28 February 2013

Another "All Your Yesterdays" entry: Denisovan, or "Polar Neandertal"

They've extended the deadline for the All Your Yesterdays contest, so I've decided to do a couple more entries. I started this one years ago as a Neandertal restoration. Since that time, new genomic discoveries showed that the speculative pigmentation was incorrect. But, other discoveries identified a new candidate for the subject matter!

Known from a few scrappy pieces, the Siberian Denisovans (Homo sp. or Homo sapiens ssp., depending on how large you like your species) are a true challenge to reconstruct. We have their entire genome, but know almost nothing about their anatomy. The few fossil elements we have are not morphologically distinct from Neandertals (Homo neanderthalensis or Homo sapiens neanderthalensis) or humans (Homo sapiens sapiens).
But the genomic facts are highly intriguing:
  1. Some Oceanian humans have inherited up to 6% of their nuclear DNA from Denisovans (with the highest ratios in Meganesia [Australia and New Guinea]). 
  2. The nuclear DNA indicates a common ancestor with Neandertals, shortly after the split from proto-humans.
  3. But the mitochondrial DNA indicates a motherline that branched off much earlier. (Possibly Homo erectus?)
  4. Genes for pigments are consistent with dark skin.
Here I've imagined a Siberian Denisovan as a sort of "polar Neandertal". As with polar bears, his skin is dark, trapping heat, but his pelage is light, allowing for camouflage against the taiga and tundra. He is the last of his kind — his southern kin mixed with the strange, baby-faced people who keep invading from the west. But he does not welcome them. He will fight to his death.


  1. Hi Mike,
    Cool reconstruction, but I am sorry to say, we actually have a lot of information on what Denisovans could have look like, thanks to the recently published high coverage genome. There is a freely accessible supplementary table to the paper with inferred pigmentation patterns at odds with your reconstruction (http://www.sciencemag.org/content/suppl/2012/08/29/science.1224344.DC1/Meyer.SM.pdf). They were inferred to have dark skin, dark hair and brown eyes (and not freckles!). Not bad for a pinky finger.
    I also imagine them much more human than your reconstruction.
    Josie Hammet got it right in an illustration for a piece in the Guardian
    Good luck with your entry though.
    The Denisova genome paper is:
    Meyer, M., Kircher, M., Gansauge, M. T., Li, H., Racimo, F., Mallick, S., ... & Pääbo, S. (2012). A high-coverage genome sequence from an archaic denisovan individual. Science.

    1. He is actually pretty human-looking once you get past the expression and the hair distribution.

      As to the other point ... hmmm ... in that case I'll posit a mutation elsewhere in the regulatory network, previously unidentified due to its uniformity in humans! And maybe I'll place it on the Y chromosome, currently unknown for Denisovans! Yeah!

    2. Aha!

      "However, we note that predicted phenotypes in Cerqueira et al. did not show a strong concordance with observed phenotypes in modern humans (total percentage of agreement = 59%) and the reliability of the inference for the Denisovan individual (belonging to an archaic group that might have had other pigmentation-associated SNPs not present in modern human variation) may be even lower."

      So event at best we aren't even 60% sure of the inferred pigmentation.

    3. FWIW, 23andMe predicted my eyes to be brown. (They're green.)

    4. A polar Neandertal. Maybe a good guess?
      The Denisovan genes is not only found in Australia etc.
      My I1 haplogroup (Norwegian) have 4.2% Denisovan genes, according to Geno 2.0

    5. There are a few problems with what you say.

      1) I-M253 (formerly I1) is a Y-DNA haplogroup. To the best of my knowledge, there is no Y-DNA known for Denisovans. (The only sampled individual was female.)

      2) All known human Y-DNA haplogroups (including the recently discovered A00 branch) originated after the initial split between humans and Denisovans.

      3) If someone did have a Denisovan Y chromosome, it would be over 95% Denisovan, since the Y chromosome does not recombine. (The only exception is some [noncoding] parts of the telomeres, which do recombine with the X chromosome. Although the X chromosome has very little Denisovan DNA even in native Australians and New Guineans.)

      Are you sure you're reading the report right?

  2. This comment has been removed by a blog administrator.