Please ignore mtDNA and Y chromosomal haplogroups

By Razib Khan | May 20, 2013 11:03 am

Guess what, we’re related! Credit: Wapondaponda 

This is a public service announcement. If you are a user of direct-to-consumer personal genomics services, please do not pay any attention to your mtDNA and Y chromosomal haplogroups. Why? Because they hardly tell you anything about your individual ancestry. What do I mean by this? Your mtDNA comes down from your mother’s-mother’s-mother’s-mother… and similarly for your Y chromosomal lineage if you are a male. These few individuals are not any more likely to contribute to your ancestry than all those multitudes and multitudes who do not contribute to your mtDNA or Y lineages; also known as almost all your ancestors! What you should pay attention to are your autosomal results. Inferences made from most of your genome. These results may be more difficult to parse, but difficulty is no sin, and elegant ease is no virtue, in this case. That’s because you are interested in your ancestry, not a convenient interpretable story.

Of course I am not saying that mtDNA and Y chromosomal haplogroups are useless. They are useful for population scale phylogeography. But please don’t make inferences about yourself from one data point. At least in most cases.

CATEGORIZED UNDER: Personal Genomics
  • Paul Conroy

    If you are interested in health or general traits about yourself, then you would do well – for the most part, not entirely – to ignore uni-parental markers – NRY and mtDNA.

    However if you are interested in Genetic Genealogy – tracing your ancestors through DNA – or population level history of some of your ancestors, then Y-DNA is crucial, as it forms an unbroken trail back through the generations.

    If for instance your last-name (aka surname), is “Smith” it can reveal exact relationships between people with the same name, and exclude the millions of false leads – when you are searching for your direct male line ancestors.

    Likewise for widely dispersed Y-DNA haplogroups – like R1a1a, found from Scotland to Bengal and across Central Asia, Russia and into NW China – it can reveal the hidden structure within the millions of men who carry it, based on SNP analysis, and point towards specific population histories, migrations, invasions and the like.

    Full Disclosure:
    I have recently launched a genetic startup – – which tests the FULL Y Chromosome. We are planning a range of other products in the coming months, stay tuned.

    • razibkhan

      or population level history of some of your ancestors,

      i think the key is population level. most people don’t understand this, and think that uniparental lineages are substantively relevant on an individual scale. they’re not. though they may be spiritually relevant.

  • Anthony_A

    In general, you’re right, but there will be special cases where MtDNA tells you something. I’m of mostly European descent, with supposedly 10-13% American Indian ancestry. Earlier estimates from 23andMe said more like 6%.

    Knowing that I’m MtDNA haplogroup A2 tells me *which* ancestor was Indian.

    • razibkhan


    • hypnosifl

      Similarly, my mother’s side of the family is from Argentina but the family doesn’t know anything about any non-European ancestors, and 23andMe’s overall estimate for Native American ancestry was 1.7% which was might be within the margin of error, but my MtDNA came back as haplogroup C which I take as much better evidence that I do have some ancestry on that line.

      • razibkhan

        1.7% is not within the margin of error from the number of markers they have from what i know. also my own runs of admixture, etc., seem to indicate 1%> of this in europeans is very informative.

  • razibkhan

    the genetic genealogy community is different, and a very small minority. they actually know what they’re doing, and the relevance of what they’re doing. they are not the 100,000+ average customers of 23andMe etc. who are totally confused by all this.

    • notageneticist

      I hope I am not misinterpreting your beliefs regarding DTC Y-DNA &mt-DNA DNA Tests for “individual level” recent ancestry purposes (<500 YBP) at 23andme, FTDNA, NG Geno 2.0,, etc.

      You seem now to not favor "individual level" testing for Y-DNA and mt-DNA to "more easily" determine some major and informative recent ancestry starting points and seek instead to convince uninformed (your words – " totally confused") consumers to initially embark on Autosomal testing. You said: "What you should pay attention to are your autosomal results. Inferences made from most of your genome."

      Mark Thomas has wrongly painted ALL lay Genetic Genealogists and DTC DNA Ancestry Consumers as "Genetic Astrologers". Now you add to the mix and say we are "Totally Confused" and that Y-DNA & mt-DNA are merely "Spiritually Relevant."

      Your arguments did not mention that there are many weak or non-existent "Reference Population Groups" for consumer Autosomal DNA Analysis.

      In regards to "Totally Confused" … I hope you would have provided some information or studies to back up that or at least compare DTC Ancestry Testing to Parents who are confused about my serious DNA Prenatal Testing for DS and other diseases.

      • Andrew Lancaster

        FWIW, as a genetic genealogist:
        1. I agree that Y DNA (not mt DNA) is very useful in genealogy. It is certainly a point worth making which adds something to what Razib is saying.
        2. OTOH, while I agree Mark Thomas over-generalized, I also disagree that genetic genealogists generally do not very often (not always of course) misunderstand the importance of uniparental markers in determining something about “who they are”. You do not have to look at a GG forum for long to see a bit of “astrology”. 😀

  • Karch_Buttreau

    Thanks for your post. I totally agree with it and gosh, if one just reads comments at 23andMe, it’s clear that the average person there has no clue and is being somewhat misled. For $99, still great value, but it’s bad enough there that I had to stop accepting share invites and hide myself from that relative finder thing… I just got tired of explaining things to the neophytes.

    I have 3 main beefs with this consumer level genetic service stuff. 1) the stuff you just wrote about – the presentation of population study data as individually relevant; 2) the absence of phasing which makes for lots of false positives with relatively small amounts (like 0.1%) of possible matches; and 3) the presentation of a single SNP as establishing the phenotype for a polygenic trait (memory, height, muscles…).

    It all falls on deaf ears… a service will present whatever grows their business the best.

  • GuestOfGuests

    I kind of disagree, especially for populations with an effective history of endogamy… An individual belonging to an endogamous family will be much more likely to find his haplogroup all over the paper trail.

    For instance, I get 101 J1e cousins… This is the biggest clade amongst all of my matches on RF as well as my own haplogroup, the second most common one is E-M123 with 97 cousins.

    • Andrew Lancaster

      I think you are right that uniparental markers can be useful for population genetics. This is because they spread in ways which will be different from the average chunk of DNA in non-random ways. (Think of the non random way in which Y DNA will spread in periods of technical or political change, or that in which mt DNA will tend to represent the opposite more stable extreme.)
      Maybe Razib’s post could be taken as disagreeing with that although I think it is just not what he is writing about. Anyway it seems to be to be a point worth making.

      But it remains true that what is useful as a single statistic in a database, might be meaningless when looked at in isolation. So looking at Y DNA in a whole country might be useful, but looking at one single case of a certain haplogroup might not be useful at all.

  • Tyrone Bowes

    If you have had a Y-DNA test then look at the surnames of the people you share a common male ancestor with. You will notice that they are all different and this is because those surnames arose among a group of related males living in a very specific area. Find out where the surnames that appear in your Y-DNA results originate and you’ll reveal a common area of association or literally where your ancestor lived when he picked his surname (typically 1000 years ago)

    • Paul Ó Dubhthaigh

      Of course it does depend on when surnames are adopted in a region, we here in Ireland are bit spoilt in that choice given majority of irish surnames have been around in some form up to 1000 years.

      Razib regarding large scale population studies I see Flanders study have published new report they finally tested for R1b-L21 which they had left out the last time around. They also tested for R1b-Z195.

      “Increasing phylogenetic resolution still informative for Y chromosomal studies on West-European populations”


  • badJim

    It’s possible that my ancestry is so simple that the Y-chromosome (Irish) and MtDNA (Swedish) specify nearly 90% of my genetic inheritance, assuming that the Irish and Swedish populations have been pretty isolated since the end of the last ice age.

    There may be some surprises when I get the results from 23andme in a month or so, but I’m not optimistic.


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