Recovering 1,000 Year Old Viking mtDNA

Today’s big anthropological headline is based upon this brand new open access PLoS One paper, “Evidence of Authentic DNA from Danish Viking Age Skeletons Untouched by Humans for 1,000 Years.” The paper is written well, has lots of controls, and establishes some protocols on how to recovery ancient DNA. Despite these good things, I think I have found some minor holes in this paper, which I’ll address in this blog post.

The above photo is of the excavation of the skeletal remains sampled in this current study. The remains come from the Gladegil burial site near Otterup, Denmark. Gladegil has yielded 3 males, 4 females, and 3 other individuals whose sex can’t be determined. Carbon-14 dating has revealed that the site is approximately 1,000 years old and is a Christian burial site. As you can see the excavators are in sterile bunny-suits, worn to prevent contamination of modern DNA during the exhuming process. I wonder if this new image of archaeologist will replace the dusty-rugged Indiana Jones type-caricature?

The authors of this paper extracted only mtDNA only from the teeth and associated artifacts, so the title of this paper is misleading, it should have been rewritten “Evidence of Authentic mtDNA…” You may not see this as a significant issue, but I think it is. Other researchers and editors have made sure titles are as accurate as possible, so why then did this one get the green light? While mtDNA is DNA, mtDNA is different from nuclear DNA. First of, there are a lot more copies of the mitochondrial genome compared to the nuclear genome, that’s because cells have one nucleus (for the most part) but many mitochondrion. The mitochondrial genome is much smaller and also is a completely different entity from the nuclear one. It is circular, whereas the nuclear genome is linear. Lastly, the nuclear genome is less stable and that makes extracting and sequencing it from any sample, ancient or not, a much more challenging task.

Anyways, mtDNA was isolated in a clean room by grinding the teeth and what not into a pulp. The highest quality reagents and kits were used, according to the authors. Another misleading aspect of this study is that the entire mtDNA genome was not sequenced, only a portion of the genome, the hyper-variable region 1 (HVR-1) was amplified by quantitative PCR. The HVR-1 is a pretty informative region of the mitochondrial genome, some haplogroups can be identified but it is by no means a high-resolution indicator. If the authors extracted the HVR-2, a region with a lot more haplogroups, the analysis woulda been much more thorough.

I didn’t find any discussion on how the samples were sequenced. Given that the HVR-1 is small, roughly 560 base pairs, I’m thinking a standard dideoxy chain terminated method was used. I guess it is so common nowadays that people don’t even write it in their preliminary publication. I didn’t check the supporting materials. The HVR-1 sequences from the skeletal were aligned based upon known Inuit mtDNA.

The DNA of everyone who was involved with the excavation, extraction, sequencing was also sequenced, to screen for any possible contamination. I give the team kudos for including that in their study. It is critical for anyone doing ancient DNA work, also forensic DNA work, to not only keep track of everyone that handled the samples but also include their DNA in the sequence analysis.

Based upon the haplogroup distribution, the samples were attributed to be of Viking ancestry. To clarify, the presence of haplogroup I and G6 have also been seen relatively higher frequencies in other Viking populations. Hg I is not observed in any ancient Italian, Spanish, British, and central European populations. Again, if HVR-2 was included, this woulda been a more robust association of ancestry…

In general this was a graceful little paper. But after reading it, I don’t see what and where the big umph of this paper is to warrant so much play in the press. There have been much more challenging successes in ancient DNA work that didn’t get as much attention as this paper has gotten. And like I said, it is ultimately a report of an isolation and sequence comparison of only 500 base pairs… which is much easier to do than a nuclear one. I’d like to have seen at least the HVR-2, that woulda given me more of a meal-deal. Perhaps more will be done is a good template for ancient DNA work? At the very least it is an effective template for how to excavated ancient-but-recent DNA and what controls to include in the analysis.

    Melchior, L., Kivisild, T., Lynnerup, N., Dissing, J., Ahmed, N. (2008). Evidence of Authentic DNA from Danish Viking Age Skeletons Untouched by Humans for 1,000 Years. PLoS ONE, 3(5), e2214. DOI: 10.1371/journal.pone.0002214

6 thoughts on “Recovering 1,000 Year Old Viking mtDNA

  1. I’ll believe it when someone replicates the analysis of the DNA. All the other DNA that has supposedly been found has been from contamination.

  2. Kevin, I don’t quite understand your comment. Care to rephrase it? Are you suggesting the mtDNA from this study is strictly from the excavators, or through the excavation, and not directly from the samples?


  3. Kevin, I don’t quite understand your comment. Care to rephrase it?

    He obviously means that it’s proper academic procedure to replicate all tests independently to check for any possible errors, specially in such a delicate and often controversial field.

    Personally I think they did their best to prevent contamination but certainly replication is a must in any case. Even if contamination is discarded there are other possible sources of errors, like DNA degradation.

  4. Luis, thanks for clarification. My first comment was written in haste, I now better understand Kevin’s comment and I agree, the data will need to be reconfirmed in another lab.

    I remember reading in another aDNA paper where the lab that did the preliminary testing sent the samples to another, unassociated lab (with new primers, reagents, equipment) and asked them to replicate the results. I’m surprised it wasn’t done in this paper!


  5. The paper does not mention any replicated results or analysis. I still think this is a good test for other reasons the authors state (similitude with previous regional aDNA tests, large array of resulting haplogroups, extreme unprecedented profilactic measures since the digging…) but it seems like a methodological flaw in any case. Other aDNA analysis lacking replication (notably Australian Lake Mungo remains, that yielded quite odd results and will probably never be tested again because of native burial laws) have been subject of heavy criticisms. Obviously this is more likely to happen if the results are very unexpected, not if they fall within expectations as in this case.

Comments are closed.

A Website.

Up ↑

%d bloggers like this: