GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2003-11 > 1069234145


From: "Malcolm Dodd" <>
Subject: [DNA] END of Thread for me re Viking ancestry
Date: Wed, 19 Nov 2003 09:29:05 -0000


Our excellent list administrator, Ann previously posted this

>William Curry, who had his Y chromosome tested at Oxford Ancestors, made
some
interesting observations about the distribution of "I" haplotypes at the
Y-STR database http://ystr.org. His haplotype (in Y-STR order 19; 389i;
389ii;
390; 391; 392; 393) is 14-12-28-22-10-11-13.

-> . . haplotype with DYS390=23 (instead of 22).

He checked the distribution of these two haplotypes in "Anglo Saxon"
territory (Netherlands, Northern Germany, Denmark, and Estonia) and "Norse
Viking"
territory (Norway, Sweden, and Finland). There were a total of
858 samples in the
Anglo Saxon territory, and 1164 in the Norse Viking territory.

The distribution appears to be lopsided -- in other words, the "22" and "23"
don't look random, but show some population structure.

"22" "23"
A-S 60 16
N-V 39 130

I plugged this numbers into the chi-square formula, and the probability of
this distribution is << .001 (assuming the chi-square
is the appropriate test to
use).

This is somewhat analogous to the DNAPrint method for finding "Ancestry
Informative Markers." They look for markers
which have higher frequencies in some
populations than others.>

MY ORIGINAL PROPOSITION
I am merely trying to point out the fact that someone from the British Isles
with "I" haplogroup is just as likely to be Anglo-Saxon as Viking.
COMMENTS BY OTHERS
The probable reason why Family Tree DNA (as well as myself) contest
your assertion about haplogroup I is plain and simple - because you are
entirely wrong.
CONCLUSION -
1 There is SOME population structure.
2 In the World database for DYS390 there are 14 "Anglo-Saxon" DYS390 - 22
and there are 37 "Viking" DYS390 - 23 for the England (London) population.

We can get a feel for origins (SOME structure) but we cannot be specific or
pedantic - as some web sites both Genealogy and Test Companies tend to be.
Malcolm


This thread: