Archiver > GENEALOGY-DNA > 2010-02 > 1266805880

From: Bonnie Schrack <>
Subject: Re: [DNA] Y chromosome SNP occurring in More than one PaternalLineages
Date: Sun, 21 Feb 2010 21:31:20 -0500

Ken rightly pointed out that what I said was exaggerated:

> If a SNP were *not* a product of recombination, and were found to have
> mutated in more than one haplogroup, these multiple mutations could be
> caused by an unstable location on the Y chromosome which was prone to
> shift back and forth unpredictably between two alleles. This would
> make such a location entirely unsuitable for use on the phylogenetic
> tree.

I was trying to draw a strong contrast with the more stable recombined
blocks. Actually the SNPs that have mutated more than once aren't
always that unstable, and can certainly be useful in certain contexts on
the phylotree.

As an example of an unstable SNP, I was thinking of P25, where it says
in the ISOGG R tree page,

> P25 is an unreliable SNP with a 2% reversion rate within R1b1. It
> should not be used by itself to classify individuals.

But other SNPs that occur more than once in the tree are not necessarily
ones that are likely to revert. I'm in favor of using them in the
ISOGG tree when evidence indicates that they are reasonably stable.
Sometimes we need every marker we can get to discern subclades.

I'm not sure what the story will be with L69, which was found to have a
back-mutation to T in two members of J2a4*. I hope we'll learn
eventually whether this is a consistent mutation in other members of
their clusters. It will help if more of them will order it.


This thread: