View Full Version : French and German on the X chromosome

07-27-2017, 09:27 PM
French and German on the X chromosome. Only appears at 50% speculative so not sure how seriously I should take it. I know complete ancestry up to 3gt level which is British.


07-27-2017, 09:51 PM
Eupedia has these maps (http://www.eupedia.com/europe/autosomal_maps_dodecad.shtml#French_German) of 23andme categories, maybe it can help.

07-27-2017, 11:09 PM
It's interesting that one of the places it peaks is Switzerland. I know of no Swiss ancestry , but a location search for Switzerland on Ancestry gives me 11 pages of matches .

07-27-2017, 11:36 PM
It's difficult to know how to read the X chromosome where Ancestry Composition is concerned. On the one hand, you're dealing with the one chromosome that is basically "self-phased", if you're male.

Because a male has only one X chromosome, it doesn't actually need to be phased to begin with. Yes, I'm aware of the presence of PAR1 and PAR2, but these should make no real difference. The total length of PAR1 is only about 2.7 Mb, and PAR2 is only 330 kb. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4102438/

In any case, even if these regions did make a difference, it should only be at the extreme ends of the chromosome. So to phase a daughter with an attached father, nothing more should be needed that just subtracting out the father's SNPs. These SNPs would represent the daughter's paternal X chromosome, and the remaining SNPs would represent her maternal X.

There is therefore no reason I can think of why a daughter's paternal X chromosome shouldn't have exactly the same calls in Ancestry Composition as her father. Yet for my sisters, this is not the case.

There is broad agreement -- that is, the first 3/4 of my father's X chromosome is identified as "Southern European" and the last 1/4 is "Northwestern European"; and the same is true for my sisters. Yet at the "country" level, they do not get the same results.

The majority of my father's "Southern European" on the X chromosome is further classified as "Italian". My sisters have no Italian on their paternal X chromosome, however, though two of them do have a "Balkan" segment. The third sister can't be classified any more precisely on her paternal X chromosome than just "Southern European".

I expect that someone will suggest the cause is "interference" from their 2nd X chromosome, but I can't understand why this should be the case for the X chromosome. In the case of autosomes, there may be some uncertainty about the correct phasing due to ambiguous SNPs -- for example, if both parent and child have "AC". This cannot be true of the X chromosome, where except for the tiny PAR1 and PAR2 a man has only 1 SNP at every position on the X. No ambiguity exists.

So once again, why do daughters not have exactly the same results on their paternal X chromosomes in Ancestry Composition as do their fathers on their only X? I mean, obviously there is some sort of "interference" from my sisters' second X -- but why?

EDIT: I apologize that this may not be exactly on point, except that I believe it does speak to something "odd" going on with the X chromosome as it relates to Ancestry Composition.