DNA-R1B1C7-L Archives

Archiver > DNA-R1B1C7 > 2009-12 > 1260355469


From: "Sandy Paterson" <>
Subject: Re: [R-M222] Coyne and Golden project families
Date: Wed, 9 Dec 2009 10:44:29 -0000
References: <mailman.9945.1260277289.2036.dna-r1b1c7@rootsweb.com><001c01ca789d$a43229f0$aa00a8c0@Sisko>
In-Reply-To: <001c01ca789d$a43229f0$aa00a8c0@Sisko>


Aha!

@
The bottom line: 1) Within the limitations of probability, the likelyhood of
relatedness between two
individuals within a haplogroup is characterized by the genetic distance
between the haplotypes of the
two individuals, regardless of how close they are, or are not, to some
calculated modal value.
@


OK, let's examine the evidence.

Let us pretend that we know beyond doubt that the M222+ modal is exactly as
estimated, ie that the entire population of M222+ is descended from one man
with the haplotype

13,25,14,11,11,....

Let us assume further that the age of M222+ is indeed 60 generations (1800
years).

On the one hand, we compare Mackenzie 26917 (who matches the M222+ modal
67/67) to Coyne N49082 (who matches M222 66/67). They are a gd of 1 from
each other.
On the other hand, we compare Paterson 118913 (yours truly, 60/67 with M222)
and Lamont 94014 (59/67 with M222). They too are a gd of 1 from each other.

According to your statement above, in the absence of any further
information, the likelihood of Mackenzie being related to Coyne is exactly
the same as the likelihood of Paterson being related to Lamont.

Well, let's compare Mackenzie with Coyne. Coyne has DYS576 = 17 compared
with the modal value of 18. That's the only difference. On the face of it
therefore, it could be argued (as you seem to do) that they are closely
related. However, there is no evidence whatsoever of any common mutations
from the M222+ modal. That is a big problem because that's the only way to
logically conclude probable kinship.

Now compare Paterson with Lamont. Paterson has (compared to Lamont and the
M222+ modal):

PatersonLamontM222

DYS439131312
DYS449313130
DYS570181817
CDYa393938
DYS481262625
DYS446151513
DYS487141413

So Paterson and Lamont match on all 7 of the markers where Paterson is
off-modal. The evidence of common mutations is both obvious and compelling.
Lamont has one additional mutation, but that doesn't detract from the
strength of the reasoning.

I think your use of the phrase "..some calculated modal value" is the clue
to where (I believe) you are going wrong. It suggests that in your opinion,
a modal is some meaningless set of numbers that have no relevance to
probability of kinship. Whilst I agree that it is impossible to be sure of
the haplotypes of the founder male (and simulations of 67 marker haplotypes
using Monte Carlo seldom deliver all 67 of the starting haplotypes as the
end modal), I'd suggest that modal value are indeed the best estimate in
statistical terms of the founder haplotype.

So. There is no evidence whatsoever of Mackenzie and Coyne having shared any
mutations, whereas there is clear and compelling evidence that Paterson and
Lamont have indeed shared mutations. It is remotely possible, I suppose,
that all seven of the shared mutations came about independently (so called
parallel mutations) but as you have pointed out, it's a question of
probability.

There is no doubt in my mind as to where the balance of probability lies.

I look forward to your reply.


Sandy



























-----Original Message-----
From:
[mailto:] On Behalf Of Gregory Brown
Sent: 09 December 2009 07:03
To:
Subject: Re: [R-M222] Coyne and Golden project families

I've been sitting on the sidelines for a while waiting for a definitive
resolution of this discussion.
Sandy, I have to respectfully disagree with one of your comments suggesting
that a 67-67 match that is
identical to the "modal haplotype" does not suggest an equally close
relationship between two individuals
as a 67-67 match that is off-modal:

> It would be incorrect to say that your father is not related
> to Mackenzie
> and Sinclair. All M222+ are related. But there is no evidence
> that your
> father is more closely related to Mackenzie than is evidenced
> by the fact
> that they are both M222+. I haven't seen Sinclair's haplotypes. If he
> matches your father on one or more markers that are not on
> the M222+ modal,
> that would be some evidence of a closer kinship. Evidence,
> mind. Not proof.


While off-modal matches are quite significant when comparing within a group
who share membership in a
given "modal haplotype", I don't believe they are any more (or less!)
significant than matches that are
close to the modal haplotype. The modal haplotype is merely an estimation
of a possible "ancestral"
haplotype...an estimation that is based upon those whose individual
haplotypes are a fairly close match
and who share the same "haplogroup", in the present case M-222+. The
haplogroup is defined by SNP
testing for the presence or absence of specific "Unique Event
Polymorphisms"...mutations which presumably
occurred only once in history. That is how they are able to trace ancestral
origins and ancient
migrations.

Getting back to modal haplotypes: There is absolutley nothing particularly
special about a modal
haplotype. It can be estimated for any group or sub-group of related
people. It is simply a mathmatical
estimation...you could even call it an average (simplistically)...but the
alleles that comprise the modal
have no more "staying power" or immunity from mutation than any other
haplotype. Thus, if two people
have a 67-67 match that is identical to the modal haplotype, it has the same
significance of
"relatedness" as two people who have a 67-67 match to each other but are
2-3-4 or more away from the
modal haplotype! Yes, the *likelyhood* that the latter two are more closely
related to each other than
they are to someone who matches the modal precisely is true...but no less
true that the fact that those
who match the modal 67-67 are *likely* to be more closely related to each
other than they are to the two
off-modal 67-67 matches. The significance of genetic distance does not care
about modal
haplotypes...modal haplotypes are simply constructs.

All of this, of course, must be taken with a grain of the salt of
probablilities.

The bottom line: 1) Within the limitations of probability, the likelyhood of
relatedness between two
individuals within a haplogroup is characterized by the genetic distance
between the haplotypes of the
two individuals, regardless of how close they are, or are not, to some
calculated modal value. 2)
Further, the accuracy of the estimation of relatedness naturally increases
with the number of markers
tested. A 65-67 match is much more significant than a 12-12 or 25-25 match.
Any other interpretations
defy intuition, statistics and my understanding of genetics (which may,
granted, be somewhat limited and
skewed since I am just a paleontologist!)

Also looking for the aspirin, and respectfully,

Greg



R1b1c7 Research and Links:

http://clanmaclochlainn.com/R1b1c7/
-------------------------------
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: