July 26, 2004
Surname Studies of Ancient Patrilinal Population Structure
American Journal of Physical Anthropology
Early View (Articles online in advance of print)
New method for surname studies of ancient patrilineal population structures, and possible application to improvement of Y-chromosome sampling
Franz Manni et al.
Abstract
Several studies showed that surnames are good markers to infer patrilineal genetic structures of populations, both on regional and microregional scales. As a case study, the spatial patterns of the 9,929 most common surnames of the Netherlands were analyzed by a clustering method called self-organizing maps (SOMs). The resulting clusters grouped surnames with a similar geographic distribution and origin. The analysis was shown to be in agreement with already known features of Dutch surnames, such as 1) the geographic distribution of some well-known locative suffixes, 2) historical census data, 3) the distribution of foreign surnames, and 4) polyphyletic surnames. Thus, these results validate the SOM clustering of surnames, and allow for the generalization of the technique. This method can be applied as a new strategy for a better Y-chromosome sampling design in retrospective population genetics studies, since the idenfication of surnames with a defined geographic origin enables the selection of the living descendants of those families settled, centuries ago, in a given area. In other words, it becomes possible to virtually sample the population as it was when surnames started to be in use. We show that, in a given location, the descendants of those individuals who inhabited the area at the time of origin of surnames can be as low as 20%. This finding suggests 1) the major role played by recent migrations that are likely to have distorted or even defaced ancient genetic patterns, and 2) that standard-designed samplings can hardly portray a reliable picture of the ancient Y-chromosome variability of European populations.
Posted by Dienekes at July 26, 2004 10:53 PM | PermaLink>>the major role played by recent migrations that are likely to have distorted or even defaced ancient genetic patterns
I find this a very salient point, as I have noticed anomalies in the explanation of distributions of various haplotypes across Europe, and feel that this may be the reason.
For instance, it is often postulated that ancient Scandanavia was first populated by Slavic types (R1a), then later by more Germanic types (I and R1b) - this is supposed to explain the 25% R1a in Norwegians. However R1a is found much less in Sweeden and Denmark, so I would suggest that in Norway the high figure is a result of their practice of taking slaves from the slavic areas and bringing them back to Norway. Afterall South Western Norway, where the land is poor, was the birthplace of Vikings, so this is where one would expect the most R1a, and this is exactly what one sees on the ground.
In a similar fashion, farming is often postulated to have reached Southern Spain very early on, based on the high percentage of Middle Eastern DNA in the area, when in reality this area was settled by the Phoenicians of modern day Lebanon and later by North African Arabs, both of Middle Eastern descent.
Comments anyone?
Posted by: pconroy at July 27, 2004 07:04 AMI think that inferring events of the past based on the present-day distribution of genes has severe limitations. There are all sorts of demographic events which could result in the present-day gene distributions, and geneticists follow parsimony methods to infer the origin of mutations, their direction/age of spread etc.
Ultimately, palaeogenetics will shed much light into the history of populations. Already some studies, such as that on Ancient Basques contradict inferences made based on the modern populations.
Posted by: Dienekes at July 28, 2004 01:12 AMI don't know about other European countries, but in England the historical records (rent rolls, etc) show that people moved around quite a lot in medieval and early modern times. (By 'quite a lot' I mean at least 5 per cent migration into a town or village every generation.) The idea that our peasant ancestors just stayed in the same place since time immemorial is one of those romantic myths. Migration of 5 percent per generation would cumulatively produce a lot of population mixing over a time scale of a millennium or so.
Posted by: David B at July 28, 2004 03:14 PM