandgroper

    “It has been reasonably estimated that around 5,000 years ago everybody who was alive was either the common ancestor of everybody alive today, or of nobody alive today; at this point in history we all share exactly the same set of ancestors.”

    Complete drivel. I can prove to him that he is wrong. Pot kettle black.

    • Other than a few isolated islands I think that his assertion is spot on. Of course by throwing that statement out without qualification he’s of engaging in the same astrology that he accuses others of. While we might all have the same set of people on our family tree we have them in different proportions.

  5. I am also coming across people who share significantly autosomal half-segments but “know” that we can’t be related because we have different uniparental markers.

  6. dtvmcdonald

    Not all claims about ancestry are bogus. For example take the “Somerled”
    claims of the Clan Donald. Disclaimer: I’m the one making the claims.
    We are claiming that R1a Y chromosomes which are 19-21 at YCAII and
    8-10 at DYS459 and generally match our “Somerled” haplotype … or
    are L176.1+ … “mostly” descend from John “Lord of the Isles” (d. 1386) ,
    and that Somerled himself likely was John’s male line ancestor,
    and that he was in fact a Viking descendant. The claim is that

    the DYS459 and L176.1 mutations occurred a few generations
    before Somerled, as did the “break” in what was likely an R1b Gaelic chief line. The data is on our web site for all to see.

    In any event, it is quite clear that this line is not Gaelic. It is clearly descended from Scandinavian roots.

    As to autosomal matters: I’ve tested about 4000 people using

    about 300,000 markers from the “standard” Illumina set. In some cases its possible to tell, quite accurately, what single group someone comes from. When someone is mixed, its vastly harder and usually (but not always) impossible to be sure of the exact mix. But in most cases a good general idea of the region(s) they come from can be figured out.

  7. I have a dim view of popular genomics as science.

    Think for example the way they keep insisting the Lemba are a Bantu x Israelite cross FFS.

  8. villandra

    There are actual Y DNA lineages in Britain that got there with the Romans. I’m probably descended from one of them. My Smith lineage belongs to a small, sparsely distributed, cluster with MCRA of around 1700 years, whose distribution in England, Scotland and the middle Rhine closely lines up with Roman forts, and is otherwise very hard to explain. I agree with you 100%. Critters like this are a major problem in genetic anthropology and they undercut the field. For instance a British columnist rightly took Moffat and Wilson apart, but their main argument is that genetic anthropology has neither interest nor basis in fact.

    With that said, I’ve personally been in touch with Moffatt and Wilson and Co, and they’re nuts. For instance, my brother in law’s lineage is Isles-Scottish, an I-M223 branch that originated in Galloway, Scotland around 300 AD, and was confined there until the Scotch-Irish migration. Jim Wilson excitedly told me on the phone that this is a “Pictish” group, and southwestern Scotland is the Picts’ center. I didn’t misunderstand a word he said, and he didn’t misunderstand what I was talking about. First of all, Picts as an ethnic group don’t exist; this is a military confederation that formed in the north end of Scotland in the last part of the 3rd century AD in reaction to the Romans. Second, the Pictish confederation spread across northern, central, and northeastern Scotland. It left southwestern Scotland completely alone because other groups successively held southwestern Scotland.

    Maybe it should also be pointed out that the SNP that Wilson and Moffat base their Roman soldiers claim on, S28/ U152, is a major central European SNP, not specifically an Italian SNP, and it spread with the Celts. It spread to Britain with the Celts. If it is scarcer in Ireland, that is important information – about Celtic history in Britain. However, it was pointed out on the Genealogy-DNA list at Rootsweb that Wilson et al’s claim contradicts other evidence on this subject. David Faux uses S28/ U152 to trace Celtic history in western Europe.

    It’s very true that Wilson and Moffatt often make outrageous claims with insufficient evidence and in the face of clear evidence to the contrary. The motive HAS to be to sell books and get funds, especially when scientific data is inadequately and exclusively presented in books one has to buy in order to see it. Oppenheimer and Sykes, working out of Oxford and maybe Cambridge as well, have been doing pretty much the same thing, only their work mostly bases conclusions on dangerously superficial, inadequate research.

    In the U.S., Spencer Wells with the Genographic project, has raised millions of dollars by claiming, over a period of years, that all of us are descended from “Phoenicians”, and he’s actually been running around the Mediterranean, spending great tons of money trying to prove this notion! Common themes better based in reality are that a major branch of the Neolithic migrated out of the nearby Near East and surrounding areas, along the Mediterranean coast, to Spain, Britain, western France adn Denmark, founding the Megalith culture; all western Europeans are substantially descended from them. It is also the case that Semitic peoples spread all over the Mediterranean coast, and early Neolithic haplogroups like E3b, and J spread with them. None of that has a whole lot to do with Phoenicians, but probably it is a substantial amount of what if anything Wells is basing his notion on. The actual geographic center of these people surrounded the Near East and the Balkans and extended across the Ukraine region. Phoenicians, on the other hand, were by the time they were one of a number of major powers in the region, an ethnic amalgam of Semites, southwestern Asiatic people and Indo-Europeans.

    Wells’ current Genographic 2 project isn’t even being done responsibly. For $200 apiece, people can get tested for some long list of SNPs that they can’t even learn what they are, though they do learn what they tested positive for, once they have tested. This test is known for instance to OMIT the M284 SNP, which is a major subclade of I-M223, a parent SNP of the Isles-Scottish haplotype, found at 2% incidence across Britain – maybe. There are important questions about its distribution and also its point of origin.

    Thank you for taking critters like this apart!


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar