March 16, 2016

H. heidelbergensis is Neanderthal ancestor and not 'Denisovan' cousin


The unprecedented sequencing of a small fraction of the autosomal DNA of Homo heidelbergensis from the Sima de los Huesos of Atapuerca proves that they are in direct ancestral line to H. neanderthalensis and not particularly related to Denisovans.

Matthias Meyer et al., Nuclear DNA sequences from the Middle Pleistocene Sima de los Huesos hominins. Nature 2015. Pay per viewLINK [doi:10.1038/nature17405]


A unique assemblage of 28 hominin individuals, found in Sima de los Huesos in the Sierra de Atapuerca in Spain, has recently been dated to approximately 430,000 years ago1. An interesting question is how these Middle Pleistocene hominins were related to those who lived in the Late Pleistocene epoch, in particular to Neanderthals in western Eurasia and to Denisovans, a sister group of Neanderthals so far known only from southern Siberia. While the Sima de los Huesos hominins share some derived morphological features with Neanderthals, the mitochondrial genome retrieved from one individual from Sima de los Huesos is more closely related to the mitochondrial DNA of Denisovans than to that of Neanderthals2. However, since the mitochondrial DNA does not reveal the full picture of relationships among populations, we have investigated DNA preservation in several individuals found at Sima de los Huesos. Here we recover nuclear DNA sequences from two specimens, which show that the Sima de los Huesos hominins were related to Neanderthals rather than to Denisovans, indicating that the population divergence between Neanderthals and Denisovans predates 430,000 years ago. A mitochondrial DNA recovered from one of the specimens shares the previously described relationship to Denisovan mitochondrial DNAs, suggesting, among other possibilities, that the mitochondrial DNA gene pool of Neanderthals turned over later in their history.

Some articles that describe the findings:
at Público (in Spanish)

Matthieson also found that the Sima de los Huesos hominids were closer to Denisovans and Neanderthals in mtDNA two years ago. But this sequencing of their nuclear DNA puts them much closer to Neanderthals instead.

Prüffer et al. found in 2013 that Neanderthals form a cline with "Denisovans" in nuclear DNA but not in mtDNA, in which they are closer to us. This one is a very interesting read for background, as it explores in great detail the various possible scenarios.

That "Denisovans" could be closely related to H. erectus (a catch-all term for most archaic populations, particularly in Asia) has been considered as very possible before (Waddell et al. 2012) but there is no genetic confirmation so far, neither strong rejection. Getting DNA from such ancient specimens is considered a breakthrough and this partial sequencing of 400,000 years ago is believed to be within the very limits of absolute possibility.

[Conclusions edited on Mar 19th because I got it all wrong and don't wish to keep confusing anybody else. Instead I listed several relevant background studies, judge yourself].


  1. As reasonable an interpretation as any. Thanks for that.

  2. I don't get it, why Homo erectus? It just says more than 430 000 years ago. Denisovans are still the sister group to Neanderthals in nuclear DNA, within the Neanderthal-Denisovan-MH clade. That can hardly include Homo erectus. Rather something like the Narmada human or Dali man.

    Denisovans did have minor ancestry from a much more divergent human, that could be some Homo erectus admixture.

    1. Damn! You're absolutely right. Conclusion: I should not give opinion on stuff I have not meditated enough about (like this "quickie"), as I got all the info mixed up and therefore the conclusions. Thank you, Capra: I'll delete the random thoughts presented as conclusions. I hate it but when you're right and I'm wrong that's it. Thank you for the correction.

    2. Edited... who said what? Not me! ;)

      Seriously: I think that I particularly appreciate, in spite of the initial instinctive resistance, when someone proves me wrong as easily as you just did. It does not happen often but it should happen now and then, so thank you very much for pointing the obvious error in my logic.

    3. So what's the possible conclusion then? These H. heidelbergensis are close to Neanderthals (also in physiognomy) but carry Denisovan mtDNA. H. Neanderthalensis are closer to us than to Denisovans in mtDNA but not autosomal DNA... Do you think that there was a proto-sapiens population migrating to Eurasia that got admixed with pre-existent "H. erectus" populations producing Neanderthals? What are the "Denisovans" then in classical taxonomy? Shouldn't they be something like Solo man?

      When we talk about H. erectus (sometimes called H. ergaster in Africa but not in Asia), we're talking of a huge continuum of human populations (with clearly growing brain size along time, up to very close the modern human average: 1200 cc vs 1350 cc) and I'd dare say that H. erectus is not a different branch as such but rather the wider catch-all clade of all humans, except some of the very early branches (habilis, georgicus, floresiensis). Dali is generally considered H. erectus for example, Narmada also appears as such in some papers, even if it's clearly an outlier with greater brain capacity and Neanderthal-like traits.

      The issue of "Denisovan" as H. erectus has been raised before anyhow. Are Neanderthals pre-sapiens with strong "Asian Erectus" admixture? Is that the reason why Neanderthals form a cline with "Denisovans" and Heidelbergensis carries "Denisovan" mtDNA?

      It would also imply, I guess, that "Denisovans" and their plausible cousins like Solo man (a good candidate for "Denisovan" introgression in modern humans) carried Neanderthal admixture. Something I've been wondering since this "Denisovan" admixture was detected in Oceanians is whether Neanderthal admixture is not being counted twice: once as Neanderthal proper and the other as "Denisovan", i.e. that both admixture scores should not be added but rather considered in overlap (although no confirmation till now, also no strong rejection).

  3. Aargh, it either ate my previous comment or it will show up twice.

    You are right, Homo erectus seems to be a catch-all for "not later western humans", i.e. both early forms and late eastern forms. Which is very annoying, because in these analyses how are we supposed to tell whether Homo erectus means "share features of later eastern humans" or only "lacking features of later western humans"?

    I have been doing some reading up on the subject and really there seems to be little overall agreement or clarity, so I am as confused as ever. There are not firm dates for Solo Man either, could be 50 000 years ago or could be 500 000 years ago, so it is not clear whether Sundaland was occupied by pure late Homo erectus (who could probably not be the Denisovans) or some hybrid as you suggest.

    I have found some speculations by palaeoanthropologists on the identity of the Denisovans and mostly they seem to focus on Narmada and Dali/Jinniushan/Maba. But that does not narrow things down very much because there is no agreement on what those people were either.

    If you have seen the latest paper on Neanderthal and Denisovan admixture they break down the archaic ancestry in Melanesians into certainly Neanderthal, certainly Denisovan, and some which is ambiguous. So that should be able to detect any overlap or twice-counted material.

    1. Yeah, the concept H. erectus is both a problem and mediocre solution to our inability to properly categorize archaic humans. There are just not enough remains and certainly not enough academic consensus to change the terminology today anyhow. But, on the other side, we are learning that strict species barriers within the Homo genus did not exist. In a very lax sense we are all H. erectus, albeit primarily of a very specific and distinctive branch we call H. sapiens. The case of Neanderthals may well be the same.

      In any case H. erectus almost certainly should be distinguished in various categories, depending for example in brain capacity (Solo man and others were close to us and Neanderthals, within the low range of present variability, while early H. erectus were closer to H. habilis instead).

      I mention Solo man because he (his population) could well be the missing link in the Oceania-concentrated "Denisovan" admixture. The most parsimonious explanation is clearly that there were "Denisovans" in Sundaland or nearby areas (Wallacea or is it India, Indochina?) who were the ones directly implied in the admixture episode (and not Altai "Denisovans" quite obviously).

      "If you have seen the latest paper on Neanderthal and Denisovan admixture they break down the archaic ancestry in Melanesians into certainly Neanderthal, certainly Denisovan, and some which is ambiguous".

      I haven't seen, and that says much about my growing inability to keep up to date. It looks interesting because it appears to address my doubts on the matter, so thank you for the heads up.


Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).