GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-07 > 1183310781


From: "William Hurst" <>
Subject: Re: [DNA] Mtdna SNPs?
Date: Sun, 01 Jul 2007 13:26:21 -0400
In-Reply-To: <4687CD71.3050705@mchsi.com>


Hi Rebekah, Ken and all,

You might also look at the first full paragraph on p. 1091 of the paper
where it discusses the probability that parts of the control region are not
as junky as one might think. Something besides heteroplasmy has to explain
why some HVR positions never mutate and some mutate often.

The emphasis of the paper is on haplogroups, but I'm more interested in
subclades. I only see my K Project members after their haplogroup has been
determined by the procedure outlined in the paper. But then the most
important part of mtDNA to determine the subclades of K is not HVR1, not the
coding region, but HVR2. Over 60% of K is in K1a which is defined by 497T -
another mutation which is virtually a UEP. K1c is determined by 498-,
otherwise found only in small numbers in L. These are extremely stable
mutations; more stable than the 10550 used by FTDNA/Genographic to determine
K. I could list more of these, but they are of lesser importance.

I posted a memo this morning which I sent to the K Project members yesterday
concerning to Genographic paper. There might be something of interest to
those not in K:
http://freepages.genealogy.rootsweb.com/~wrhurst/mtdna-k/genopapermemo.htm

Bill Hurst

>:-) Welcome to mtDNA... But we only use the non pathogenic coding region
>SNPs and they are much more reliable than the control region SNPs. I
>think part of the problem though is that some of the coding region is
>more um vital/non-redundant than other parts and thus looks slower.
>
>Of course when we have a few thousand complete yDNA chromosomes
>sequenced we might have a much different opinion of yDNA SNPs.
>
>Rebekah
>
>Ken Nordtvedt wrote:
> > Now that's peachy --- using SNPs in coding regions subject to normal
> > selection as identifiers of mtdna haplogroups. Doesn't seem like a very
> > lasting tag to me?
> >
> > Ken


This thread: