GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-03 > 1111760144

From: "Sharon Bryant" <>
Subject: Re: [DNA] How significant is a 35/36 match with a person of another surname?
Date: Fri, 25 Mar 2005 09:15:44 -0500
References: <195.3ba14424.2f750e1d@aol.com>

I'm sorry; stupid question, I know -- but what does ASD stand for?

Sharon
----- Original Message -----
From: <>
To: <>
Sent: Friday, March 25, 2005 1:47 AM
Subject: Re: [DNA] How significant is a 35/36 match with a person of another
surname?

>
>
> Jim ...
>
> Ken has posted several msgs with the formula for ASD:
>
> ASD for a specific marker = SUM over all pairs ( n(r)*n(r1)* (r)^2 *
> (r1)^2) * ....)/ (N*(N-1)) where "n" is the count of individuals with
> the
> repeat count "r" for the DYS marker, n1 is the count of individuals for
> the repeat
> count r1, etc. N = n + n1 +..., i.e. the total number of individuals
> in
> the sample for which you're calculating the ASD.
>
> Note ASD can be calculated with all pairs for different repeat counts for
> the marker being examined: no need to assume an ancestral haplotype. (tho
> you
> can :-)
>
> You can peruse the archives for Ken's suggested usage of ASD for TMRCA for
> haplogroups, starting with "I". The initial idea was to look at ratios of
> ASD
> for subclades (you don't need to actually calculate a mutation rate, tho
> you
> can). Also described in a paper by Di Giacomo in 2004 estimating
> subgroups
> of "J", you can estimate the time of sub-haplogroups as a fraction of the
> total lifetime of the parental haplogroup.
>
> ASD theoretically grows as 2*m*g (2 times mutation rate times generation
> count), so the mutation rate estimate between 2 generations is (ASD2 -
> ASD1)/(2*g).
>
> And if you have a rate you believe, TMRCA can be estimated as
> (ASD2-ASD1)/(2*m).
>
> There are several msgs from Ken in the archives describing all this.
> ------------------------------------
>
> By parsimonious I mean the simplest method of estimating an ancestral
> haplotype, yielding the fewest number of mutations, collectively, in your
> sample of
> haplotypes. Charles has a good description of this in his website. It's
> important to have an accurate family tree diagram.
>
> Mike ...
> ========================================================
>
>
> In a message dated 3/24/2005 5:58:33 PM Pacific Standard Time,
> writes:
>
> Mike
>
> It appears that I may have to study possible matches to a paper trail
> that
> goes back to the Norman invasion. Can you give me an idea of how to bone
> up
> on such matters as use of ASD to estimate TMRCA? And how do you define
> parsimonious?
>
> Joe Fox
>
>
>
>
> =======================================================
> http://humphreygenealogy.com
> multi-family project:Humphrey, Liles, Morton
>
>
> ==============================