Archiver > GENEALOGY-DNA > 2010-03 > 1270077661

From: (John Chandler)
Subject: Re: [DNA] Questions about alternate means of searchingDNA Genealogy -creating a modal haplotype from a few samples
Date: Wed, 31 Mar 2010 19:27:53 -0400
References: <d8c.1d027bc.38e4b9ac@UNKNOWN><><>
In-Reply-To: <>(message from Johnathan Clayborn on Wed, 31 Mar 2010 08:21:07 -0700)

John wrote:
> My 389-I is 11,
> which according to SMGF happens 0.38% of the time, and my 389-II is 27,
> which happens 0.95% of the time.

This is a perfect example of the need to watch for correlations between
rare values. As it happens, DYS389ii is the sum of two pieces, one of
which is DYS389i. That means your information is really 389i=11 and
389ii-i=16. The former is rare, but the latter is very common. In
this case, only the 389i is unuusual.

More generally, though, you should always focus on *all* of the
markers you tested and not restrict your view to just a few. You
don't know how far back in time your rare alleles first appeared in
your line, and so you might be "barking up the wrong tree" if you
guide your search by them. If, for example, your grandfather was the
one who experienced a 12->11 mutation on DYS389i, it would be
completely useless to look for not-immediate-relatives with 11.

> From what I understand the markers at 389 also have very low mutation rates,

Not really. Lower than average, but not much lower. That's why you
really do have to worry about the time of acquisition of your rare
alleles. Indeed, you yourself could be the first in your line to
display any of these. The problem here is one of finding relatives,
not of identifying a criminal suspect -- that means you are looking
for a range of neighboring alleles, not for an identical match. When
it comes to a range of values, every marker you compare has an equal
(well, almost exactly equal) contribution to the overall pattern.

John Chandler

This thread: