GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-11 > 1289598931


From: "Ken Nordtvedt" <>
Subject: Re: [DNA] Fwd: Fwd: First Neolithic Y-DNA published
Date: Fri, 12 Nov 2010 14:55:37 -0700
References: <AANLkTi=R-uQYBmw7vDz3Tsfjzv7_c6JD__MT_n4i6XvE@mail.gmail.com><625030.15518.qm@web25901.mail.ukl.yahoo.com><AANLkTin314jXQN2fOafYjTPKv=kBE8CmNG-bh8FPRoG-@mail.gmail.com><AANLkTimAuM7fw-Byb1M=kafpZ4BZZKGxN6O1QxuYCQQY@mail.gmail.com><016901cb8298$82e64500$bb579245@Ken1><AANLkTin_Tao5XEW7uyUqhNy9Lt4KxknRY9LiS-Rsa0op@mail.gmail.com>


----- Original Message -----
From: "Mike W" <>

I think Vince Vizachero's graphic on R-U106
using the Myres data is an interesting one for those who are most
interested in R-U106.

[[ Perhaps it is being very familar and comfortable with numbers, but i am
staring right now at the underlying table from which the interpolated
graphic is produced. I feel on much firmer ground with the data (and can
see the numbers which drive the graphic), although I think there are serious
limitations to it including incorrect statistical confidence intervals. But
these interpolation algorithms take fuzzy or spotty data and make even
fuzzier pictures from it. I am reminded of an early paper on haplogroup I
which involved absolutely no British Isles data at all, yet included a
graphic from interpolation algorithm which showed some haplogroup's
frequency throughout the British Isles as well in some other countries where
no data existed in the paper's tables. You would at least have thought the
authors would have blacked out from the map the places where no data
existed.

But probably what's at the bottom of my suspicions concerning the S21
haplogroup data is the over-interpretation of the small quantities of it
showing up on the eastern fringes. Throughout the last several centuries,
just about every country to the east and southeast from Germany and Austria
has been accepting repeated flows of German immigrants in order to beef up
their economies (enrich the rulers). The cumulative impact of this is to
bring some haplotypes from the Germanic lands into these eastern regions.
And when we are talking about data of small quantities in these regions,
this impact can be non-negligible concerning both the frequency and variance
of what is found.

Anyway; there are a number of y haplogroups whose places of highest
frequency are basically in Germany but are found sprinkled all over eastern
Europe in diminished frequency. The challenge is to figure out a way to
figure out how much of that can be explained by people flow of the last
several centuries.

And we have not even touched on the question of whether the Myres et al
variances and ages are supposed to be coalescence ages or actual tmrcas?
The former of course are only indirectly related to something specific like
age to founding events; such constructs also fold in some generally unknown
post-founding demographic history. Do you have an understanding of exactly
what kind of variances they are using in their paper?

Ken ]]



This thread: