GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2008-05 > 1209682115


From: (John Chandler)
Subject: Re: [DNA] question re ftdna common ancestor estimates
Date: Thu, 1 May 2008 18:48:40 -0400
References: <235112.80340.qm@web86601.mail.ird.yahoo.com>
In-Reply-To: <235112.80340.qm@web86601.mail.ird.yahoo.com> (message from AlanR on Thu, 1 May 2008 21:29:54 +0000 (GMT))


Alan wrote:
> If a tester has matches at 25 markers at a genetic distance of 1 or
> 2, ftdna seems to imply a common ancestor at a high probability
> within say 2-300 years.

You must bear in mind that FTDNA is doing the calculation with
mutation rates that are way too high, leading to seriously
overoptimistic TMRCA estimates. For 25-marker comparisons,
the appropriate scale factor is 1.7. In other words, if
FTDNA says the likelihood of finding a common ancestor within
8 generations is X percent, you should interpret that as
saying the probability of finding a common ancestor within 14
generations is X percent.

Another factor to consider is the closeness of the haplotypes in
question to the nearest modal haplotype of a major clade. If
they are *very* close to a mode, then the assumption of flat
priors breaks down, and the time estimates need to be scaled
up further.

John Chandler

This thread: