Re: Hamilton's rule
- From: "Jim McGinn" <jimmcginn@xxxxxxxxx>
- Date: Fri, 21 Oct 2005 13:47:27 -0400 (EDT)
Tim Tyler wrote:
> Joe Felsenstein <joe@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote or quoted:
> > Tim Tyler <tim@xxxxxxxxxxx> wrote:
> > >Joe Felsenstein <joe@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote or quoted:
>
> > >> I'm astounded that everyone here has allowed McGinn to divert
> > >> the discussion to the useless topic of "what is a gene".
> > >> It really has nothing whatsoever to do with whether Hamilton's
> > >> Rule works.
> > >
> > >It matters for the 1964 version of Hamilton's rule - which uses
> > >identity by descent. *If* a base pairs are genes, then 98% of
> > >our "genes" are identical by descent with those of other humans,
> > >we are all closely related to one another - and that would impact
> > >Hamilton's 1964 expression of his rule.
> >
> > You misunderstand IBD.
>
> I don't think so - though the concept has a few subtleties.
>
> "Identity By Descent" refers to genes which are identical with
> other genes in relatives as a result of descent from one another
> or common descent from a shared ancestor.
Yes, obviously, Joe is the one that doesn't understand IBD. He's the
one that made the following statement which we know to be false: "The
reason it (IBD) is used here is that it allows us to compute the
conditional probability that an allele present in you is also present
in your relative." What's really silly is that further along in the
same paragraph he recognizes that this is an erroneous statement and
then goes on to make silly excuse: "There may be other copies of G
present in the recipient as well, there purely at random, but they
don't count." Even Catherine failed to pick up on it. I asked Joe to
explain to us why these genes don't count. Undoubtedly the evasive
twit will never respond.
>
> > The reason it is used here is that it
> > allows us to compute the conditional probability that an allele
> > present in you is also present in your relative. If 1-p of
> > everyone in the population has an A at a certain position, p
> > have a G, and you (the prospective altruist) have a single G, and we
> > consider a potential recipient who happens to have half a chance
> > of being your sibling, and half a chance of being a random other
> > individual from the population, then the number of extra copies of G
> > present in the recipient owing to its possible relationship is
> > 1/4, no matter what the gene frequency p is.
This last statement by Joe is plainly, demonstrably, false.
> > There may be other
> > copies of G present in the recipient as well, there purely at random,
> > but they don't count. That is all that is needed. And no one needs
> > to look further up or down the chromosome, or consider "what is a gene".
>
> If that was an attempt to explain how I "misunderstand IBD", then
> it has not been very effective - since it failed to even mention
> the concept.
See if you can get him to explain this statement: "There may be other
copies of G present in the recipient as well, there purely at random,
but they don't count." Why don't they count? Because they are at
random? How convenient.
>
> Hamilton explained that the notion that r = genes IBD only made good
> sense in an outbred population.
>
> I was discussing a population where many genes are shared. It is
> no suprise that genes IBD is now a very bad measure of relatedness -
> and is inappropriate for use in Hamilton's rule - that's one of
> the reason's why the notion of defining relatedness in terms of
> genes identical by descent was subsequently revisited.
Subsequently revisited?
>
> > >Of course, treating base pairs as genes in this context is ridiculous -
> > >and the 1964 version of Hamilton's rule now looks a bit dated - but
> > >I don't think you can claim that the "what is a gene" issue is
> > >*irrelevant* to discussion of Hamilton's rule.
> >
> > I can claim that, do claim it, and have claimed it.
>
> OK - you can claim it - but the 1964 "r = genes IBD" statement
> in Hamilton's paper makes it incorrect.
And you're suggesting regression supposedly salvaged it? More smoke
and mirrors.
Jim
.
- Follow-Ups:
- Re: Hamilton's rule
- From: Tim Tyler
- Re: Hamilton's rule
- References:
- Hamilton's rule
- From: Catherine Woodgold
- Hamilton's rule
- Prev by Date: Re: Hamilton's rule
- Next by Date: Re: Hamilton's rule
- Previous by thread: Re: Hamilton's rule
- Next by thread: Re: Hamilton's rule
- Index(es):
Relevant Pages
|
