GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2006-04 > 1146346296


From:
Subject: Re: [DNA] Re: R1b SNP page updated
Date: Sat, 29 Apr 2006 22:31:36 +0100


Ken

no, S26 is a special case - it is the "DYS439 null" phenomenon, and was
discovered precisely because it affected the testing of a nearby STR. Most
SNPs will not be anywhere near a STR and so won't be discoverable in this
way.

My comment just meant that we know S26 is single copy because of the way it
was discovered, and so it can't revert by recLoH.

Gareth


----- Original Message -----
From: "Ken Nordtvedt" <>
To: <>
Sent: Saturday, April 29, 2006 10:23 PM
Subject: Re: [DNA] Re: R1b SNP page updated


> Did you mean to leave the impression that all known SNPs are "associated"
> with STRs? I guess that could be the effective situation if all searches
> for SNPs up until now have been done near STRs, but I did not know that was
> the case, although searches near STRs could be common.
>
> Ken
> ----- Original Message -----
> From: <>
> To: <>
> Sent: Saturday, April 29, 2006 3:16 PM
> Subject: Re: [DNA] Re: R1b SNP page updated
>
>
> > Joe
> > gene conversion is one form of recLOH and it can happen in P25 because P25
> > is actually a 3 copy locus, only one copy of which has a SNP. So the SNP
> > is occasionally reversed when one of the unmutated copies overwrites the
> > mutated one. IMO if this was a newly discovered variation it would not be
> > acceptable as a haplogroup branch defining SNP. (Yes - we have just
> > published a new version of the tree and P25 is still there but despite the
> > many changes we have had to be quite conservative in places).
> > S21 is different, it is a single copy locus and I do not believe it has
> > been observed to revert. All SNPs can revert in principle but without a
> > backup copy elsewhere the chances of this happening are very very small. I
> > think John meant that P25 could revert equally in men who are S21+ and
> > S21-, not that S21 would revert as well.
> > I assume S26 is similarly stable as it is associated with a single copy
> > STR. SNPs associated with multicopy STRs will be subject to gene
> > conversion - e.g. the G to C SNP underlying the DYS464X test sometimes
> > appears in two copies or all four copies rather than the usual three out
> > of four.
> > Gareth


This thread: