GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2004-10 > 1097695252


From: "Whit Athey" <>
Subject: RE: [DNA] Differing mutation rates
Date: Wed, 13 Oct 2004 15:20:52 -0400
In-Reply-To: <LAEMICGGMBKIADLLMIJEMELDEBAA.terry@bartonsite.org>


Hi Terry. I am not the best person to ask this question, but we will always
be faced with a probabilistic argument in the end. There is a way to modify
the regular confidence interval to account for many trials (corresponding to
many surname projects in this case). Maybe if you only consider your own
project, no modification would be needed. However, if we looked at 100
projects in the same way, we would likely see about 5 that had rates outside
the 95% confidence interval just on chance alone. Maybe some of our real
statisticians could provide some better guidance on this.

Whit


-----Original Message-----
From: Terry Barton [mailto:]
Sent: Wednesday, October 13, 2004 11:26 AM
To:
Subject: RE: [DNA] Differing mutation rates

Whit: how many results would I have to have before the lower average
mutation rate we are seeing would be statistically valid?

Terry

-----Original Message-----
From: Whit Athey [mailto:]
Sent: Wednesday, October 13, 2004 10:54 AM
To:
Subject: RE: [DNA] Differing mutation rates


I agree that differing mutation rates in different family lines remain a
theoretical possibility. I have come around to that view a little
reluctantly, but it appears that in mtDNA, at least, that a few mutations
can introduce "instability" into neighboring locations, causing greater
mutation rates. This has been observed in mt-Haplogroup T, where a mutation
at 16296, just two locations from one of the defining mutations for T at
16294, can cause instability and back-mutations in the region 16292-16298
(Malyarchuk & Derenko, Ann Hum Genet 63:489-497 [1999]). One of my project
participants has mutations at 16294, 16296, and 16298, for example. I
assume that something similar might happen on the Y-chromosome, even though
the copying and error checking may be a little different. However, the
value that Terry got for his project STR rate seems entirely consistent with
the average values that most people are using. With small numbers of
mutations, there is sure to be a fair amount of uncertainty in the derived
rates. I haven't yet seen any direct evidence that the "theoretical
possibility" discussed above has actually been observed. With so many
surname projects, we are sure to see a few average rates with 95% confidence
intervals that don't include the value of 0.0023, just due to chance, so I'm
not sure how we would ever know if real differing rates were being observed.


Whit




-----Original Message-----
From: Hikaru Kitabayashi [mailto:]
Sent: Wednesday, October 13, 2004 12:08 AM
To:
Subject: [DNA] Differing mutation rates

Just a comment on the possibility of differing
mutation rates. I feel this should not be dismissed as
out of hand. DNA genealogy is still in its infancy and
genetics as a science is hardly out of its swaddling
clothes. Fast strides are being made, but there is
still far more which is not understood than
understood. Whether or not it proves to be true in the
future, at this early stage of DNA genealogy, keeping
the possibility of differing mutation rates alive in
the back of one's head is a reasonable proposition and
could one day be the source of a yet better
explanation of how DNA works.

Hikaru Kitabayashi

> Date: Tue, 12 Oct 2004 20:57:34 -0400
> From: "Terry Barton" <>
> To:
> Subject: RE: [DNA] New YHRD results - AMH & R1b
> Mutation Rates
>
> John, I don't base this idea on the number of
> incidental matches (which are
> considerable), but on the actual low rate of
> mutations we see in our main
> R1b Lineage. We have 65 men in the Barton Lineage I
> representing at least
> 25 different families, with most pedigrees going
> back to the 1600s and
> 1700s. Those 65 men have generated a total of only
> 24 mutations. (the 65
> men range from 25 to 48 markers, with an average of
> 31 markers reported) I
> haven't calculated a mutation rate lately, but last
> time I did, it was about
> 0.0016. Recent results probably lower that rate a
> little.
>
> Ken did address my thought directly and suggested
> that there is no evidence
> of it.
>
> So, with you two both dismissing the thought of R1b
> or AMH having a lower
> mutation rate for different reasons, I'll drop it.
> And, I'll keep
> struggling with two Surname Projects with R1bs with
> low mutation rates and
> hope that an explanation emerges one of these days.
>
> Terry
>
>
> -----Original Message-----
> From: John Chandler
> [mailto:]
> Sent: Tuesday, October 12, 2004 4:27 PM
> To:
> Subject: Re: [DNA] New YHRD results - AMH & R1b
> Mutation Rates
>
>
> Terry wrote:
> > Adam, I'll toss out another possibility on AMH.
> Maybe the R1b haplogroup
> > mutates more slowly than the "average"? Or maybe,
> it's just the AMH 6
> > marker portion of it?
>
> I think you're trying to find an explanation for
> something that isn't
> even true. There is plenty of haplotype diversity
> within R1b. It's
> just that there are so many R1b's around that
> matches are common. If
> you look only at people who are *on* the mode, then,
> by definition, you
> have excluded everybody who is different. That's no
> way to determine
> the degree of diversity.
>
> John Chandler
>
>
> ==============================
> You can manage your RootsWeb-Review subscription
> from
> http://newsletters.rootsweb.com/


==============================
Gain access to over two billion names including the new Immigration
Collection with an Ancestry.com free trial. Click to learn more.
http://www.ancestry.com/rd/redir.asp?targetid=4930&sourceid=1237





==============================
Gain access to over two billion names including the new Immigration
Collection with an Ancestry.com free trial. Click to learn more.
http://www.ancestry.com/rd/redir.asp?targetid=4930&sourceid=1237


==============================
Gain access to over two billion names including the new Immigration
Collection with an Ancestry.com free trial. Click to learn more.
http://www.ancestry.com/rd/redir.asp?targetid=4930&sourceid=1237




This thread: