Archiver > GENEALOGY-DNA > 2011-08 > 1314691830

From: "Sandy Paterson" <>
Subject: Re: [DNA] exact match at 67
Date: Tue, 30 Aug 2011 09:10:30 +0100
References: <005801cc574b$96bb1af0$c43150d0$> <> <950D783B573A4FE78DD7CD26F8C8DB53@jimpc> <> <> <009e01cc5853$e3c54d30$ab4fe790$> <> <00f201cc587a$cf36c7f0$6da457d0$> <> <001201cc5b2c$fd9634f0$f8c29ed0$> <> <006901cc5bdf$fb374600$f1a5d200$> <> <014001cc63b4$31e9c9d0$95bd5d70$><>
In-Reply-To: <>

I couldn't agree less.

You seem to have ignored (or missed) the fact that the two haplotypes not
only match 67/67, but have 7 off-modal matches.

That changes the probabilities completely, to such an extent that
Fr{TMRCA<=5} is approximately .855

So my opinion is completely opposite to yours, and I would suggest (in
agreement with DJM and Dave Johnston) that the 1750 ancestor is highly
unlikely to be their most recent common ancestor. A common ancestor,
obviously, but the most recent common ancestor is far more likely to be a
descendant of the 1750 ancestor.


-----Original Message-----
[mailto:] On Behalf Of John Chandler
Sent: 30 August 2011 04:30
Subject: Re: [DNA] exact match at 67

Diana wrote:

> No it doesn't, not when the sample size is too small. And my advice
> was good. But for the sake of our fellow subscribers, let's just
> agree to disagree.

That's a favorite ploy of yours, and, indeed, proposing to "agree to
disagree" is a standard way of ending intractable arguments amicably,
but you haven't quite mastered the technique. For one thing, it
implies a promise to say no more on the topic thereafter, but your
track record for maintaining silence is less than impressive.
Similarly, the appropriate time to make this offer is IN LIEU OF yet
another statement of one's position, and the above example is a
classic case of making the offer in bad faith. Boil it down, and it
says: "You're wrong. I'm right. Let's have no more discussion."

In point of fact, though, your so-called "good" advice was the worst
possible. Recall the question that opened this thread: a 67/67 match
exists between two men, one tracing back to an ancestor in the 1750s
and the other with a more limited pedigree -- should they operate on
the assumption that the earliest known ancestor of the one is also an
ancestor of the other? Your advice reached the following crescendo:

> Yes, of course you can calculate probabilities from mutation rates,
> which are based on large sample sizes. But in the case at hand (a
> 67/67 match between two persons), those probabilities are of virtually
> no *practical* use to the genealogist or project admin because all
> they tell you is that these two individuals almost certainly have a
> common ancestor in genealogical time.

In a word, your answer is an emphatic "NO". However, the correct
answer is "YES". As I pointed out in response, the odds are extremely
good that the proposed assumption is true.

As for your other claim, I must repeat that correlation is a
statistical property. If you have enough cases that it's actually
meaningful to utter the word "correlate" in the first place, your
sample is large enough. I'll freely stipulate that a sample size of
three closely related men tested on 37 markers is too small to get
meaningful statistics. However, ten men should be plenty to reveal
the correlation between genetic distance and relatedness. Note that
testing more (or "faster") markers will lessen the number of men
needed to get meaningful statistics.

John Chandler

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: