GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-06 > 1118202247


From: (David Faux)
Subject: Re: [DNA] Re: Genographic Project Upload Tab - Now Available!
Date: Wed, 08 Jun 2005 03:44:07 +0000


(I don't know why that virus check message previously so will resend again)

I am not so sure Bonnie. The rate of uncertainty is probably comparable for
mtDNA and Y-STR haplotypes.

I have just gone through the lengthy task of taking a string of nucleotide bases
for 503 Shetland samples, locating those that deviated from the CRS and then
using the best set of algorithms available (my contacts are the top mt
researchers in the field). I was able to assign haplogroups to about 80% of the
sequences, some were entirely impervious since they have apparently not been
seen in the literature before or are ambiguous (could be either one haplogroup
or another).

Simoni et al. (2000) found precisely the same problem in the attempt to assign
2619 European sequences to a haplogroup. They observed that "78% of the
sequences of our database could be unambiguously assigned to one haplogroup".
Of these 17% total were assigned to "Other" and the rest to macro groups such as
IWX which I would argue is questionable.

Nothing much has changed, I found about the same figures but my approach was to
just leave them as "unknown" although most were probably H. I therefore wonder
where NGS is going to obtain a foolproof way to pigeon hole all sequences -
frankly it is not possible without testing the coding region which adds
immensely to the expense unless there is a new development about which I am
unfamiliar (entirely possible), but I would like them to describe how they
derived these Y and mt algorithms (now that I uploaded 4 to GenoGraphic and am
pleased to be officially part of this landmark study).

David F.



-------------- Original message --------------
<snipped>
> All along, FTDNA has been willing to perform further testing of mtDNA
> beyond analyzing HVR1 and 2, when necessary, without charge, to
> determine haplogroups. Their system for determining haplogroups and
> deciding who needs further testing has now been renovated (personal
> communication, Bennett Greenspan). In general their philosophy is that
> they feel sure of everyone's reported mtDNA haplogroup, either through
> analysis of HVR mutations or further testing. I have found errors in
> their system in the past, but I believe that things will be improving
> now, with their Genographic Project involvement, which is bringing in
> new scientists. The point is that they don't consider your reported
> mtDNA haplogroup an "estimate" the way the Y-DNA haplogroup can be.
>
> Bonnie Schrack


This thread: