View Full Version : What should population geneticists at universities be researching? My recommendations

08-02-14, 13:40
I am starting to get annoyed by the number of useless studies published recently in the field of population genetics that do not tell us anything that we didn't already know, and do not provide any worthwhile new data (e.g. Pickrell et al. 2014 (http://www.eupedia.com/forum/threads/28940-The-Khoi-San?p=426061&viewfull=1#post426061), Sverrisdóttir et al. 2014 (http://www.eupedia.com/forum/threads/29484-The-mystery-of-Lactase-Persistence-%28LP%29-in-Europeans), Pierron et al. 2013 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE53445), and that's just for the last two months). Then you have the case of mtDNA studies that concentrate on completely outdated HVR sequences, which are pretty useless for any purpose nowadays. That is the case of Lee et al. 2013 (http://link.springer.com/article/10.1007/s12520-013-0173-0), who tested a handful of ancient mtDNA samples from Neolithic Germany (Rössen culture), and Deguilloux et al. 2013 (http://www.sciencedirect.com/science/article/pii/S0305440313003191), who tested a few Merovingian mtDNA samples. Neither study yielded informative results.

By nature, I don't like wasting time, energy or money. It's even more irritating when potential for enlightening scientific studies is wasted. This is why I have decided to make a list of suggestions of what would be really interesting to study.

Mitochondrial DNA studies

The price of DNA sequencing has been plummeting and it would be unthinkable to make any new mtDNA studies without testing the full mtDNA sequence, or at least all the phylogenetically relevant mutations (not that it would change much in terms of costs). Knowing that someone belongs to haplogroup H or K is almost as useless as not knowing anything. Only deep subclades can shed light on historical population movements and differences between ethnic groups.

I suppose that university researchers do check what populations have been tested in the past, and do not choose to focus their new studies whimsically on population they "like" or are easy to sample, but on those that have been understudied to date and who require further investigations. I have made a compendium of mtDNA frequencies by country (http://www.eupedia.com/europe/european_mtdna_haplogroups_frequency.shtml) which can immediately show what country or region remains undersampled. Among the top European and Middle Eastern countries that need to be tested in greater details are:

The Netherlands: not a single proper mtDNA study to date. The listed data is from commercial tests only.
Belgium: the only study is De Corte 1996, which only had 31 samples that could fit within the haplogroups used today. All other samples are from commercial tests.
Syria: only 118 research samples to date, ridiculously low for a region with so much ancient history.
Kurdistan: very few research samples despite the region's prominent role in ancient history (Assyrians, Hurrians).
Albania: only 149 research samples to date, and apparently a huge diversity of minor H subclades that could shed light on the origins of this haplogroup.
Serbia: only 117 research samples to date, and a lot of presumed diversity due to the prominent role of the region during the Neolithic.
Wales: only 92 research samples to date, which give an improbably high percentage of hg H (60%).
France: lots of samples, but not geographically diversified (over-representation of Brittany and southwest France). Most of central and eastern France remains unsampled.
Russia: same as France. Many regions left completely unsampled in European Russia, including all the Pontic Steppe (anywhere south of the Kursk-Saratov axis), which is a shame since it is the Indo-European homeland and could be an invaluable source of information.

What we need are high-quality studies like Kushniarevich et al. 2013 (http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0066499) or Hernández et al 2014 (http://www.biomedcentral.com/1471-2156/15/11), which report deep subclades for hundreds of individuals in a relatively small region and compare frequencies across sub-regions.

Y-chromosomal DNA studies

Most European countries are fairly well sampled are present. Those that need more effort are:

France: very few research samples available and 2/3 of the regions remain completely untested. Even commercial tests are scarce due to the prohibition of DNA tests in France, so that most commercial tests are heavily biased towards Canadians of French descent (who are not representative at all of the whole country).
Austria: possibly the most undersampled country in Europe. Even Greenland is better sampled.
Switzerland: Only one official study (Zürich area) + some commercial samples from the north. I expect to find huge regional differences between cantons and valleys, just as has been the case for the Italian Alps. Therefore Switzerland should be a priority for Y-DNA studies.
Kosovo: only one small study to date which yielded oddly high frequencies of E-V13 (possible due to sampling bias).
Moldova: potentially very interesting region, but only the Gagauzes in the south were tested.
Russia: like for mtDNA, most studies focus on the north-west and leave the Pontic Steppe mostly untested.

In the Middle East the most important countries to study are:

Syria: the least studied country in the region at the moment. Probably not the best time to go ask for volunteers though.
Iraq: the only important study focuses on Marsh Arabs, who aren't representative of the country as a whole. Considering the huge importance of Mesopotamia in ancient history, a detailed regional analysis of Iraq is a must.
Turkey: only one average-size 10-year-old study by Cinnioglu et al. 2003. Too little for a big and diverse country which played such a central role in both Middle Eastern and European prehistory.

Ancient DNA studies

Ancient Y-DNA and/or autosomal DNA from Mesopotamia (especially from the Neolithic, Chalcolithic and Bronze Age, including Sumer, Akkad, Old Babylonia and Assyria).
Ancient Y-DNA, mtDNA and/or autosomal DNA from the Natufian culture and Pre-Pottery Neolithic Levant.
Ancient Y-DNA, mtDNA and/or autosomal DNA from the Göbekli Tepe, Arslantepe, Çatalhöyük and other major Neolithic Anatolian sites.
Ancient Y-DNA and/or autosomal DNA from the Levant from the Neolithic to the Bronze Age.
Ancient Y-DNA, mtDNA and/or autosomal DNA from Maykop, Yamna and Kura-Araxes cultures.
Ancient Y-DNA and/or autosomal DNA from Rome or Latium from the 8th to 2nd century BCE.
Ancient Y-DNA and/or autosomal DNA from the Etruscan civilization.
Ancient Y-DNA and/or autosomal DNA from the Neolithic Balkans and Carpathians, especially from the Vinča, Varna and Cucuteni-Tripyllian cultures.

Please do not waste time and funds testing any ancient samples newer than the Iron Age.

08-02-14, 19:17
Well, the south-tyrol, trentino areas are very highly sampled, but very rarely initiated by the people, but by the regional authorities . Unsure why the sample so heavily. ( this info to me by relatives )

09-02-14, 23:25
More detailed mesolithic DNA from Sweden, especially skin colour and eye colour related. A sample from Sungir. A sample from the Swifterband culture. Samples from the two burials of Téviec?

From recent finds of Groß Fredenwalde palaeogenetic studies were taken, according to this:


09-02-14, 23:42
Great topic. I Think more detailed research of The indigenous people of The Pacific Northwest Coast, like the Salish and Wakashan people's needs a more througouh study done, But more specifically The Haida. The Haida Have a language and culture completely distinct from all other Pacific North West Coast people's that surround them. They Have a culture That Is most similar to East Asian aboriginals like The Ainu and Nivikh people. I think collecting DNA from The Hadia Could shed a lot of light on The earliest migrations of humans into The America's.