GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-09 > 1125582210


From: "Peter A. Kincaid" <>
Subject: Re: [DNA] "Irish" R1b variety
Date: Thu, 1 Sep 2005 10:43:30 -0300
References: <000c01c5ae77$fe6a5590$71509045@Ken1> <003e01c5ae9a$e01f6e40$a2129a8e@PeterAKincaid> <00c801c5aea1$7c36eff0$71509045@Ken1>


You are the one who claims to found a unique subclade. Any
proper scientific analysis includes an explanation on
methodology so that one can repeat the experiment in a static
environment. In this case the popluation is changing so it is
only proper to present the data as others will not be able to
extract exactly the same data.

In this particular case you are presenting a pattern based on
5 markers. Yet Sorensen requires you to enter 7. Assuming
a minimum of 5 allelles for each marker it is clear that you
have not presented a frequency distribution for 70,000 plus
combinations. At the very least you fixed 2 of the markers
and worked out frequencies of the other five based on one
or two allelles maximum per marker. What possible
scientific conclusion can be drawn from when modifying
your fixed markers can dramatically alter your results.

Thus, in effect you have concluded that you have located
an Irish variety or Scots variety when these have very
little to do with the populations of these countries. All they
deal with is the geographic distribution of apparently one or two
haplotypes whose origin may or may not have anything to do
with these countries. Instead what one is really saying is
that assumming that x haplotype originated in y1 country today
it makes up n2% of the population in y2 country, n3% in y3
country etc.

Thus, my argument all along in that the labels chosen are
completely misleading. If you want some credibility to
your statements then at the very least the onus is on you
to present your data used. I believe everyone appreciates
you taking time out to do some heavy analysis for the
benefit of others. However, it appears to be all for
nothing but adding to confusion.

Well I have beaten this dog to death. It is clear to me
why you continue to persist in your presentation.

Peter

P.S. I am indeed not helpless. I am using my time to
first identify a pool of surnames that originated in the
Strathclyde area of Scotland at least 600 years ago.
I am slowly building up a significant database of these
names from Ysearch so as to get a picture of the makeup
of the area. Once done the data will be presented for
all to see and pick apart as is only proper.

----- Original Message -----
From: "Ken Nordtvedt" <>
To: <>
Sent: Thursday, September 01, 2005 12:01 AM
Subject: Re: [DNA] "Irish" R1b variety


> You are not helpless. Go to Sorenson yourself and look at them. I said
> the
> others were Americans (or Canadians; that means folks who don't know
> their
> ancestry across the Atlantic, though that's where they are from.)
>
>
> ----- Original Message -----
> From: "Peter A. Kincaid" <>
> To: <>
> Sent: Wednesday, August 31, 2005 8:14 PM
> Subject: Re: [DNA] "Irish" R1b variety
>
>
>>
>> ----- Original Message -----
>> From: "Ken Nordtvedt" <>
>> To: <>
>> Sent: Wednesday, August 31, 2005 7:04 PM
>> Subject: [DNA] "Irish" R1b variety
>>
>>
>> > Sorenson database has 127 haplotypes of the 25/11/14/11/13 variety at
>> > DYS390,391,392,385a,b. Origins as stated by owners of these haplotypes
>> > from whom Sorenson demands detailed pedigrees are:
>> >
>> > Ireland 20, England 11, Scotland 11, Denmark 1, France 1, Sweden 1,
>> > Germany 1, and the rest American (and I resisted ascribing any of them
> to
>> > an old world country of origin in spite of many give away surnames).
>> > Hence an "Irish" variety of R1b seems an apt nickname.
>> >
>>
>> Here you admit to some selective bias on your part. You have
>> 81 (or 67%) haplotypes unclassified but chose to label this
>> an Irish variety. Will you post all 127 haplotypes (with surnames)
>> so we can see how selective you were?
>>
>> Peter
>

This thread: