GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2004-12 > 1104338858


From:
Subject: Re: [DNA] 385a,b Upheaval in R1b (WAS: Can 385a,b = 14,15 be 15/14?)
Date: Wed, 29 Dec 2004 11:47:38 EST


First let me hasten to see that the problems with multi-copy markers are not
of great concern for projects dealing with the genealogical time frame. If you
have a cluster of people who match on most markers, you don't really need to
worry about whether some of them are really DYS385a/b 11,14 while others are
really 14,11.

In fact, I've only encountered one case where the difference might have been
significant for a haplogroup assignment, where a person was estimated to be
R1a but turned out to be R1b on a SNP test. He was not very close to the modal
values for either haplogroup. I wrote this up in some detail last November:

http://archiver.rootsweb.com/th/read/GENEALOGY-DNA/2003-11/1068521518

That was before Whit came out with his haplogroup predictor. If someone has
time to run Don's numbers through the predictor, that could be interesting.

As far as population geneticists are concerned, I've seen articles where the
authors eliminate multi-copy markers from their phylogenetic analyses. I've
always recommended omitting DYS464 from Fluxus calculations, too. The modal
values for R1b are 15-15-17-17, which makes it particularly problematic to know
which marker changed if someone has 14-15-17-717 or 15-15-16-17.

For Ken's work on Haplogroup I, I'm not familiar enough with the ins and outs
of all his analysis to comment on whether he should eliminate the multi-copy
markers. But I've always been a bit uneasy with making deductions from modal
values which are close to each other, even when they are on distinct markers.
It gets very fuzzy when results could be identical by descent (IBD) or
identical by state (IBS).

Ann Turner - GENEALOGY-DNA List Administrator
Search or Browse the archives, Subscribe or Unsubscribe at
http://lists.rootsweb.com/index/other/DNA/GENEALOGY-DNA.html






This thread: