Dubious commercial claims
We've discussed elsewhere on this website the problematic and exaggerated media coverage generated by the genetic ancestry testing company BritainsDNA. They are not by any means the only company making misleading claims, though they are, to our knowledge, the only one to resort to legal threats in an attempt to silence criticism (see BritainsDNA saga). We highlight below other dubious claims from businesses related to genetic ancestry testing.
The DNA satnav
The Daily Mail, BBC News, The Washington Post and many other sources carried the story in May 2014 of a new DNA "satnav" from Prosapia Genetics which purports to pinpoint the village your ancestors once lived in a thousand years ago. The underlying research on which this tool is based was published in a peer-reviewed paper, but the university press releases over-hyped the claims (see the press releases from the University of Sheffield and the Children's Hospital Los Angeles). In the paper the median distance from the true origin was 450 kilometres, which is hardly the village level precision claimed. Dr Tatiana Tatarinova, one of the lead authors of the paper, launched the new Prosapia website on the back of the research offering customers the chance to order an analysis using the "new GPS time machine" (this website has since been removed although the GPS tool is available on Tatarinova's lab website). Unfortunately the test does not live up to the hype, and from the start customers have been dissatisfied. For the full story see Debbie Kennett's blog post Driving in the wrong direction with a dodgy DNA satnav. She highlights the very large number of ancestors each of us had 1,000 years ago, and they are very unlikely to have all come from the same place, so that it is "... meaningless to try and pinpoint a single geographical location as the origin of all those diverse ancestors one thousand years ago". In any case the patterns exploited by the DNA satnav are from the DNA of people alive today, and were formed by many processes at many times in the past, so that any reference to a specific time depth such as 1,000 years is implausible.
Genetic homeland case reports
Genetic homeland case reports are offered by Tyrone Bowes through his Irish Origenes, Scottish Origenes and English Origenes websites. The reports purport to pinpoint the place of origin of one's patrilineal ancestors one thousand years ago based on the results of a 37-marker Y-DNA test from Family Tree DNA. However, the methodology used in the reports does not appear to have been subjected to systematic testing or scientific peer review, and the reports make a number of false assumptions and fail to quantify or report uncertainty. See Debbie Kennett's critique of the Origenes' genetic homeland case reports for a summary of the problems. See also Debbie's letter in the August 2014 issue of Family Tree Magazine.
ConnectMyDNA is an American company which "combines a DNA test with a personality test to identify those most like you". Your DNA profile is then transformed into a GeneRing, your "personalized genetic emblem". This is genetic astrology at its worst. There have been many dissatisfied customers. See the complaints on RipOff Report and ReviewsTalk. See also the article A DNA test not to bother with by the "Legal Genealogist" Judy Russell.
Historical haplotypes – Genghis Khan and Niall of the Nine Hostages
A number of companies offer Y-DNA tests for customers who might wish to find out if they are related to historical figures such as Genghis Khan or Niall of the Nine Hostages (see for example here, here, here and here). The purported Y-DNA signatures of Genghis Khan and Niall of the Nine Hostages have been published in scientific papers but these studies merely found unusually prevalent Y-chromosome lineages, and proposed the association with Genghis Khan and Niall of the Nine Hostages as a possible explanation. The hypotheses have not been confirmed, and even if these historical figures possessed the Y haplotype attributed to them, it is likely that many of their contemporaries also carried it so that direct descent cannot be established simply from carrying that haplotype.
The Genghis Khan study was published in 2003 by Zerjal et al. The authors provided evidence of a predominant Y-chromosome lineage found throughout a large region of Asia, which they attributed to descent from the Mongol emperor Genghis Khan. The Niall of the Nine Hostages claim derives from a 2006 study by Moore et al who identified a Y-chromosome signature that was found at an unusually high frequency in northwestern Ireland. They suggested an association with the Uí Néill dynasty which traces back to a "possibly mythological 5th-century warlord" known as Niall of the Nine Hostages.
Ideally one would like to extract DNA from the remains of the historical figure under investigation, but that was not possible in either of these cases. It is not known that Niall of the Nine Hostages actually existed, and the burial place of Genghis Khan has not been identified, although there has been much speculation. In the absence of ancient DNA we are reliant on testing the DNA of living people and using statistical methods to estimate the time to the most recent common ancestor (TMRCA). There is considerable uncertainty in all such estimates, and both these studies were based on low-resolution Y-STR testing (17 and 16 markers respectively) with no SNP testing to filter out Y-STR matches due to "chance" rather than recent shared ancestry. Consequently there is a wide error margin in any such TMRCA estimates. Even if it were possible to estimate the TMRCA accurately the link with a historical figure remains speculative (Royal et al 2010). There are other possible explanations for the high prevalence of a particular haplotype. If you take a Y-DNA test and you match a Niall or a Genghis Khan signature then there is a possibility that the famous man was your ancestor but there is no way to be sure and it is also possible that all your ancestors had no claim to fame beyond their success in contributing to what is today a highly prevalent haplotype.
See also the article by Paul Frijters Why Genghis Khan won't have had 16 million descendants.