Neanderthal and Denisovan retroviruses in YOU!

Well, some of you! And maybe me!

Last year, scientists combed through Neanderthal and Denisovian DNA sequences (yeah, those exist!) and found fourteen ERVs that were in Neanderthals and/or Denisovians, but NOT humans.

Neandertal and Denisovan retroviruses


Modern humans (Homo sapiens) last shared a common ancestor with two types of archaic hominins, Neandertals and Denisovans, roughly 800,000 years ago, and the population leading to modern H. sapiens separated from that leading to Neandertals and Denisovans roughly 400,000 years ago [1,2,3,4]. Genome sequences for these two types of archaic hominins have been reported [1,2]. They were determined by sequencing ancient DNAs using techniques that generated many short sequence reads. Here, we analyzed individual sequence reads used to assemble the published Neandertal and Denisovan genomes for insertions of Human Endogenous Retrovirus K (HERV-K) DNA. Virus–host DNA junctions were identified that defined 14 proviruses where modern humans contain the corresponding, empty, preintegration site. Thus, HERV-K reinfected germ lineage cells of Neandertals and Denisovans multiple times, and these events occurred around the time of or subsequent to the divergence of the archaic hominin lineages from that leading to modern humans. One of the proviruses was shared by Neandertals and Denisovans, which is consistent with the hypothesis that these archaic humans shared a common ancestor more recently than they shared one with the lineage leading to modern humans.

Remember, HERV-K ERVs are usually the 'youngest', thus the most likely to be polymorphic between humans, Neanderthals, and Denisovans. In humans, there was no evidence these 14 Neanderthal/Denisovan ERVs were ever there (not a broken ERV, not a solo LTR, just regular ol DNA sequence).

The conclusion was, was that these Neanderthal and/or Denisovan ERVs must have inserted themselves into their respective genomes after the ancient relatives/human split.


Speciation isnt a 'clean' break. Its not as if one day there is a human/Neanderthal/Denisovan ancestor, and the next day there are humans and Neanderthals and Denisovans.

Speciation is a time consuming, tangled up mess.

So this new information isnt surprising at all:

Neanderthal and Denisovan retroviruses in modern humans

The human genome project, the human reference genome, does not contain the genomes of all humans. It contains the genomes of a select few humans.

These researchers hunted through the genetic sequences of individuals whos genomes were sequenced for other reasons (a cancer project) and they ended up finding most of the 'absent' ERVs! Not in every patient, but some patients had one, some patients had others, etc.

What does this mean?

With the exception of co-opted ERV loci such as syncytins [5], which could increase in frequency due to positive selection, we assume ERV loci become common by genetic drift, and the average time for a neutral allele to go to fixation is 4Ne generations (where Ne is the effective population size). Given estimates of long-term human generation time and population size [6], this is ∼800,000 years. The population divergence of modern humans from the Denisovan/Neanderthal lineage is more recent, between 170,000 and 700,000 years according to a more recent — and much deeper —sequencing of the above Denisovan fossil [7], so many loci will have persisted at fluctuating frequencies in all three lineages.

It means unless an ERV is really positive (and selected for) or really negative (and selected against), ERVs just drift. They will remain polymorphic in a population until a given number of generations, depending on the size of the population (that is, all humans have the same really old ERVs, these younger ERVs are different between humans).

So some of us have more in common with our Neanderthal and Denisovan ancient cousins than others.


More like this

The implication seems to be that these ERVs found in a minority of modern humans are leftover from several hundreds of thousands of years previously and moving, slowly, towards fixed absence. An alternative hypothesis is that they were later acquired by the interbreeding that has now been pretty well demonstrated to have occurred to a small degree. If the former hypothesis were true, you would expect some of them to be found in Africans, who are more genetically diverse than Eurasians and their descendants; if the latter, you'd expect sub-Saharan Africans not to have any (though perhaps to have others acquired from the recently reported probable interbreeding with archaic humans there).

"fourteen ERVs that were in Neanderthals and/or Denisovians, but NOT humans"

They were humans :-(((

By El PaleoFreak (not verified) on 21 Nov 2013 #permalink

I haven't made the trip to the library to see this paper yet. Can they tell from the haplotype around each insertion whether the ERV was first inserted in modern human or Neandertal/Denisovan? Without this, it seems possible that the insertion could have occurred in either lineage and been transferred during mating ,or even that the insertion occurred before they divergence of the lineages with incomplete lineage sorting determining what we see today.

While I will not pretend to understand everything she is talking about with ERVs, and how a virus can introduce its DNA into ours, as I am not a scientist; It appears many, many Scientists spent a lot of time in a Lab, but perhaps should have taken a one or two Philosophy classes.
You do understand the absurdity of saying that because entity X has attribute 1 , and entity Y has attribute 1; therefore entity Y came from entity X?
I guess a very simple analogy would be "My cat has the common cold, I have the common cold, therefore I came from my cat..." It's equally absurd to say "Cave man x has an attribute of ERV X, I have an attribute of ERV X, therefore I must have came from, am related to, or have any connection what so ever to cave man x." It may or may not be true however, but the like attributes do not prove it.
Possessing like attributes, even when talking about progeny, does nothing to prove causation. Just thought you should know this...
Actually, here's a bit of a challenge for you. Can you give me any example, of anything, anywhere, other than the one you provided, where we would look at two entities and determine that because they have some like attribute, that one caused the other? The emphasis here is that the similarities are the crux that prove the causation.

Hey Matt,
The difference between your examples and what scientists use do determine relatedness (descended from, or sharing common ancestry) is that you are talking about phenotypes - hair colour, proteins, placentas - and these can be produced from different sents of instructions, such as birds and insects both having wings - or you are talking about things that are not passed genetically, like the common cold. What is evaluated in these papers is the DNA - that is, the particular source of a phenotype. The DNA that codes the (very complex) information for a bird's wing is drastically different and easily distinguishable from the sequence that codes for an insect wing, so we know that these two things, although similar in appearance, arose from different sources and are not evidence for genetic relatedness. The ERVs discussed here, when found in ancient Neanderthal and Denisovian, or modern human, DNA, have sequences that are similar enough that we can say with confidence "They are the same in Neanderthals as in modern humans and must have a common origin." Much like the DNA that codes for syncytin (coded by an ERV that is found in placental mammals and is key in us having a placenta), or the various sequences that together give us fingers.