About the SA Blog Network



Opinion, arguments & analyses from the editors of Scientific American
Observations HomeAboutContact

Finding My Inner Neandertal

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

Credit: Genographic Project

Odds are you carry DNA from a Neandertal, Denisovan or some other archaic human. Just a few years ago such a statement would have been virtually unthinkable. For decades evidence from genetics seemed to support the theory that anatomically modern humans arose as a new species in a single locale in Africa and subsequently spread out from there, replacing archaic humans throughout the Old World without mating with them. But in recent years geneticists have determined that, contrary to that conventional view, anatomically modern Homo sapiens did in fact interbreed with archaic humans, and that their DNA persists in people today. In the May issue of Scientific American, Michael Hammer of the University of Arizona in Tucson examines the latest genetic findings and explores the possibility that DNA from these extinct relatives helped H. sapiens become the wildly successful species it is today.

As Scientific American’s anthropology editor, I have an enduring interest in the rise of H. sapiens; and as longtime readers of this blog may know, I’m fascinated (you might even say obsessed) with Neandertals. So naturally I’ve been keen to find out how much, if any, Neandertal DNA I have in my own genome. Several consumer genetic testing companies now test for Neandertal genetic markers as part of their broader ancestry analysis, and after 23andMe lowered the price of their kit to $99 in December, I decided to take the plunge. As it happens, National Geographic’s Genographic Project had recently updated their own genetic test to look for Neandertal DNA, and they sent me a kit (retail price: $299) for editorial review, much as publishers do with new books. And so it was on a chilly Saturday in late January that I found myself spitting into a test tube for 23andMe and swabbing my cheek for the Genographic Project.

Of course the two tests look at far more than one’s Neandertal ancestry. 23andMe provides a wealth of health information, testing for variations in DNA that might affect disease risk and drug performance as well as mutations that could cause disease in one’s children. (I’ve decided to not look at those health results for now because I’m a worrier, although I may change my mind eventually.) Genographic’s test does not look for health information. Both tests trace one’s maternal lineage (and paternal lineage, for males) to beyond 10,000 years ago and reveal what percentage of one’s recent ancestry comes from various regions around the world.

Based on my mitochondrial DNA, which is passed down exclusively along the maternal line, both tests traced my maternal lineage to the Mediterranean 16,000 years ago (Genographic takes it all the way back to East Africa 70,000 years ago). I was interested to learn that my branch of the mitochondrial DNA tree is found in only about 3 percent of people in the British Isles, where my maternal great-grandmother came from.

My recent ancestry results were pretty much as expected—roughly half of my ancestry is European and half is east Asian; both tests further broke these regions down into smaller subregions. Intriguingly, the Genographic test hinted at a 2 percent Native American component, which I did not expect based on family lore. Yet this may not be what it seems. The Genographic site explains in a case study of a boy with a European mother and a Japanese father, “the 2 percent Native American actually reflects the fact that the ancestors of today’s Native Americans came from Asia, and reveals that there are still genetic patterns that they share from thousands of years ago.” The same may well apply to my result.

But on to the Neandertal business, because that’s what got me interested in these tests in the first place. Well, I’m pleased to report that both tests determined that I do, in fact, have Neandertal DNA. But interestingly, the tests differed rather significantly on how much I have. According to the Genographic Project, I’m 2.1 percent Neandertal (average for non-Africans, their site informed me). In contrast 23andMe found 2.9 percent Neandertal content. So what gives? Why the different results? As I pondered the results, other questions came to mind. Is there any way of knowing which stretches of one’s DNA have Neandertal origins? Will scientists eventually be able to tie one’s Neandertal DNA to observable traits? Might it be possible some day to pinpoint when and where my H. sapiens ancestors mated with Neandertals?

Genographic also found that I am 1.4 percent Denisovan, but they note on their site that they are still refining their approach to assessing Denisovan contributions. Frankly, I think it’s amazing that they assess this at all, given that the mysterious Denisovans were only discovered in 2010. I wonder how good they think their estimate is, and how it can be improved in the future. And I’d love to know how data from these personal genetic testing services are driving bigger research questions about the rise of modern humans (Hammer’s article includes a cool example of this.)

I’ll be reaching out to both 23andMe and the Genographic Project with these and other questions about their ancestry tests. Have you had your DNA tested for insights into your deep ancestry? Do you have general questions about how these tests work or what the results mean? Let me know in the comments below and I may pose them to the experts along with my own queries.

Kate Wong About the Author: Kate Wong is an editor and writer at Scientific American covering paleontology, archaeology and life sciences. Follow on Twitter @katewong.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 13 Comments

Add Comment
  1. 1. syzygyygyzys 2:47 pm 04/19/2013

    Interesting. Thanks for this.

    Link to this
  2. 2. Happy Hal 3:16 pm 04/19/2013

    Both my father, and I have always had troubles with hats, my father both with a homburg, and fedora, me, just the fedora. The problem was that even on special order (‘long oval’) they weren’t long enough oval, which has made me wonder if the ’round heads’ were Cro Magnon, and if the more northern humans had more Neanderthal in their DNA, because these hat styles were based on ‘Borsalino’, German/Italian head shapes, and our heritage was Viking, whom, I suspect, all had longer skulls, as described as Neaderthal. The net result was that blood circulation to the upper front skull was restricted, and our frontal hair died.

    Link to this
  3. 3. N a g n o s t i c 10:32 pm 04/19/2013

    Funny, the graphic says thal, the author says tal.

    Get it straight.

    Link to this
  4. 4. N a g n o s t i c 10:36 pm 04/19/2013

    I could’ve told the author she was half-East Asian for way less than $99. I’d have just pointed out her last name.

    Link to this
  5. 5. Califia 11:32 pm 04/19/2013

    I had my analysis done by 23andme. Just like you, I decided to forgo looking at the health risk portion of the results. My Neanderthal percentage is 3.1% (95th percentile). Fun! My ancestry results came back as anticipated, except the 10.3% Native American. I do have a great-grandmother from Mexico but she was not 100% native, rather a mixture of Spanish and Mayan. There must be another source of Native American in the family for my percentage to be so high. In exploring the possibility of further refining the source of my “Native American” DNA, I have come to understand that refinement is unlikely due to resistance on the part of Native North American populations to DNA sequencing. The results of these ancestry tests are dependent on the database to which a given sample is compared.

    Link to this
  6. 6. ErkkiRuohtula 5:55 am 04/20/2013

    Intriguing. How precise are the results regionally? I mean if I tried the test and it told me I’m Northern European, that would be fairly useless, since I already know that from genealogies going back about 200 years, and my country also has not had any significant immigration from outside Europe (except only in recent years). But can it pick out smaller regions within the continent? That would make it interesting, in addition to the “Neanderthal content”. I could not find useful information about this in the 23andme web site.

    Link to this
  7. 7. jonesge 10:36 am 04/20/2013

    Note to Califia and other readers – You said:

    “(1)In exploring the possibility of further refining the source of my “Native American” DNA, I have come to understand that refinement is unlikely due to RESISTANCE on the part of Native North American populations to DNA sequencing.”
    “(2) The results of these ancestry tests are DEPENDENT on the database to which a given sample is compared.”

    In regards to (1) … the vast majority of USA Native Americans already know which federally registered tribe they are officially registered in …. it’s not a matter of taking a DNA test!

    In regards to (2), you are confused on what an Autosomal DNA Test can and cannot show. The FIRST & LARGEST DEPENDENT Variable is how many Generations back in time you are seeking to identify an admixture event. Once you go back about 7 generations things get very dicey because of the Autosomal DNA process called Recombination. Once you go past this 7 Generation / 1 % Threshold you are in muddy waters unless you are a Professional Genetic Genealogist.

    Generation = 7
    Relationship to you = GGGG-Grandparents
    Aprox. % of Their DNA You Carry = 0.78%

    In summary, Autosomal DNA testing will pick up relationships reliably back to about the 6th or 7th generations, and sporadically beyond that.

    Link to this
  8. 8. rugeirn 7:55 pm 04/22/2013

    I wonder: Why jonesge thinks the reply is relevant to the passages cited? Whether or not you know what tribe you belong to has little or nothing to do with disliking the idea of taking a DNA test. As for #2, it is perfectly obvious that if you compare a DNA sample to database A, the result will be different than a comparison to database B if the two databases differ. So what’s the point of all that other irrelevant stuff?

    Link to this
  9. 9. afarensis74 6:41 am 04/23/2013

    I was wondering isn’t the percentage of DNA in line with what you would expect if both H.Sapiens and H.Neandertalis shared a common ancestor?

    What I’m asking is how do you know that the DNA that has been flaged is specific to Neandertals and not simply a shared common trait of an earlier common ancestor.

    Link to this
  10. 10. macgrant 3:48 pm 05/13/2013


    A very cogent point, and one that has been extensively considered in the literature. In short several lines of evidence point to the conclusion that the current genome is the result of (relatively) recent admixture, not a shared population structure that predates the H.n. and H.s. split.

    See Genetics, Vol. 194, 199–209 May 2013, and the papers it references, for more details.

    Link to this
  11. 11. LMB0217 6:54 pm 06/25/2013

    I am so pleased to read in your blog you did the 23andme test. I recently did it and got my results. I was most intrigued by the Neanderthal results (the inner geek in me–is there a gene for that?) since mine was 3.1% putting me in the 98th percentile of the population. I wondered if this has any statistical significance. My google search led me to your blog.

    Link to this
  12. 12. liveoaklinda 9:20 pm 01/14/2014

    I’ve done this too:
    love your blog.

    Link to this
  13. 13. kelpypaddy 1:16 am 02/25/2014

    note to jonesge: the tests determine “deep Ancestry” – (which goes back way older than any federal registration roles were ever developed, and indeed, is before there was a federal government or United states of America. Ancient Native sites discovered and dated by archaeologists at Meadowcroft Rockshelter (Penn.) Buttermilk Creek(Texas), & “On -your knees Cave(Oregon)date between 13,000 – 15,500 years ago. The people living there had no modern tribal names & for sure no Federal Registration. Califia’s statement is correct.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article