GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2002-01 > 1011375506


From: "Dell Ariola" <>
Subject: [DNA] From: Ann Turner re: the Kincaid Project
Date: Fri, 18 Jan 2002 11:41:04 -0600
References: <16f.758563c.2979b3ff@aol.com>



----- Original Message -----
From: <>
To: <>
Sent: Friday, January 18, 2002 11:23 AM
Subject: Re: [DNA] Interesting DNA result for Kincaid project


> In a message dated 01/17/02 5:06:14 PM Pacific Standard Time,
> writes:
>
>
> > is it fair to
> > say that samples 1255 and 1474 represent the oldest
> > line with sample 1494 being one branch of it and
> > the other six matching samples being another branch?
> >
>
> Referring to http://www.alphalink.com.au/~kincaid/dnaresul.htm :
>
> I'd say that's a good working hypothesis.
>
> One technique used by population geneticists is to construct network diagrams
> which show all possible ways of getting from one sample to another with
> single-step mutations. You then see what the network looks like if you strip
> away the long paths and keep the shortest paths (= the smallest number of
> mutations), a technique called "maximum parsimony."
>
> Sometimes the network shows alternative paths, where you can't really
> determine the shortest route. But sometimes the network is clearly
> star-shaped -- one central point with several branches radiating off of it.
> The center of the star is considered to be the ancestral haplotype. For an
> illustration of a star, the Oxford Ancestors mtDNA report has several stars
> http://www.oxfordancestors.com/daughters.html#matriline
>
> You're on your way to a star, with 1255/1474 at the center and two branches
> radiating off it. One thing that makes your situation a little trickier is
> that one of the branches has a two-step mutation, but it's still basically
> the foundation for a star shape. More branches off the center would
> strengthen your hypothesis.
>
> The other technique, or rule of thumb really, is to look at the frequency of
> each haplotype. The most common haplotype is considered to be the ancestral
> one. This is because there are probably other people who share the haplotype,
> and they all continue to have descendants, most of them adding more people to
> the same haplotype. But each mutation starts a new branch with just one
> person, so they're not likely to catch up in numbers with the ancestral type.
>
> If we look at the Y STR database, your 1255/1474 set has 131 matches; your
> 1494 has 7 matches, and your set of 6 has 11 matches. That's consistent with
> the hypothesis that 1255/1474 has the ancestral haplotype.
>
> Random events can play quite a strong role in the way numbers fall out, so
> this is just a rule of thumb. For example, a large number of sons in the
> early generations cause one offshoot to dominate numerically.
>
> I checked the Stidham results, where the common ancestor is known, and the
> numbers do fall out according to theory. The ancestral haplotype had 41
> matches, and two branches with single-step mutations had 8 and 10 matches.
> Would other people like to check this rule of thumb on their project data?
>
> Anyway, it's always nice when two different techniques come up with the same
> answer. More data is nice, too!
>
>
>
>
>
> ==============================
> To join Ancestry.com and access our 1.2 billion online genealogy records, go to:
> http://www.ancestry.com/rd/redir.asp?targetid=571&sourceid=1237
>
>


This thread: