Archiver > GENEALOGY-DNA > 2005-06 > 1118466334

Subject: Re: [DNA] High Mutation Rate in Second Cousins
Date: Sat, 11 Jun 2005 01:05:34 EDT

It may help also to keep in mind the papers & data supporting the notion
that the average mutation rates & direction of individual markers can vary as a
function of the actual allele length. No one has hazarded general estimates
(that I know) of the increased rates near the end points of the observed total
interval of allele length variation, but Dupuy, e.g., has published specific
rates based on the 1766 Norwegian father/son pairs he sampled in his 2004
paper. (You can find pointers in our List archives - discussed several times
over the past several months). Also, it's observed that the mutations of allele
length near the top end of the interval tend to mutate down, and allele
lengths near the bottom end tend to mutate upwards.

There's also the correlation observed between haplogroup and observed
mutation rate of some markers, but on closer inspection, this was due to the
stronger relationship of allele length & rate. Some haplogroups have "typical"
allele lengths of, e.g. DYS 455, closer to the end points of the generally
observed interval. Discussed in List archives recently.

I think it may true that in some specific time-windows, some paternal lines
may have tendencies for some specific markers to mutate faster or slower than
other families observed in the same time-window. Or is it the case that as
we see more & more actual test results, we will see all kinds of
probabilistically feasible, but "unusual-appearing" cases.

CDYa & 442 are moving fast enough to delineate 3rd cousins in my
11-generation Humphrey lineage. I'm not surprised that it can also be rock solid for
more than 11 generations in other paternal lines. 570 is moving fast enough to
delineate 2nd cousins in the Liles project. 576 is rock solid in their line.
Again, no surprise.

Mike ...

In a message dated 6/10/2005 8:18:47 PM Pacific Standard Time,


Thanks. Considering this an other sources it would seem prudent to exclude
these markers when looking for matches to other surnames for those who are
brickwalled. What is odd is that the also fast mutating CDYa and b have held
fast in every generation at least in the three Fauxes tested so far. I know
that the ever popular concept "chance" can easily be invoked here; and that it
is a natural human tendency to want to see meaning where there may be none -
but red flags are popping up here.

David F.

-------------- Original message --------------

> 570 and 576 can move around quite a bit. I think you have run into a
> situation in which fast mutators have done what it is they do. They just
> happened to do it within the last few generations of a family that you
> follow closely.
> David W.

Admin: FTDNA Humphrey project
Co-Admin: FTDNA Liles project
multi-family project:Humphrey, Liles, Morton

This thread: