GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-07 > 1184357097


From: (John Chandler)
Subject: Re: [DNA] Help please with marker 458
Date: Fri, 13 Jul 2007 16:04:57 -0400 (EDT)
References: <008701c7b415$794b0120$2401a8c0@your447023ae6b><REME20070705181857@alum.mit.edu><a81622ac0707051724y16754956obfb853b6686e74ed@mail.gmail.com><REME20070707152459@alum.mit.edu><a81622ac0707071304x11373a0en3b0b4e52a7b2e3c@mail.gmail.com><REME20070707183721@alum.mit.edu><a81622ac0707071637o34a3421dr36660529da9f75a4@mail.gmail.com><REME20070709202718@alum.mit.edu><a81622ac0707092003u290a5dd4s96c4a87a62fe3230@mail.gmail.com><REME20070711175004@alum.mit.edu><a81622ac0707120434g62d3796bk63d9ce33299bb67c@mail.gmail.com>
In-Reply-To: <a81622ac0707120434g62d3796bk63d9ce33299bb67c@mail.gmail.com>(jmknapp@gmail.com)


Joe wrote:
> What are the frequencies of interest? There's power at the lower
> frequencies mostly due to the initial spike in birth rate as the
> population builds up to the steady state, plus of course the "DC"
> term.

The relevant frequencies are 0.02 to 0.05 cycle per year.

I'm not sure why there would be an initial spike in birth rate.
Didn't you say that the population starts out at 10,000 (the same
as the quota)?

> Well then, how much does the exponential constant matter? A flat
> population has a constant of 0. Would a tiny growth rate of 0.001/gen
> make any difference in the result? I.e., at what point does is the
> growth rate large enough to show the effect you want?

A growth rate of 2.8% per generation would expand from 1 pair to
20,000 people in about 10,000 years. Because the population is
small at the beginning, the total number of simulated persons
is much smaller in this type of scenario. As long as the growth
rate is much larger than the mutation rate, the simulation would
steer clear of the draconian weeding-out that characterizes a
static population.

> 10-tribe case:
>
> male#/descendants
> 36 486
> 45 526
> 78 867
> 125 142
> 324 949
> 338 492
> 369 489
> 408 6
> 450 498
> 479 523

There's something I don't understand here. If each tribe has an equal
quota of 1,000, and there is no mixing between tribes, why don't you
have very close to 1,000 living descendants for each of the ten men who
end up saturating their respective tribes? Or, if the above tally is
only the male descendants (adding up to nearly 5,000), how can there be
one tribe with only 6 men and 994 women? Or, if the tribes did not
have quotas of their own, how many of the tribes vanished completely
in the course of this simulation?

> 1-tribe case:
>
> male#/descendants
> 3 343
> 120 1343
> 125 172
> 131 565
> 141 412
> 146 15
> 150 501
> 187 667
> 315 248
> 469 637

Don't you find it remarkable that the number of surviving male lineages
is exactly the same in this case?

John Chandler

This thread: