Archiver > GENEALOGY-DNA > 2006-08 > 1154875534

From: "Ken Nordtvedt" <>
Subject: Re: [DNA] ISOGG 2006 tree
Date: Sun, 6 Aug 2006 08:45:34 -0600
References: <> <>

There are a number of examples in haplogroup I where people on this list
commissioned tests to prove the opposite --- that SNPs which the pros called
non-equivalent turned out to be equivalent. Other tests commissioned by
doubting hobbyists showed that SNPs which had been located downstream of
another SNP turned out to be parallel. The labs and academics make mistakes
about this because they rushed things into print in broad brushstroke papers
in this relatively new subject, and in that rush did not perform all the
cross checks necessary.

Even today in haplogroup I, there is no longer any evidence whatsoever that
SNP P38 should not be collapased back to equivalence with P19,M170,M258 ---
yet it lingers there in the Tree separating I* from I1* because no one
really wants to rename all the clades in I so soon after going through this
6 months ago --- but we should.

The better solution is probably to phase out the naming system (which does,
however, have its associated loss of interesting information) and just refer
to clades by their most downstream known SNP. Even that naming system
changes when further downstream SNPs are discovered, but there is not
cataclysmic change of an entire bush of the Tree's names when upstream SNPs
are added.

I think the folks working on haplogroup R1b may have stories where reported
equivalent SNPs have been split?

Many of us believe that many presently equivalent SNPs will be found to
split by a small population when the databases get much larger. This is
based on estimates of actually how often each male who has ever lived has a
fresh SNP mutation occur in his non-recombining ydna.

On the other hand, some males could very well have had more than one fresh
SNP, or he could be the only male descendant of his great grandfather, so he
has had a chance to accumulate a unique collection of surviving SNPs over
several past generations. A male like this becomes the founder of a
haplogroup with truly equivlant defining SNPs. If his haplogroup has become
large due to the accidents of nature and history, his equivalent SNPs could
all be found today.


----- Original Message ----- > Indeed. As a newcomer to all this I wonder at
how strong the "phylogenetically
> equivalent" wall actually is. How many samples have had all tests applied
> to
> prove (=test) the equivalence?
> --
> John Cartmell john@ followed by 0845 006 8822
> Qercus magazine FAX +44 (0)8700-519-527
> Qercus - the best guide to RISC OS computing
> ==============================
> Search the US Census Collection. Over 140 million records added in the
> last 12 months. Largest online collection in the world. Learn more:

This thread: