Yes, Neanderthals are us!

October 23, 2014 • 9:03 am

by Greg Mayer

In a paper published today in Nature, Qiaomei Fu and colleagues report a high quality genome sequence derived from a 45,000 year old, anatomically modern human femur found in western Siberia. “Ust’-Ishim Man” has provided the oldest known genome of an anatomically modern human (there are earlier genomes of archaic humans).

Usht'-Ishim Man's femur (from Nature).
Ust’-Ishim Man’s femur (from Nature).

So, why is this interesting? First, it is a marvelous technical achievement to be able to get a high quality sequence out of a bone of such great age recovered from a riverbank. Kudos to Fu and her colleagues for this achievement. Second, Ust’-Ishim Man proves to be very interesting phylogenetically. While definitely non-African in his genetic affinities, he appears to be equidistant from both modern Europeans and modern East Asians. Fu et al. interpret him as being at or near the point in time when the split occurred between these two branches of humanity, making him part of the lineage of modern humans that had left Africa, but had not yet split into European and East Asian sub-lineages.  Third, by being able to identify the genetic differences between Ust’-Ishim and modern man, they were able to estimate the mutation rate in both the nuclear and mitochondrial genomes. The autosomal mutation rate was about .5X10E-9 per site per year, the Y chromosome mutation rate was higher, about .75X10E-9 per site per year, and the mitochondrial rate much higher, about 2.5X10E-8 per site per year. These rates and their mutual relations are about exactly what we would expect, but it’s nice to have fairly direct estimates, over a long time base, to confirm estimates based on short term de novo mutation studies and comparison of contemporaneous sequences.

And, finally, there’s what we learn about the interbreeding between anatomically modern humans and Neanderthals. As Jerry, John Hawks, and I have all argued before (and as I recently summarized at The Dish—  see the update at end of Andrew’s post), Neanderthals and early non-African anatomically modern humans (along with Denisovans), were all parts of a group of interbreeding populations in nature, and thus were all members of the species Homo sapiens. Ust’-Ishim Man’s genome is about 2% Neanderthal, just like modern Europeans and East Asians. This means that the level of admixture characterizing modern populations was already in place by 45,000 years ago. This is not too surprising. Neanderthals were going or gone by about then, so whatever interbreeding occurred should have (mostly) occurred by then. So Neanderthals are us.

Figure 5: Regions of Neanderthal ancestry on chromosome 12 in the Ust’-Ishim individual and fifteen present-day non-Africans. The analysis is based on SNPs where African genomes carry the ancestral allele and the Neanderthal genome carries the derived allele. Homozygous ancestral alleles are black, heterozygous derived alleles yellow, and homozygous derived alleles blue. (From Fu et al. 2014).
Figure 5: Regions of Neanderthal ancestry on chromosome 12 in the Ust’-Ishim individual and fifteen present-day non-Africans. The analysis is based on SNPs where African genomes carry the ancestral allele and the Neanderthal genome carries the derived allele. Homozygous ancestral alleles are black, heterozygous derived alleles yellow, and homozygous derived alleles blue. (From Fu et al. 2014).

But that’s not all. Modern humans are separated by some tens of thousands of years, and thousands of generations, from the time our forebears interbred with one another. During this time, recombination between the chromosomes of our anatomically modern and Neanderthal ancestors will have broken up the originally contiguous chromosome segments, dispersing the two sets among one another. Since the great majority of our genome is from anatomically modern ancestors, this will most easily be seen in our Neanderthal genetic component, which will become scattered throughout the anatomically modern part. This is exactly what is seen in the 15 modern non-African genomes in the figure above– the yellow and blue Neanderthal segments are scattered throughout the black anatomically modern background.

But when interbreeding first occurs, the two genomes will be separate. The first “hybrid” child will have one set of Neanderthal chromosomes, and one set of anatomically modern chromosomes. When that child produces gametes, its chromosomes will undergo crossing over— an exchange of chromosome segments– during meiosis, so that its children will receive a chromosomal gemisch: each chromosome will consist of alternating stretches of Neanderthal and anatomically modern parts. In subsequent generations, crossing over occurs again, so the contiguous segments from the founding generation keep getting broken up into smaller and smaller bits. So, if we catch the genome fairly soon after the genetic admixture has occurred, we should see that the chromosome segments occur in larger, contiguous blocks– and that’s exactly what Fu and colleagues found!

Look at the top row in the figure above. That’s Ust’-Ishim Man– note that his Neanderthal DNA occurs in larger blocks, indicating that it has not yet been fully broken up by crossing over. His genome represents an earlier stage in the genetic admixture of Neanderthals and anatomically modern humans. The actual interbreeding has already occurred– he’s 2% Neanderthal– but his Neanderthal DNA still largely occurs in unrecombined blocks. Based on this, Fu and colleagues have been able to calculate about how long before Ust’-Ishim Man the interbreeding occurred, and come up with a figure of about 300 generations, or about 10,000 years before Ust’-Ishim Man. So, the interbreeding occurred on the order of 50-60,000 years ago.

You might also wonder why our genomes are mostly from anatomically modern humans. If they and Neanderthals interbred, shouldn’t it be 50-50? Well, no– it would be 50-50 only if there were an equal number of ancestors from the two groups, but that’s not necessarily the case (in fact, we know it’s not the case in this instance). Most of the “hybrids” must have backcrossed (i.e. had children) with anatomically modern humans. There are many instance in history of two modern human groups meeting and interbreeding, but with a rather unqequal genetic contribution to the descendant populations. In the case of Neanderthals, the ratio was about 1 to 49. It’s easy to imagine how this might happen– a lone Neanderthal being adopted into a modern group, with its descendants therefore breeding mostly with the numerically predominant moderns. Many other scenarios could be posited, but they would be mostly speculative.

Whenever I see interesting results in human evolution, I always check to see what John Hawks has to say, but he’s not posted on this yet; fortunately Carl Zimmer at the NY Times has been able to get a hold of John personally, and ask him what he thinks:

“It’s irreplaceable evidence of what once existed that we can’t reconstruct from what people are now,” said John Hawks, a paleoanthropologist at the University of Wisconsin who was not involved in the study. “It speaks to us with information about a time that’s lost to us.”

That’s absolutely right of course, but I’d like to hear more of what he has to say, and I hope he will post something on the new discoveries.


Fu, Q., et al. 2014.Genome sequence of a 45,000-year-old modern human from western Siberia. Nature 514:445-449. abstract

39 thoughts on “Yes, Neanderthals are us!

      1. “Yes, Neanderthals are us!” – “Us” means homo sapiens, but also non-Africans. If I were black, I’d be a little bit distressed about the implied syllogism.

  1. The actual interbreeding has already occurred– he’s 2% Neanderthal– but his Neanderthal DNA still largely occurs in unrecombined blocks.

    Can we tell anything based on the location of the block? I.e., does it give any information on which traits or genes this individual most likely inherited from neanderthals?

    1. What an interesting question! But it is probably very hard to answer it precisely, since there are probably hundreds of genes in those long yellow stretches throughout the genome. Keep in mind that the figure just shows one fragment of an autosome. I just had a quick look at the supplementary information in the manuscript and they analyzed 221,000 SNPs that are presumed to be of Neanderthal ancestry, so it would be hard to describe them all in terms of particular traits.

    2. I seem to remember that Pääbo has mentioned coupling to traits as the next area he wants to see work on, and in the NYT video he says it (again).

  2. I’m amusing myself by imagining Ken Ham reading

    “The autosomal mutation rate was about .5X10E-9 per site per year, the Y chromosome mutation rate was higher, about .75X10E-9 per site per year, and the mitochondrial rate much higher, about 2.5X10E-8 per site per year.”

    and trying to shoehorn it into his (Neanderthal) world view….

    1. Well, yes, it does mean more than that – but I thought this post was pretty clear about what it tells us about the timing of the splits in European and Asian populations, the dates by which “Modern” and Neanderthal had completed their interbreeding, etc.

      Since modern humans and Neanderthals interbred, then Neanderthals aren’t considered a separate species. But then I don’t know if they ever really were considered to be a separate species (others may know).

      1. We are traditionally classified as separate species (H. sapiens and H. neanderthalis. But the interbreeding data shows we are one species in terms of the biological species concept. So now we can say we are subspecies of H. sapiens.

        1. I don’t think it even means that. Under the BSC, separate species don’t normally interbreed but populations of the same species normally do (unless separated by a physical barrier). What the ancient DNA shows is that sapiens and neanderthalensis (and Denisovans, and possibly other archaic lineages) maintained separate identities without interbreeding for hundreds of thousands of years on separate parts of a single connected landmass (Africa-Eurasia), which suggests they were ecologically quite distinct and consistent with acting as separate species. Interbreeding happened right at the end, when there were few Neandertals left, and it wasn’t a matter of wholesale intermarrying with the new neighbours but (more likely) a few stragglers being absorbed after the old people were nearly wiped out. For the new people coming out of Africa, the idea that ‘Neandertals are us’ was probably a minority view.

          1. But of course there’s no real dividing line between ‘species’ and ‘subspecies’, which was one of Darwin’s arguments that they are points along a temporal continuum.

    2. This work tells us not only that we have some Neanderthal DNA in us, but how it got there, and when. Based on the genomes of modern humans and the fragments of Neanderthal genomes that had been sequenced it could be predicted that the genome of an anatomically modern human of that period should look like this, with large “chunks” of Neanderthal DNA in the genome. It is nice to see this hypothesis demonstrated.

  3. What this drives home to me is how incredibly brief human history is. We only have, what, 10,000 years give or take of history? These people that interbred with neanderthals lived over four and a half human histories ago. We know virtually nothing about their struggles their wars, their hardships, their victories. We don’t know their names, we don’t know what they considered virtues and vices, we don’t know about their heroes and monsters. But we are their descendants.

    We tend to think of our meager 10,000 years as being the be all end all of humanity and our world. In 45,000 years, odds are nothing we’ve done will be remembered. Given the time spans we tend to think in, 45,000 years is an incomprehensible amount of time for humanity. It is a sobering thought.

    1. actually a quick google search says that 10,000 years is a gross over estimate. We have 5-6 thousand years of recorded history, and a bit longer of pieced together history.

      1. Yes, Timothy Hughbanks, sooner or later some intellectually consistent Catholic scientist, probably a Jesuit, is going to have to posit that all life forms have a soul. Right back to the prokaryotes. And he will probably be excommunicated for Jainist tendencies. x

    2. I can think of several things we’ve done in the last century or two that our remote descendants will remember, including mass extinctions, global climate change, and nuclear waste. In the coming century or two we will likely have novel organisms created from scratch by genetic engineering, human habitation of Mars and other solar system bodies, digital minds of human or superhuman intelligence, and other game-changing wonders that will leave permanent marks on history.

      So even 45,000 years from now, historians will point to our era as a time of significant change. Unless of course we’re extinct, in which case nobody will be around to remember anything.

      1. If I understand John Hawks article correctly he is stating that modern humans did not have Ust’-Ishim Man as a direct ancestor. They were yet another population of humans in our family tree which also interbred with Neanderthals.

        I got distracted a bunch of times while reading it so I may just be confused.

        1. I don’t think that you’ve got it quite right. I think he’s saying that non-African modern populations and Ust’-Ishim shared a common ancestor group that had interbred with Neanderthals. Some of the descendant populations survived, some didn’t. Us’-Ishim is among the ones that don’t have living descendants.

  4. A small amount of introgression between two groups doesn’t necessarily make them one species. The boundary between two species and one species is of course arbitrary and ambiguous. I can’t say with confidence that the data show us to be two species, but can you say with confidence that they show us to be one species?

  5. As I understand it, current taxonomy posits that Neandertal and modern humans were closely related but separate species with a common ancestor, H. heidelbergensis. The issue of being a single variable species relates more to our classification systems, and is thus a matter for due discussion. But an interesting question.

    1. In 2001, J. Hei enumerated 25 different definitions of species (Trends in Ecology & Evolution 16(7), 326- 329). The problem is that we (I mean, our human brains) work with mutually exclusive categories, while the limits of species (= genetic units of evolution) are hopelessly fuzzy. It is frequent to observe, in a given species, populations which are still exchanging some genes with a neighbour species and, simultaneously, populations differenciated from each other enough to show very restricted gene flows. Species is a handy, even necessary concept, but a definitively, intrisically vague one as well.

  6. So where did Neanderthals originate?
    We know we Europeans came out of Africa and we know that negroes don’t have Neanderthal genes. Knowing also that all humans were created on the 6th day of creation, and that we were made in God’s image it must be that God is a Neanderthal.

Leave a Reply