GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-03 > 1269879553


From: Stephen Forrest <>
Subject: Re: [DNA] Family Finder Test
Date: Mon, 29 Mar 2010 12:24:30 -0400
References: <C03DB1E5-A1C7-440F-87D9-43BD49DFFD15@earthlink.net>
In-Reply-To: <C03DB1E5-A1C7-440F-87D9-43BD49DFFD15@earthlink.net>


A question I've wondered about for some time is the following:

We know that the mean amount of nuclear DNA shared between two full siblings
is 50% of the total. What is the variance?

I imagine this depends considerably on crossover frequencies, etc. and thus
is not a completely straightforward calculation but surely there must be
estimates, especially now that we can compare shared segments! In your post
you allude to a small but nonzero probability of no shared autosomal DNA
between siblings, so presumably the variance is high enough to not
completely exclude that.

regards,

Stephen

On 27 March 2010 18:11, John Carr <> wrote:

> Elizabeth,
>
> Your idea will not work because all of the descendants of a man and a women
> will not share exactly the same autosomal DNA. In fact, two siblings could
> inherit totally different autosomal segments from each parent and appear to
> not match each other at all (low probability, but it could happen). Your
> idea may be used to exclude people from potential shared lineage within 5
> generations based on low probability, but the results will be so varied that
> they will not confirm shared lineage. The results would show a high
> probability of shared lineage if there are matching autosomal segments, but
> you already have that from the YSTR results. The same paper trail required
> for YSTR results would be required here as well. There would be one
> advantage, the autosomal results would indicate the probability of a shared
> ancestor was within a few generations, where the matching YSTR results may
> indicate shared ancestry from within a few generations to many generations
> ago.
>
> This is all still new technology, perhaps in time someone will develop a
> way to distinguish shared lineage using autosomal SNP results, I believe
> paternity tests already claim this ability, but for now, all we see is the
> possibility of a potential shared biological lineage.
>
> Hope this makes it more clear for you,
>
> John Carr
>
> >
> > From: Elizabeth Bennett <>
> > Subject: Re: [DNA] Family Finder Test
> > Date: Sat, 27 Mar 2010 16:36:11 -0200
> > References: <
> >,<>,<
> >
> > In-Reply-To: <>
> >
> > I'm sorry if I was unclear. I do not assume that if autosomal results
> show a relationship that it is in the paternal line. IF there is an
> autosomal relationship and IF they are assumed to be cousins in the paternal
> line, CAN these results be used to group Y DNA results in smaller groups
> within a large one. If you look at the R1a network charts on the Clan Donald
> DNA page there are not many clusters.
> > Clan Donald has at least one chief who was Donald Gorm; Donald the blue
> eyed so I have exchanged information on eye colour and height with some of
> my father's matches.
> > Elizabeth
>
>
>
> -------------------------------
> To unsubscribe from the list, please send an email to
> with the word 'unsubscribe' without the
> quotes in the subject and the body of the message
>


This thread: