Archiver > GENEALOGY-DNA > 2010-02 > 1266252769

From: "Lancaster-Boon" <>
Subject: [DNA] Y Tree SNPs can not be counted
Date: Mon, 15 Feb 2010 17:52:49 +0100

Dear Anatole

Please read all the way down and do not judge too quickly because there are
several steps here.

> The issue of the discussion was focused at two items: (1) do counting of
mutations in the contemporary haplotypes give us an adequate tool to
calculate TMRCA?, and (2) are margins of error which I use "too optimistic"?

This is a useful attempt to summarize the subject, which is sometimes
important when things are getting confusing. Going over your 1 and 2:-

1) OK, so my quiet translation to standard terminology is "does genetic
distance from modal give us an adequate tool to calculate TMRCA?" I do not
know if this was ever a separate question from 2. Anyway, a basic and
already known answer is, I suppose, that it often is adequate, but not

I would like to propose a different definition of question 1. The more
interesting question is how we can tell for sure which kind of case it is,
or can we only guess? Let's call this re-wroding of the question "1A".

2) Yes. This is a question. Notice how the answer to 1A changes everything.
If we can only guess whether we have a "first order" case or not, then our
real range of possible errors can be quite big.

>In this case everyone in the dataset is a DIRECT descendant of the common
ancestor. This is what I call an "uncomplicated" dataset. This is what I
call a "first-order" dataset.

At first sight this is indeed what the conversation needs, a definition of
what makes a case "complicated" or not. It is the same as 1A above.
But then when I read this, I do not see your answer.

Why is there no apparent answer? Well, just for starters, is it not true
that for EVERY data set "everyone in the dataset is a DIRECT descendant of
the common ancestor"? Are there others types of cases? We are looking at Y
DNA, and by definition everyone descends "directly" from their common
ancestors with anyone else.

Possibly this is another definition problem. Are you defining "common
ancestor" in some special way like Sasson suggested?
523 . See point 3.)
Or possibly is there some special meaning implied by "directly"?


> If the dual method gives you the same TMRCA for both linear (mutation
counting) and logarithmic (base haplotype counting) methods, you are all
set. It is a first-order tree.

I have noticed that you say things like this quite consistently. Maybe this
is a good lead to your code. Normally when you say it, I read you as
mentioning this as a property of complex and simple cases worth remarking
but not as the definition of what distinguishes them per se.
However, maybe this is not your intention.

So, do you intend the above as a kind of formal and not fuzzy definition for
the distinction you make between simple and complex cases? (Are they the
cases where linear and logarithmic methods disagree?)


To David Ewing you also write:
> Indeed, some minor branches allow a multiple choice. Mutations are
mutations, and they follow dumb statistics. Some haplotypes venture into a
neighboring branch, or sit on a borderline. To sort them out you need actual
family archives.

This paragraph sounds so much like what other people are telling you, and
what you SEEM to disagree with, that it seems like a good opportunity for a
cross check. Can you confirm therefore whether the "some cases" that you
mention to David are the "complicated" cases you distinguish in your posts
to me, and that these can be defined and identified as the cases where the
linear and logarithmic methods disagree?

Your answer could possibly help increase understanding.

Best Regards

This thread: