GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2009-12 > 1260185851


From: "Lawrence Mayka" <>
Subject: Re: [DNA] R-U152 and R-L21 on the European Continent
Date: Mon, 7 Dec 2009 05:37:31 -0600
References: <59915.56775.qm@web81103.mail.mud.yahoo.com> <06F79C98-398C-4200-8D9C-5BA976ABD233@vizachero.com><ea3bd9560912061730y581da805v54dec4a73733efee@mail.gmail.com><006701ca76ef$d0ab50b0$7201f210$@org> <003201ca76f3$ee2dcd30$6400a8c0@Ken1><91E2D25B95804417B98E786B9CB9E69B@bobPC>
In-Reply-To: <91E2D25B95804417B98E786B9CB9E69B@bobPC>


The Zhivotovsky camp does not significantly dispute the known germline
mutation rates for often-used STRs, though one might get that impression
from their misleading talk of "effective" or "evolutionary" mutation rates.

They insist, though, on the application of a fudge factor, roughly x3, to
account for (a) the interval from the first occurrence of an SNP to the MRCA
of the sample, and (b) unaccounted population structure within the sample.
Nordtvedt's interclade method eliminates these two reasons to fudge, by
bracketing the age of the SNP between a lower and upper bound, calculated
exclusively from the known population structure.

When M458 was announced, we were thus easily able to bracket its age. An
interclade calculation between P vs. N types (both M458+) gave us roughly
3200 years as a lower bound, and an interclade calculation between P+N vs. K
types (K is M458-) yielded roughly 4300 years as an upper bound--all using
67-marker haplotypes.

I have not yet seen any response to this method from the Zhivotovsky camp,
which instead clings entirely to the intraclade method with its known
inaccuracies.

> From: [mailto:genealogy-dna-
> ] On Behalf Of Bob Bootle
> It is a real nuisance that
> mutations vary
> rates of mutation vary
> and average mutation rates vary.



This thread: