GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2009-12 > 1260532514


From: Vincent Vizachero <>
Subject: Re: [DNA] R-U152 and R-L21 on the European Continent
Date: Fri, 11 Dec 2009 05:55:14 -0600
References: <200912110730.nBB7UqKV015592@mail.rootsweb.com>
In-Reply-To: <200912110730.nBB7UqKV015592@mail.rootsweb.com>


I don't think Ken was ignoring this point, but rather looking past it
to the day when we can model this non-linearity accurately and
(therefore) treat even the fastest markers as useful instead of simply
ignoring them for older estimates (which, today, is really the only
approach we have).

Even if markers are not accumulating variance linearly, as is the case
with fast markers over longer periods, as long as they are
accumulating variance somehow then we can theoretically find a way to
use them. Measuring the manner in which linearity begins to fail is,
at root, an empirical undertaking that should be possible.

VV



On Dec 11, 2009, at 1:30 AM, Tim Janzen wrote:

> However, as I pointed out in various messages this summer,
> including a lot of fast mutating markers in the group of markers
> used for
> interclade TMRCA estimates for subclades or haplogroups 10,000 or
> more years
> old will invariably skew the ages to be lower than they are likely
> to be in
> actuality.


This thread: