December 23, 2010

Is the Denisova mtDNA sequence modern?

So says a paper that was posted in Nature Precedings. People more knowledgeable than myself about mtDNA sequence alignments and such are requested to chime in.

If true this would be very significant, because it would mean that the very archaic Denisova mtDNA reported in the new Nature paper is not really archaic.

Anyway, here's an excerpt from the paper:
The recognition of the Neandertal inserts in the Denisova sequence changes the reading considerably and indicates that the sequence before the insert at 16194, perhaps ending at 16181 is corrupt either from degradation or during preparation for sequencing. There may be another explanation for the lack of sequence alignment before that location and the substantial agreement after it. Nevertheless, this finding argues for modern human status of the Denisova sample and against a new species designation as suggested from the original analysis (Krause, et al., 2010). Variations in mtDNA in populations and their significance given the natural history of mitochondria have been noted by Ballard and Whitlock (2004) with a caution of their use to build phylogenetic relationships.

Niccolo Caldararo


Andrew Oh-Willeke said...

Per the paper: "This changes the appearance of the Denisova sequence and makes it appear to reflect a combination of Neandertal sequences and AMH sequences[.]"

Since mtDNA is uniparental, a mix of sequences from two species doesn't make sense. The paper could be Neanderthal confused by degradation and mtDNA from AMH researchers. But, unless there were some Neanderthal grad students on the project (perhaps they smelled and looked like it, but this hypothesis seems unlikely), I don't see how this could be a AMH mtDNA sequence given the Nature commentary writer's admissions.

pconroy said...


You'll probably be interested in what Joao Zilhao is quoted as saying about this, over at the Loom:

So, I suspect the “Denisovans” are in fact “modern humans.” The future will tell, but that should surprise no one, because I see no reason to suppose that genetic lineages (however defined, namely via mtDNA [mitochondrial DNA]) that are very old and cannot be found any more among present day humans did not survive until a very recent past (in evolutionary terms). Until “globalization” events occurring in the last 10,000 years confused everything, that would be the thing to expect under a scenario where, at any given point in paleontological time, all humans belonged to a single biological species but one where, due to strong population structure related to geographical isolation, local extinctions and frequent recolonizations with attendant founder effects, there was a lot more variation in morphology and in genetics than there is today. To me, that’s what the fossils have bee saying for the last 15 years, and the genetic evidence is simply adding further detail and further complexity to that scenario.

clusteredmaps said...

If this sequence is not archaic at all it would have to be either a member of Haplogroup M or Haplogroup N.

Andrew Oh-Willeke said...

"If this sequence is not archaic at all it would have to be either a member of Haplogroup M or Haplogroup N."

It could be a member of Haplogroup L as well. The fact that there are no remaining trace of African mtDNA halogroups in Eurasia doesn't mean that sometime, somehow, some woman with an African mtDNA haplogroup couldn't have left Africa ca. 50,000+ years ago, whose descendants formed an isolated population that died out in Southern Siberia, perhaps after admixture with Neanderthals that left that population with an archiac morphology.

But, your essential reasoning is sound. If its mtDNA doesn't fit anywhere on the phylogony of modern human DNA then the break between Denisovians and modern humans has to be older than mitochondrial eve, and the case that Denisovians were modern humans starts to look very weak.

German Dziebel said...

"It could be a member of Haplogroup L as well."

Unlikely. For once, we haven't found any Ls in small pockets outside of Africa at low frequencies. The comparably old mtDNA from Kostenki is U2. The Denisova sequence, if proven modern, could instead represent an undifferentiated M/N lineage and an antecedent to African L lineages. East Asia, America, Australia and Melanesia are characterized by populations with a mosaic of M and N lineages. (Compare Europe only has N lineages, especially those belonging to the R subclade of N.) This mosaic is often explained as the hybridization of M clans with N clans but it can also mean that their common denominator outside of Africa has simply got lost. At the same time, there are no autochthonous M or N lineages in Africa (M1 and U6 found in North and East Africa are secondary arrivals from outside of Africa). So, it's been always puzzling that M and N derive separately from L3, don't share a common ancestor specific to M and N, but then they get mixed up everywhere outside of Africa, including such remote places as Australia and South America. Denisova, if proven modern, may just fill in the phylogenetic gap that Africa couldn't.

Anonymous said...

I think it quite normal and expected that the finding of the archaic nature of Denisovan mtDNA would be challenged.

Think of the Piltdown Man episode. Findings have to be questioned particularly with ancient dna with by its age is degraded, however the degradation process is well known and corrective measures are taken. Eventually more specimens will be found, and tested. Mitochondrial dna is simple being similar to bacterial dna and consists of a single strand of dna in a circle. There is plenty of mtDNA to work on, and it should be easier to see the effects of degradation and correct for it.

Scientists are fallible, and unfortunately all too human. Many of them want to hang on to old paradigms which in this case is the Out of Africa one with no interbreeding between the "Africans" and the older hominid residents of Eurasia.

eurologist said...

Yes, it is preposterous to claim that modern human mtDNA variation in very small populations in Central Asia 60,000 years ago was vastly - by orders of magnitude - wider than the entire extant human population.

Yes, most (sub-) lineages became extinct - but not many outside of today's trees and especially not outside of Africa's. If such high diversity had been the case, extant lineages wold be a mess to reconstruct compared to the rather simple structure we see, today.

As to the author of the note - Niccolo Caldararo is an Anthropologist and Museum collection worker, and (my guess) about as much of a dilettante in mtDNA as I am. I know a few groups here in the US who are "highly critical" (to put it mildly) of Svante Pääbo's work - in my view, unprofessionally confusing personal resentment with scientific critique.

The note is very poorly worded - similar to the sometimes quite emotional rebuttals to peer paper reviews. Also, his main arguments are hard to follow and are without concrete references. To me, to make sense, he basically says all ancient mtDNA is highly degraded but except for a couple of "Neanderthal insertions" otherwise identical to modern human mtDNA (he may mean something different, but that is not apparent from the "research" note). Well, that doesn't make any sense, and it's not like mtDNA researchers are not trying to find the best alignments.

Anonymous said...

Andrew Oh-Willeke : "The fact that there are no remaining trace of African mtDNA halogroups in Eurasia"

It's rare but some have found where they were not expected.

Reconstructing the phylogeny of African mitochondrial DNA lineages in Slavs

Boris A Malyarchuk et. al

To elucidate the origin of African-specific mtDNA lineages, revealed previously in Slavonic populations (at frequency of about 0.4%), we completely sequenced eight African genomes belonging to haplogroups L1b, L2a, L3b, L3d and M1 gathered from Russians, Czechs, Slovaks and Poles. Results of phylogeographic analysis suggest that at least part of the African mtDNA lineages found in Slavs (such as L1b, L3b1, L3d) appears to be of West African origin, testifying to an opportunity of their occurrence as a result of migrations to Eastern Europe through Iberia. However, a prehistoric introgression of African mtDNA lineages into Eastern Europe (approximately 10 000 years ago) seems to be probable only for Europeanspecific subclade L2a1a, defined by coding region mutations at positions 6722 and 12903 and detected in Czechs and Slovaks. Further studies of the nature of African admixture in gene pools of Europeans require the essential enlargement of databases of African complete mitochondrial genomes.

European Journal of Human Genetics advance 9 April 2008;

One mtDNA L2 among 74 Norwegian samples.

Different genetic components in the Norwegian population revealed by the analysis of mtDNA and
Y chromosome polymorphisms

"One mtDNA molecule displayed an L2 haplotype, that is typically African, but is also sporadically found in Southern Europe and is likely to be a result of more recent gene flow"

eurologist said...

From the Krause et al. Denisova paper:

The final assembled mtDNA sequence was identical irrespective of whether the reference mtDNA sequence used for the assembly was modern human or Neanderthal. Coverage across the mtDNA was high (mean = 156-fold, lowest 2-fold, highest 602-fold; Supplementary Fig. 1).

Several different approaches were used to assess the reliability of the mtDNA sequence determined. First, we made a second DNA extract from the Denisova phalanx and determined 9,908 mtDNA fragments by shotgun sequencing on the Illumina platform. From the reads we assembled a complete mtDNA sequence that was identical to the one retrieved with the PEC approach (Supplementary Information). This shows that the mtDNA sequence retrieved is reproducible and not dependent on the particular primers used for PEC. Second, we used 276 sequence positions where the Denisova hominin mtDNA is different from >99% of present-day human mtDNAs to assess whether the mtDNA recovered with PEC comes from a single individual. Among the 15,008 fragments that overlap these 276 positions, 14,961 carry the bases of the Denisova hominin mtDNA. This suggests that 99.7% (95% confidence interval; 99.6–99.8%) of all mtDNA fragments in the bone come from a single individual.

[Thirdly, details of degradation also match as expected].

n/a said...

If we rule out fraud or unbelievable levels of incompetence on the part of the Max Planck people, there's zero chance of this guy being correct -- especially now that there are two "Denisovan" mtDNA sequences differing by only two mutations. I think the mtDNA from the finger bone has been sequenced at over 100 times coverage, and the coverage for the tooth is about 60 fold. It's unlikely to say the least that both samples would display nearly identical patterns of contamination and DNA degradation.

caldararo said...

It is unfortunate that so few people seem to understand the nature of aDNA contamination and degradation. Much of my career has been taken up with studying the degradation processes affecting collections materials, especially those composed of organic materials as Eurologist notes. Some of the main problems, both theoretical and practical are reviewed in an article I wrote in Human Nature Review in 2002 (, but Gabow and I published a more comprehensive analysis in Ancient Biomolecules in 2000 that might help those without direct experience understand general issues.

German Dziebel said...


I red the reference you sent along. Thank you! So, what's your present take on the Mungo man DNA extraction? In the light of the Denisova study, it'd be good to revisit the nuclear insert and the Mungo man sequence.

caldararo said...

There are a number of definitions of inserts, mtDNA sequences transfered to the nDNA and mutations in sequences of both nDNA and mtDNA, the latter move the frame and can be either silent or damaging to replication of proteins. This latter case is the situation with the Neandertal insert I mention in my paper. One has to keep in mind that the published sequences are of varied quality (see for details). Inserts are common and have been catalogued by a number of researchers.

German Dziebel said...

"Inserts are common and have been catalogued by a number of researchers."

I mean this insert: ZISCHLER, H., H. GEISERT, A. VON HAESELER, and S. PA ̈ A ̈ BO. 1995. A nuclear ‘fossil’ of the mitochondrial D-loop and the origin of modern humans. Nature 378:489–492.

It was used in the Mungo man DNA extraction study. The Mungo man sequence was found to be closer to it than any extant modern human sequence, including African ones.

caldararo said...

I commented on the Mungo sequence in my paper in the Linnean in 2003, mainly about the possibilities of contamination that were being made by others about the authenticity of the results. I also cautioned people about making species designations on DNA variations. How much is necessary? In our 2000 paper Gabow and I argued that the differences between Chimpanzee populations that were interfertile were more than the Neandertal/AMH differences reported by Krings. Keep in mind that the similarity of the sequences in the Denisova sample and modern Melanesians falls in the same reltion as that of Mungo, so the question remains, how do you make the distinction? We also compared Neandertals and Humans with Wolves and dogs who are also interfertile. Recent papers have also shown Wolves and coyotes to share sequences.

German Dziebel said...

Thanks Niccolo. I checked out your Linnea paper. I like your microenvironments point and your mention of a double standard when it comes to some ancient DNA studies. The one you co-authored with Gabow is out of my reach at the moment (do you have an e-copy by any chance? could you e-mail it to me?), but I'll track it down when I get a chance. I was unaware of your publications, so thanks for coming out to the forums. As an applied anthropologist, I also took a note of your economics papers.

BTW, species definition by morphological means is equally problematic. See

caldararo said...

Dear German:
Yes, morphological assignment is and has always been problematic, just reference to the Mid Pleistocene reassignments from Neandertaloid to heidelbergensis has been based on the same measurements back and forth. Brace was one of the more astute in recognition of the variable nature.
The economic teory assumptions and the use of math is like that in the computer assumptions and phylogenetics. As John Maynard Keynes noted in this A Treatise on Probability (1921) people tend to mistake probability for reality.
I think the 2000 paper with Gabow is in pdf form on my office computer at the university but I am away until Monday. They moved my office nd I lost some files, so I hope it is still there. If so I will email it to you or if not, have the library scan it and put it on my website.

Anonymous said...

"I also cautioned people about making species designations on DNA variations. How much is necessary? In our 2000 paper Gabow and I argued that the differences between Chimpanzee populations that were interfertile were more than the Neandertal/AMH differences reported by Krings."

As long as two or more groups of animals are interfertile, they are one and the same species. So Neandertals, AMH, heidelbergensis and erectus are one and the same species, probably also habilis.

caldararo said...

It often seems that the state of molecular phylogenetics is like morphological genetics before Huxley's"new systematics" in the 1930s. I discussed the issue in a review of a book on the evolution of HIV in the American Journal of Human Biology in 2001;2-W/abstract). We see the same passion to name each difference in sequence as a species as in Darwin's day a new species was named with every difference in morphology. See Stephen J. Gould's discussion of this regarding Darwin's redefinition of von Baer's work on embryology and recapitulation (Gould's Ontogeny and Phylogeny, 1977).

German Dziebel said...

"They moved my office nd I lost some files, so I hope it is still there. If so I will email it to you or if not, have the library scan it and put it on my website."

Thanks Niccolo, no worries if it takes time. Just in case, my e-mail address is dziebelg at gmail dot com.

caldararo said...

Yes, as Strat notes, the Biological Species Concept has been considered as important in living analysis as the environmental species concept or that of the species recognition concept. But in paleospecies these are difficult to apply and instead we find a number of valid assumptions about differences standing out. I think that Dr. Conroy's very useful and instructive book (I have used it in my classes on Fossil Hominids) presents a quite logical approach to this problem that could be adopted by more investigators with profit (see: