GENEALOGY-DNA-L Archives
Archiver > GENEALOGY-DNA > 2011-12 > 1324593985
From:
Subject: [DNA] Generations [was Re: King Tut's DNA- Youngest possible
Date: Thu, 22 Dec 2011 22:46:25 +0000 (UTC)
In-Reply-To: <mailman.2440.1324586173.29924.genealogy-dna@rootsweb.com>
>From: (John Chandler)
>Therefore, trying to calibrate the mutation rate constants in
terms of years is to accept and propagate the huge uncertainty that
comes from that wide range in generation lengths.
My response:
Please define "huge". A number maybe?
Personally, I consider all those "negative comments" as empty words unless they are accompanying with a number of specific cases showing how "huge" it is. Maybe 5%? 10% 15%? 20%? I prefer to work with numbers, not with that "huge" stuff.
If you want to know how "huge" it is, take a look at my recent publication in Adv. Anthropol. (w/ Igor Rozhanskii). We analyze that "huge" thing with more than 3,000 haplotypes (most of them in the 67 marker format) comparing them with documented genealogy data. So, I have a news for you. That "huge" is fully acceptable with the correlation coefficient of 0.95 and gives a good fit with documented genealogy.
Regards,
Anatole Klyosov
***********************************************
Anatole wrote:
> Then, a generation is a "floating" number, which depends on many
> factors - cultures, habits, wars, cataclysms, diseases, etc. They vary
> in a wide range from at least 16 to at least 45, pick any number. A
> "generations" per se is a useless term to employ in historical
> calculations
You have put your finger precisely on the problem. As it happens, the
generation is the one-and-only relevant time unit in genetic
genealogy. The father-son studies have looked for, but failed to
establish, even a clear correlation between mutation likelihood and
father's age, let alone a linear relationship. In other words, the
intrinsic mutation process yields mutations per generation, not per
year. Therefore, trying to calibrate the mutation rate constants in
terms of years is to accept and propagate the huge uncertainty that
comes from that wide range in generation lengths.
John Chandler
This thread: