GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-02 > 1172154321


From: Vincent Vizachero <>
Subject: Re: [DNA] TMRCA
Date: Thu, 22 Feb 2007 08:25:21 -0600
References: <KHEKIJEABJGJEKDPFEDMAENGDDAA.elizabethod@eircom.net><898755FF-A280-4139-AE30-4B26868F55B4@vizachero.com><45D49A42.7000105@scs.uiuc.edu>
In-Reply-To: <45D49A42.7000105@scs.uiuc.edu>


On Feb 15, 2007, at 11:37 AM, Doug McDonald wrote:

> Not per se. But there is plenty of evidence that higher numbered
> alleles for a given marker mutate faster than lowered numbered
> ones. Hence,
> a haplogroup with typically higher allele counts will have that marker
> mutate faster than a haplogroup with typically lower allele counts.
> This is
> only a simple relationship for "simple" markers, that is, ones with
> a single unbroken run of a single variable unit.


I know this thread is a little old, but it took me a while to dig
into the data.

The effect of allele length on mutation rate seems to be highest for
single-copy tetranucleotide markers (e.g. DYS393, DYS455, DSY57, etc)
and the slope I plotted (using Chandler's rates) is an increase in
the rate of mutation of 0.001 per repeat.

But there are a dozen or so single-copy tetranucleotide markers the
typical STR panels. The sum of alleles over all such markers does
not vary much by haplogroup, at least in the data I examined. The
most extreme example I could find was I1a, which had an average
allele value of 0.50 repeats greater than average.

This would theoretically translate into a mutation rate 0.0005
greater than average for the affected markers, or about 0.00017 over
the whole panel. This is an effect of only about three or four
percent, which is not very significant given all the other errors and
uncertainties.

Vince

This thread: