June 24, 2009

mtDNA of Tyrolean Iceman using a multiplexed Single-Base-Extension assay

From the paper:
In addition, because the Iceman’s sequence matches the rCRS at nps 497, 498, and 5913 (diagnostic of K1a, K1c, and K1b, respectively), we can extend previous suggestions that the Iceman’s genotype does not belong to any of the three known clades of K1, and should be referred to a new (undefined) paraphyletic clade, of K1 [25] (Figure 1).


The likelihood of the current results representing the endogenous DNA of the Iceman is substantially increased by the haplotype falling outside of K1a/K1b/K1c, combined with the unique hg K transition C8137T, matching the results of [25].


The fallibility of independent replication

Given the absence of 16093C and 16362C amongst the replication results of Handt et al., the only plausible conclusion is that these derived from a second, entirely different, hg K source. These results, combined with a unique haplotype in [25] and the present study, strongly suggests that both laboratories in the original study suffered from hg K contamination. The mtDNA profiles of the staff in the primary laboratory of the first study were not provided [14]; although a subsequent publication of the replicated results [37] disclosed that a member of the secondary laboratory staff was hg K (16224C-16311C), substantially increasing the likelihood for contamination occurring.

[25] refers to Ermini et al. on the Tyrolean Iceman mtDNA sequence, [14] to Handt et al., and [37] to Richards et al.

BMC Genetics doi:10.1186/1471-2156-10-29

Genotyping human ancient mtDNA control and coding region polymorphisms with a multiplexed Single-Base-Extension assay: the singular maternal history of the Tyrolean Iceman

Phillip Endicott et al.

Abstract (provisional)


Progress in the field of human ancient DNA studies has been severely restricted due to the myriad sources of potential contamination, and because of the pronounced difficulty in identifying authentic results. Improving the robustness of human aDNA results is a necessary pre-requisite to vigorously testing hypotheses about human evolution in Europe, including possible admixture with Neanderthals. This study approaches the problem of distinguishing between authentic and contaminating sequences from common European mtDNA haplogroups by applying a multiplexed Single-Base-Extension assay, containing both control and coding region sites, to DNA extracted from the Tyrolean Iceman.


The multiplex assay developed for this study was able to test sufficient polymorphisms in one reaction to unequivocally demonstrate that the Iceman's mtDNA belongs to a new European mtDNA clade with a very limited distribution amongst modern data sets. Controlled contamination experiments show that the correct results are returned by the multiplex assay even in the presence of substantial amounts of exogenous DNA. The overall level of discrimination achieved by targeting both control and coding region polymorphisms in a single reaction provides a methodology capable of dealing with most cases of homoplasy prevalent in European haplogroups.


The new genotyping results for the Iceman confirm the extreme fallibility of human aDNA studies in general, even when authenticated by independent replication. The sensitivity and accuracy of the multiplex Single-Base-Extension methodology forms part of an emerging suite of alternative techniques for the accurate retrieval of ancient DNA sequences from both anatomically modern humans and Neanderthals. The contamination of laboratories remains a pressing concern in aDNA studies, both in the pre and post-PCR environments, and the adoption of a forensic style assessment of a priori risks would significantly improve the credibility of results.



  1. How could the previous studies have been so sensitive to contamination, given the quality and quantity of material they could work with?

  2. We are waiting for a response of the previous scholars, but this research seems good, having found a new subclade of K1, with three haplotypes rooted in Western Europe (K1d: SO44, Iceman and EU073969). As I have always thought that my K1a1b1 was born in the Italian Refugium and then expanded to Middle East (the Jewish K1a1b1a), I think that this research is a step on demonstrating this.

  3. What is K's ethnic Makeup? I don't mean WHERE they were found but what are they as a group and what are they compared to other types of Europeans?

  4. How can Mt K be Jewish and Italian,and whatever? I know there are minute or major admixtures in all of us ,but you can't be everthing at once ,so ,like,what the hell is MT K anyway?

  5. I have maintained very strongly that contamination is present in most results on ancient DNA particularly the result obtained for the bone fragments from that cave in Italy that had the mtDNA CRS. Haplogroup H was from age estimations not even present in Europe 28 kya. The CRS is not even a common mtDNA H variant today. Everything about the result screams contamination, and the contamination could have occurred years before being asceptically handled by the dna testers. It is interesting that result, the Italian bone fragment, has not been verified by other dna testers.

    Moral: take those results on ancient bones with a large bucket of salt.

  6. Except in this case, as I commented before, you are dealing with huge quantities of DNA from a well-preserved body with intact skin "stored" in sterile ice, where at least internal DNA should be very, very well protected.

    Still does not compute.

  7. I'm three and a half years late here to the party. What a great blog. You can be Italian with Jewish markers. I am. I did the GENO 2.0 test, which tests 150,000 markers. I'm northern Italian and I matched up with people from Tuscany. However on my mtDNA, I have k1a1b1a. That is supposedly Ashkenazi. However, I only had one of their markers, so I'm not sure how that works, since they want to see three or four of them. I have three markers in common with the Iceman, which is more than I have in common with the Ashkenazi. His marker, 10978 G,G, is what I have, yet the Ashkenazi have this also. Reading the numbers makes no sense if you don't list the base pairs. We all have something at 10978, but it can be G,A,T or C. Iceman did have 10978. So, these researchers aren't telling the whole story. I'm not sure why they don't know that Iceman also had 10978. It's all over the Internet. I'm one Italian with mtDNA that is k1a1b1a, but I group with Italians. Genetics is more complex than even researchers understand.


Stay on topic. Be polite. Use facts and arguments. Be Brief. Do not post back to back comments in the same thread, unless you absolutely have to. Don't quote excessively. Google before you ask.