GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2006-12 > 1165953584


From: (John Chandler)
Subject: Re: [DNA] The Basques, Iberia, Paleolithic Continuity,Y-R1b and mt-H and V
Date: Tue, 12 Dec 2006 14:59:44 -0500 (EST)
References: <031401c71e10$32f92fd0$6401a8c0@Precision360>
In-Reply-To: <031401c71e10$32f92fd0$6401a8c0@Precision360> (lgmayka@ieee.org)


Lawrence wrote:
> > [mailto:] On Behalf Of John McEwan
> > TMRCA may not have been when R1b1c mutation emerged it may
> > have been much earlier. Do the math, a constrained population
> > held at low numbers for thousands of years and subject to the
> > usual random fluctuations in offspring numbers and sex
> > distribution will typically coalesce to a single most recent
> > ancestor at a much later date.
>
> This is theoretically possible but is certainly not the most 'parsimonious'
> explanation.

Not so. It is theoretically *inevitable* and therefore *is* the most
parsimonious explanation. The key conditions are the small population
and the lack of expansion. The latter produces a probability of
virtually 100% that any given male line will daughter out or otherwise
go extinct, and the former enforces the condition that "virtual 100%
extinction" means "extinction of all lines but one". This is not the
same as a population bottleneck, but it has the same effect on the STR
variance.

> The number of SNPs that prove to be synonymous with haplogroup membership is
> indicative of the haplogroup's "gestation period"--i.e., between its
> original formation and its first persistent expansion.

Not quite. The *first* persistent expansion is the rise from a single
individual to a group. Without that expansion, the haplogroup would
surely disappear. It's really the second expansion that marks the end
of "gestation" -- the point where the population grows into many groups
which are all too big to vanish. However, it is important to note that
the extinction we're talking about is not a sudden event, and therefore
the MRCA of the resulting population must be someone who lived well
*before* the start of this expansion. In other words, the apparent STR
age of a haplogroup is not only a lower bound (on the time since
formation) but also an upper bound (on the time since expansion).

> > If you plot the ages (and errors) for all the
> > various groups defined you will find distinct "clumping" of
> > the group ages which directly contradicts Lawrence's
> > assertion above.
>
> Are you referring to the fact that different haplogroups have very different
> ages?

I think it's clear that John was referring to the fact that many
different haplogroups have the *same* few apparent ages, to within a
rough approximation. This is a striking phenomenon, but it's not clear
exactly what it means, given the complexity of relating the STR age to
events in the real world.

> We need more data, particularly from Central and Eastern Europe.

You can say that again.

John Chandler


This thread: