GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-08 > 1123306283


From: OrinWells <>
Subject: Re: SMGF [was Re: [DNA] FTDNA Genetic Distance Calculations...]
Date: Fri, 05 Aug 2005 22:31:23 -0700
References: <IGEOKAGLHNEKPCKPADIGOEIKKEAA.bbailey.lowedna@baileyconnection.com><REME20050803190044@alum.mit.edu><6.0.0.22.0.20050805102153.06137ba8@wells.org><REME20050805150049@alum.mit.edu><6.0.0.22.0.20050805123622.052bd670@wells.org><REME20050805213844@alum.mit.edu>
In-Reply-To: <REME20050805213844@alum.mit.edu>


At 06:43 PM 8/5/2005, you wrote:
>Orin wrote:
> > In this particular case, to confirm the mutations, this individual was
> > tested using a variety of tests.
>
>That's rather vague. Exactly what sorts of tests did you do? Buccal,
>for starters, of course. Blood? Semen? Only the latter would give
>you conclusive proof that it wasn't a somatic mutation.

I didn't do them. This whole thing was discovered during the BYU phase of
our project and the BYU researchers looked at all sorts of cellular level
explanations for what they were seeing. I don't even know the different
tests that were conducted other than in general terms. I will have to find
the report to refresh my memory on it. They did look at blood samples and
apparently took samples from some uncommon areas of the body. I will hunt
down the report and give you the summary of tests as defined.


> > I disagree. You can not rule this out unless you were to run a large test
> > utilizing three generations. What we are seeing in the father-son testing
> > is simply a confirmation of what we are seeing in the general
> > population.
>
>A pedigree study inevitably covers more than three generations.

Yes, but that does not necessarily apply here. You are generally not
testing at any but the last node of a tree "branch" because the earlier
ancestors have all died. It depends on the testees, their relationship to
each other and where any mutations actually happened - something you seldom
will be able to identify even in a large study without a lot of luck.

>The
>point I was making, but didn't state explicitly, is this: a father-son
>study will be sensitive to all three types of mutations that you are
>proposing, (1) "permanent" mutations, (2) "transient" mutations, and
>(3) reversions of the latter.

Sorry John, this does not make sense to me. There is no way a father-son
study can be sensitive to any of these conditions. If a transient mutation
happens between the father and the son the only thing you will see is a
mutation in the DNA of the son. 1) Even if the mutation is permanent, you
can not know this either unless you can see the DNA from a later generation
(not necessarily the grandson, but someone down stream). 2) You can not
possibly determine that this is transient unless you also examine a
grandson through the son. If you are looking at a father and son DNA set
you can not tell the difference between 1 and 2. You certainly can not
tell if there is or will be a revision of either.

A pedigree study spans a much larger scope than a father-son study. But
even if you are testing a single descendent of a branch against another
where one shows a mutation and the other does not you can not tell whether
the mutation happened at the node just below the common ancestor or in the
tested participant himself, or, for that matter, which of the two even has
the mutation unless you have more samples from the same family exhibiting
the mutation. And you can not tell if the mutation is permanent or
transient by the single sample. If, on the other hand, you have two
descendents of a common ancestor who exhibit the same mutation you are
probably nearly 100% certain it happened at the common ancestor and is
certainly permanent. The odds of the same mutation happening independently
down two branches of the same family is probably astronomical.

>Obviously, types 2 and 3 must occur with equal frequency in
>equilibrium. A pedigree study will not see any transient mutations except
>those occurring in the testees themselves and will not see any reversions
>at all. Therefore, the general agreement between pedigree studies and
>father-son studies demonstrates
>that types 2 and 3 cannot be at all common.

I submit the number of samples involved in any father-son project to date
is much too small to arrive at such a conclusion especially if it is not
possible with just the father-son samples to identify transient versus
permanent mutations. All you can say is that there was some rate of
mutation. Assuming, of course, that the pairs in the father-son study are
not a bunch of closely related individuals beyond the father-son sets.


> > The question then is how many of the mutations seen in the
> > father-son testing might have actually been transient and at the same
> ratio
> > as in the general population. I would agree it is probably a very small
> > number, but it is also probably finite.
>
>The issue is whether the number is big enough to change the statistics
>in an observable way. If not, then "finite" is effectively the same
>as "zero".
>
> > There are some interesting implications here. If this transient behavior
> > also applies to DNA on other chromosomes, and there is no reason why it
> > wouldn't, it is quite possible various identifiable traits appear and
> > disappear from generation to generation.
>
>As Ken pointed out, there is a known mechanism that does this and has
>been observed, namely, recombinant duplication. In this process, a
>sequence of DNA is copied over to replace the homologous sequence on
>the other chromosome of the pair (or, in the case of Y DNA, on the
>other side of a palindrome). Obviously, this cannot occur for
>single-copy Y markers. Also, it doesn't occur immediately, but only
>occasionally. Otherwise, it's similar to what you propose.
>
> John Chandler
>
>
>==============================
>Search the US Census Collection. Over 140 million records added in the
>last 12 months. Largest online collection in the world. Learn more:
>http://www.ancestry.com/s13965/rd.ashx


Orin R. Wells
Wells Family Research Association
P. O. Box 5427
Kent, Washington 98064-5427
<>
http://www.wells.org
Subscribe to the "Wells-L" list on RootsWeb

This thread: