GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-08 > 1123195623


From: "gareth.henson" <>
Subject: Re: [DNA] SNP mutation rates
Date: Thu, 4 Aug 2005 23:47:03 +0100
References: <IGEOKAGLHNEKPCKPADIGOEIKKEAA.bbailey.lowedna@baileyconnection.com> <REME20050803190044@alum.mit.edu> <090b01c59884$6238c5c0$0200a8c0@c452380a> <REME20050803194536@alum.mit.edu> <097001c59905$fff2b6f0$0200a8c0@c452380a> <016401c59931$32d2cc10$84d62f50@Masterbedroom> <001a01c5993f$83f4e020$71509045@Ken1>


Ken

consistent = not frequently recurring or back mutating
regularly detectable = not private or frequently liable to null results due
to primer problems

Have they thrown any away? I suspect some never got through all the
necessary quality controls (e.g. see the Perlegen Hinds et al paper, also a
similar filtering process for STRs in the Kayser paper) There are gaps in
both the M and P series.
There is one called 50f2/C (or DYS7C) which appears in some of the 1990s
papers and then was found to be so recurring as to not be very useful as an
SNP. P25 is taking a few knocks and there is a case for taking it off the
tree.

I don't think we would get a completely different tree if the process
started from scratch and found a different set of SNPs. There might be minor
differences but I think we would still have the same overall shape.

Gareth



----- Original Message -----
From: "Ken Nordtvedt" <>
To: <>
Sent: Thursday, August 04, 2005 10:57 PM
Subject: Re: [DNA] SNP mutation rates


> I was under the impression the SNPs we have are basically just what were
> found; they were not chosen? By what technique do they efficiently find
the
> SNPs "that show consistent and regularly detectable patterns of
variation"?
> (I'm actually not sure what you mean by that?) Does that mean they have
> thrown away a bunch that don't "show ...........patterns of variation."
>
> This issue has concerned me for some time: do the (deep in time) SNPs
that
> just happened to have been found determine the divisions of the tree? So
> that if all the work of the last decade or so were lost, and the SNP
hunters
> had to start all over again, would we get a substantially different tree
by
> the time we had discovered the first couple hundred SNPs the second time
> around?
>
> Ken


This thread: