09 October 2011

Enhanced autosomal DNA sharing

Warning: this item is more technical than typical in this blog. If you're not interested in genetic genealogy you may want to skip it. 

The results I've received from autosomal DNA testing by FTDNA and 23andMe have puzzled me as I seem to have more matches with people of Ashkenazi Jewish (AJ) origin than the fraction in my ancestry would justify. I've wondered if perhaps there isn't something in AJ DNA that makes it less liable to mutation than average.

A new article "The Architecture of Long-Range Haplotypes Shared within and across Populations" in the journal Molecular Biology and Evolution goes some way to explaining the phenomenon.

The article abstract, available at http://m.mbe.oxfordjournals.org/content/early/2011/10/06/molbev.msr133
states that "Ashkenazi Jewish individuals are all connected through transitive remote family ties evident by sharing of 50 cM IBD to a publicly available data set of less than 400 individuals."

The article attributes this to "a rapid expansion of about 950 diploid individuals 23 generations ago to the current hundreds of thousands" (in the AJ population). With marriages occurring within that population there is 30 times more commonality in DNA between people in the AJ population than in a comparable European sample.

The study identified a segment of DNA on chromosome 6, between 25 and 35 MB, which includes the human leukocyte antigen (HLA), as being a region of particularly high sharing within the AJ population. Several other regions of enhanced DNA sharing within various population groups are identified.


1 comment:

Ken McKinlay said...

I think this is one reason why I don't yet have faith in the use of DNA testing to determine genealogical connections. This is still a very young science and the scientists themselves are still trying to understand it. I do think it has value to exclude like-named family groups, i.e. all the Jones from the same village just might not be related.