GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-02 > 1265460996


From: "Anatole Klyosov" <>
Subject: Re: [DNA] Variance Assessment of R:U106 DYS425Null
Date: Sat, 6 Feb 2010 07:56:36 -0500
References: <mailman.373.1265443240.22367.genealogy-dna@rootsweb.com>


From: "Ken Nordtvedt" <>

>>From: "Anatole Klyosov" <>
>>If mutation counting is very mysterious to you, how can we go
>> further?

>What you call mutation counting is mysterious to me. Suppose you have a
>collection of ten haplotypes of 4 STRs
>They are shown below with counts of each haplotype variation. I can count
>number of STR differences from an assumed intial haplotype
>which shows four copies. But I personally would not consider that a count
>of the number of mutations. Some of those differences are probably the
>consequences of a singular past mutation.

>So is your "mutation count" in this example seven (7) ?

(4) 10 10 10 10
(1) 10 10 9 10
(2) 10 11 10 10
(1) 10 9 11 9
(2) 10 9 10 9


Dear Ken,

I responded to you in the preceding message in this thread, indicating that
you have deliberately deviated from a simple case of 284 of 25 marker
haplotypes. In that case the base haplotype was clearly identifiable, the
whole dataset was easily and quantitatively described, the TMRCA's were
practically identical using the mutations counting and base haplotype
counting, the error margin was not large due to a good statistics of
mutations and a high number of markers (7100 of them).

I have addressed to you a simple question: do you still call "mutation
counting" "mysterious"?

You did not answer, and in order to "prove" your negative point gave me an
example which I called "absurd" - in a sense, that it was completely
different in kind from that I had presented before. In other words, in order
not to agree with an obvious thing and not to confirm a (reasonable)
correctness of my approach, you switched to something else. I would not
call it fair modus operandi.

However, on a second thought I decided to consider your example of 10
haplotypes specifically, since many people here do not know indeed how to
handle those cases. If I, god forbid, was presented with those ten
haplotypes as representative ones for some population (suppose, they were
excavated haplotypes with a little option to get more of them), first, I
would make sure that they belong to the same haplogroup. Second, I would
conclude that they were derived from different common ancestors, and their
TMRCA cannot be correctly calculated using a standard procedure. In other
words, they represent several different lineages, they are assorted
fragments of a haplotype tree. Third, I would suggest that THEIR common
ancestor lived apparently more than 10,000 years before carriers of those
haplotypes (provided that I know their 4 markers and their average mutation
rate per marker is the same as that for the typical "scientific" 6-marker
haplotype (19, 388, 390, 391, 392, 393). I believe that information even on
that semi-quantitative level would be useful.

No, there are not seven mutations in that system, since those 4 copies do
not represent a common ancestor for the whole dataset. That is what I use my
logarithmic method for. 4 base haplotypes out of 10 would give you ln(10/4)
= 0.916, and 10 mutations (not 7, as you incorrectly implied) from the
assumed "common ancestor" in those 10 haplotypes would give you 7/10 =
0.700. There is a significant discrepancy between these two figures, hence
the 4-copy haplotype is not the ancestral one.

With a reasonable approximation I can handle the system using the
permutational method, which does not specify an ancestral haplotype, and
which is described in the reference given earlier.

Anatole Klyosov


This thread: