GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-07 > 1120248432


From: Robert Davenport <>
Subject: Re: [DNA] Back Mutation and Parallel Mutation
Date: Fri, 01 Jul 2005 13:07:12 -0700
References: <6.2.1.2.2.20050701112946.037e1910@lafn.org><20050701184723.31981.qmail@web81806.mail.mud.yahoo.com><6.2.1.2.2.20050701115851.03e3c4c0@lafn.org><00d601c57e72$916ae340$b2d2ae51@oemcomputer><REME20050701155424@alum.mit.edu>
In-Reply-To: <REME20050701155424@alum.mit.edu>


I thought perhaps there was a "rubber band" effect, in which there was a
greater probability that it might "spring back" to its original marker
value, then to mutate in a different direction?






At 12:54 PM 7/1/2005, you wrote:
>John wrote:
> > My understanding of backmutations is that they are supposed to be
> > approximately half as frequent as forward mutations.
>
>There is no "backward" or "forward" except in terms of the starting
>mutation, whatever it was. If we neglect the up-down bias, then
>backward and forward mutations should have exactly equal probability.
>Even if we include the up-down bias, it enters only at second order
>into the forward-backward bias. For example, suppose the marker in
>question has a probability of 55% of mutating +1 and 45% -1 -- if
>there is a second mutation on the same marker (with the same bias),
>then the likelihood of the two mutations being in the same direction
>is 50.5%.
>
> John Chandler
>
>
>==============================
>Search Family and Local Histories for stories about your family and the
>areas they lived. Over 85 million names added in the last 12 months.
>Learn more: http://www.ancestry.com/s13966/rd.ashx


This thread: