GENEALOGY-DNA-L ArchivesArchiver > GENEALOGY-DNA > 2010-02 > 1266252082
From: "Ken Nordtvedt" <>
Subject: Re: [DNA] TMRCA assessments
Date: Mon, 15 Feb 2010 09:41:22 -0700
----- Original Message -----
From: "Robert Stafford" <>
It [[ "parallel mutation" ]] seems to be used for ancient STRs
> in a theoretical sense.
I have no idea what you mean by "theoretical sense", but the multiple
occurrences of mutations of STRs are certainly facts of life in ancient
On another forum someone asked about why DYS462 equal to 12 or 13 was
"still" considered an indicator of ancestral or derived L22?
He was an exception to the rule of thumb and noted others.
My collection of I1 haplotypes now numbers perhaps 4000. Unfortunately,
only some of them have measured DYS462.
The tree that establishes the history of the 4000 haplotypes since the I1
MRCA contains of order 100,000 father/son transitional sites --- i.e.
mutational opportunities for each STR.
The mutational rate I use for DYS462 is about 1/2000. So the tree contains
of order 50 mutations of the slow STR DYS462.
Most of those 50 or so mutations happened in the very recent generations of
But what was special about one of those 50 mutations is that it happened
very early in the tree's history, and robust sub-trees developed on both
sides of that early mutation.
So that early and especially useful mutation very relibably (but not
infallibly) separates I1 haplotypes into two clades which happen to also
show some other STR statistical differences as well as geographical
distribution differences. The L22 snp mutation must also have happened
rather close to that unusually early DYS462 mutation. But given all that;
it is just one of many mutations of DYS462 which occured in my sample set of
I1 haplotypes. Imagine how many times DYS462 has mutated in the full tree
of the multi-million number of I1 males alive today. But even for the tree
of all millions of I1, the slowness of DYS462 mutation rate means the
additional occurrences of it mutating in the tree will on average be
well-separated in the tree from that special early mutation. One can
calculate how many of the tree's early generations had to go by before a
DYS462 mutation should come along again, and it is very large and the I1
tree population had grown to order 1000. Ken