GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2004-10 > 1098807788


From: Doug McDonald <>
Subject: Re: [DNA] DYS 388 Mutation Rate?
Date: Tue, 26 Oct 2004 11:23:11 -0500
References: <web-1587882@be-1.cluster1.bresnan.net>
In-Reply-To: <web-1587882@be-1.cluster1.bresnan.net>


ken nordtvedt wrote:

> Doug McDonald's "reverse engineered" mutation rate estimate for DYS 388
> is one of the slowest --- .0005 per generation. And my experience
> tabulating haplotype counts from databases tends to confirm the
> stability of DYS 388.
>
> But in the recent paper on haplogroup J and its subclades by DiGiacomo
> et al in Human Genetics 2004, they inferred some individual marker
> mutation rates from analyzing the found distributions of J haplotypes in
> Europe and Mideast and employing some simulation software (Batwing?).
> They concluded that DYS 388 had a higher mutation rate than other
> familar loci such as DYS 392, 393, etc. I believe their estimated rate
> for DYS 388 was almost an order of magnitude greater than the McDonald
> value.
>

Don't call that (.0005) the "McDonald value" ... it is,
as you correctly said, my reverse engineered Sorenson value.

And the slowest ones are the least accurate. It is,
however, certain tied for slowest according to Sorenson.

Now there are is in fact a set of actual data derived by
me alone, with no reference to Sorenson. This data is
calibrated as to average rate by the various published
father-son and pedigree values and the relative values
are derived using Batwing on all R1b people in the British Isles.

In that data 388 is not the slowest ... 426 is, at 0.0003 ... but
388 is very slow, at 0.0007.

But this is for R1b.

Note that rumor is that after Saturday there will be
"FTDNA" numbers.

It would be nice if, if that in fact happens, for somebody
to question them closely on how they got the numbers.

Doug McDonald

Doug McDonald

This thread: