GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-12 > 1293827001


From: (John Chandler)
Subject: Re: [DNA] The death of paragroups
Date: Fri, 31 Dec 2010 15:23:21 -0500
References: <8FB365E1-2876-4664-AA70-3CCD4CEBB9CB@vizachero.com> <000501cba78a$0850ce70$c2482dae@Ken1> <041023E5-0C60-4B09-B9C7-ED6288FC136F@vizachero.com> <AANLkTimjFkDV1HCtdJE+qQtpdP5uwhtLe7mgUwxQ-E_O@mail.gmail.com><AANLkTi=XWCJGtmgbEUZWqiyWs6vXxArwo-rBNHAAQbsy@mail.gmail.com> <003901cba846$c3a30910$4ae91b30$@dgmweb.net><00f101cba864$13034b10$c2482dae@Ken1><006c01cba86e$1939fa10$4badee30$@dgmweb.net>
In-Reply-To: <006c01cba86e$1939fa10$4badee30$@dgmweb.net> (DianaGM@dgmweb.net)


Diana wrote:
> You are assuming that the man with the ancestral allele was the last of his
> line, which is highly unlikely.

On the contrary, every man is necessarily the last of his line until
and unless he has a son. We're talking about the human male
descendancy tree here, not about some family corporation where the
chairmanship passes to the nearest surviving kin however distantly
related.

> He probably had hundreds of cousins with the
> ancestral allele, some of whom have living descendants

Perhaps so, but those were all necessarily on different lines, with
different SNPs marking the clades they and/or their recent ancestors
had founded. Again, we are talking about the actual, complete human
male tree, not about the severely incomplete successive approximations
constructed from population studies.

A bifurcation occurs exactly when, and only when, a man has two sons
who go on to leave progeny. Each such son is by definition the
founder of a clade, even if the geneticists haven't yet discovered the
characters that distinguish those clades. Similarly, a trifurcation
occurs whenever a man has three such sons. It's as simple as that.
(Of course, identical twins pose a challenge to this simplicity if
they are indeed genetically identical, but we can consider them to be
"just one son" for the purposes of this discussion.)

The key factor, from a cladistic standpoint, is the frequency of
mutations that can serve as readily discoverable characters for the
clades that exist in fact. Here is where the Y tree differs radically
from the mtDNA tree -- each new male has a very good chance of having
his own new Y SNP because the length of the Y chromosome is roughly
equal to the reciprocal of the Y mutation rate, while the mtDNA
molecule is far shorter than the reciprocal-mtDNA-mutation-rate, and,
worse yet, most of the mtDNA molecule codes for things essential to
life. This big difference means that the true branching structure of
the mtDNA tree will never be discovered, and its cladogram will always
therefore be festooned with ridiculously large numbers of seemingly
coeval sister subclades in the best-studied clades, while the Y tree
has a good chance of being reconstructed in its full ramification.

John Chandler

This thread: