DNA-R1B1C7-L Archives

Archiver > DNA-R1B1C7 > 2007-08 > 1185978607


From: "David Wilson" <>
Subject: Re: [DNA-R1B1C7] Confusing classifications
Date: Wed, 1 Aug 2007 07:30:07 -0700
References: <080120071305.9032.46B0851F0002D7810000234822068246939C029D9A0D01040D0E9F@comcast.net>
In-Reply-To: <080120071305.9032.46B0851F0002D7810000234822068246939C029D9A0D01040D0E9F@comcast.net>


I know this seems inconsistent in a big-picture view, but it is probably
consistent with the FTDNA projection algorithm, whose details are
proprietary and which is undoubtedly fairly complex.

I think the reports you provide say there is probably NO chance (which is to
say no observed or likely instances of convergence) that would permit a
12-marker haplogroup with 390=25, 385b=13 and 392=15 to be anything other
than R1b1c7. I happen to believe there is no chance that a 25-13-14
haplotype (same markers) could resemble Niall by convergence, but if their
algorithm says there is a possibility, I'll live with that for the time
being. We know that there are non-R1b1c7 haplotypes that share two of those
three markers, so maybe their algorithm has a margin-of-error element to it.

As they say, not everyone has had a DeepSNP test. As those tests increase,
the confidence levels should increase for their projections, at which point
we should see more projections at the more refined clade levels.

As an aside, I have noticed is an odd disagreement between the projected
clade on personal pages and the projected clade on compiled data tables.
There are some members of the R1b1c7 project who are projected R1b1c7 on
their personal pages but still appear as R1b1 or R1b1c on the project
summary tables available on the public web site at FTDNA or the mutation
colorization table I mentioned recently.

David Wilson



-----Original Message-----
From:
[mailto:] On Behalf Of
Sent: Wednesday, August 01, 2007 6:06 AM
To:
Subject: [DNA-R1B1C7] Confusing classifications

What don't I understand? I have two new members, both tested at 12
markers only, whom I assigned to my Northwest Cluster. One has the
identical haplotype as "Niall," and FTDNA awarded him the Niall green icon.
Yet his suggested haplotype is R1b1. When I asked FTDNA why he was not
suggested to be R1b1c7, I received this reply:
"Due to convergence, some 12 marker haplotypes who match the Niall of
Nine Hostages are not also R1b1c7 so matching the Niall lineage on 12
markers does not automatically mean belonging to R1b1c7. The males who have
had 12 markers tested and have received SNP tests are used in the comparison
of 12 markers for the purposes of predicting the haplogroup for a sample.
Not everyone who has tested in the SNP database has had the Deep SNP R1b1
performed on their sample and the R1b1s that are shown, now, outnumber the
R1b1c7s. As this changes with more people testing the prediction will
likely change to R1b1c7 on the personal pages and eventually in the chart."
OK, The other member, who came in through the Genographic project, does
not have the green icon, is only an 11/12 match with Niall (he has 15 at 292
while Niall has 14), but is suggested to be R1b1c7. Wouldn't the convergence
caveat also apply to him?

<<SNIP>>

This thread: