GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2008-05 > 1211424052


From: "Tim Janzen" <>
Subject: Re: [DNA] S21/S28 Split+m223 stuff
Date: Wed, 21 May 2008 19:40:52 -0700
In-Reply-To: <ea3bd9560805211759y332b41c7x96a78d5dd7c987f7@mail.gmail.com>


Dear David,
There is no proof in the calculations that I presented that the age
of S28 is 2500 years older the age of S116. Some of the calculated ages I
gave for S28 were younger than some of the calculated ages for S116. It may
take some time to refine these dating techniques using STR data and to
compile adequate datasets for some of the more uncommon subclades, but I
think we should remain open to what the results are telling us. Also keep
in mind that testing for the TMRCA for various subclades like I was doing is
independent of when Y Adam occurred since it is dependent on the mutation
rates for the markers but not on the age of Y Adam. The SNP counting method
is rooted in the age of Y Adam, but not the STR methods Ken and I have been
using. We are going to be limited in our use of the SNP counting method
until we have multiple complete Y chromosomes sequenced from males in all of
the subclades. Until that point in time we will have to be content with age
calculations based on STR data. My personal feeling is that the mutation
rates for some of the markers may be somewhat off, but that the sum of the
mutation rates as used in Ken's formula is probably not going to change much
over the long run even as mutation rates are better refined.
On reconsideration I also agree with Larry and Ken that we don't
know that S21 is older than S28 and the ages from the calculations would
suggest that S21 could actually be younger than S28.
Sincerely,
Tim


-----Original Message-----
From:
[mailto:] On Behalf Of David Faux
Sent: Wednesday, May 21, 2008 6:00 PM
To:
Subject: Re: [DNA] S21/S28 Split+m223 stuff

It underscores my point that these type of computations are not reliable if
only for the reason is that there is little internal consistency and hence
we cannot place the slightest reliance on their validity. It makes
absolutely no sense that the child (S28+) is 2500 years older than parent
(S116+).

The litmus test for computational methods is whether they are consistent
with the widely accepted age of circa 60,000 to 70,000 years to Y-Adam. It
is just common sense that one needs to root calculations there and go
downstream haplogroup by haplogroup until reaching the twigs where for
example S28 lies. To my mathematically
challenged mind it does not make sense to start with the twigs and move to
the root - it needs to be the other way around.

David K. Faux.

This thread: