Re: Cost of sex - a bargain!

allanm wrote:

John Harshman wrote:

allanm wrote:


John Harshman wrote:


allanm wrote:



Windy wrote:



allanm wrote:



My contention here is that meiotic defection, all other things being
equal, is selectively neutral. The two-for-one cost cannot be realised
because, as far as I can see, the defector cannot arise in such a way
that males are disfavoured with respect to the defector locus.

There are two 'alleles' - homologues, really, because there is
reproductive isolation: - Be-Asexual and Don't-Be-Asexual.
Don't-Be-Asexual is copied into every male and female descendant of a
sexual population. If not, where does it reside, in the genome?

Be-Asexual will get copied into each and every descendant.
Don't-Be-Asexual can only be sure of getting into 50% of the
descendants; in others, it may be supplanted by Don't-Be-Asexual-2,
Don't-Be-Asexual-3, or any other allele; even Be-Asexual, in case the
asexuality only manifests in some conditions.

-- w.


You really are missing it; I'm sorry to sound contentious. It doesn't
matter what gene Be-Asexual arose from.

Don't-Be-Asexual is not an 'allele' of Be-Asexual at all.
Don't-Be-Asexual is the ancestral state - all alleles, anywhere, which
do NOT promote asexual reproduction. in a functional sexual female.
"It" - a disparate set of genes - is copied to EVERY offspring of a
sexually reproducing female. There is no 50% gain for a ***defecting***
gene.

No. What there is is a 50% gain for all alleles present in any genome
that has the Be-Asexual allele. Therefore all those alleles will
increase in frequency. You could if you like consider them all as
hitchhikers except for Be-Asexual.


I do consider them as hitchhikers. More importantly, they have all
hitched the same ride. The gaining of copies by selective processes in
a sexual genome is not the same thing as the incidental gaining of
nonselected copies, whatever the ratio.

Same result, though: increase in allele frequencies in the population.

A completely irrelevant result, I have to say. If asexuals were
more-or-less identical, then no selective death for them could be
attributed to any individual allele's effect.

The selection in this case is in the population as a whole, including
both sexual and asexual individuals. The alleles in asexual individuals
all gain copies relative to the alleles in sexual individuals.

If a predator 'notices'
wild-types on the basis first of colour, then prefers larger ones to
smaller ones, there are separate selective pressures on those
characters. If a particular value of each separate variable is
incidentally frozen into the subpopulation, and a million green 1mm
bugs result, the 'increase in allele frequency' for any allele is
largely a function of the trivial fact that every member of the
population must have an allele at that locus, and there's only one
available.

I have no idea what this means either. You seem to be conflating
selection from different sources.

The wild-type, meanwhile, is in a position to make itself
inconspicuous, thanks to the variation available to it. It is not all
about copies.

Yes it is. In fact that's all it's about. You are once more conflating
different selective forces. Asexual reproduction has an initial 2:1
advantage. You seem to be claiming that because there are suggested
countervailing advantages to sex, the 2:1 advantage therefore does not
exist. Instead the standard (correct) view is that the advantage must be
overcome by some combination of the suggested advantages of sex.

Selection is blind to the distinction between sexuals and asexuals. If
there is any differential at all in any selective parameter, the
variant portion of the population (the sexuals) will undergo selective
changes in allele frequency.

No idea what this combination of words is supposed to convey.

Particularly when the mode of
assortment, and units of gene selection, from that moment on, are
completely changed from the sexual type. The whole genome of an asexual
is selected as a piece. An allele in a sexual is 'tested' by selection
as it works in conjunction and competition with many other different
alleles, separately and together, at its own locus and at others.

There is a case for saying that this is a speciation event, even if
Be-Asexual is a recessive member of the sexual genome. It can only
increase by drift, and is not phenotypically expressed in the sexual.
Every time it meets itself, it pops it and every allele in that genome
into a very different mode of operation. The sexual genome is
completely unaffected by this event - unless, in the fullness of time,
it has to deal with these individuals as competitors for food, space
etc. It is likely to be well equipped to do so, as varied sexual
genomes tend to be.

Note that it has to be well enough equipped to overcome the 2:1 advantage.

Which 2:1 advantage are we talking about here - there are several
versions of this argument.

No, there's only one version I know of.

1) The benefit to any individual is irrelevant.

No idea what that means.

2) The cost argument does not apply, if the offspring may be assumed to
cost the same.

Nor that.

3) The cost of finding a mate may be ignored, if males are obliging.

The 2:1 advantage doesn't take costs of mate finding into account.

4) The unselected allele frequency benefit is easily overcome.
Wild-types just generate extra copies of a better one!

No idea what that means either.

Which leaves the population-increase argument - well, that is indeed
what we are saying it has to be well enough equipped to overcome. But I
really cannot see how a fixed genome is going to prevail over a
population which can select and mix adaptive variation.

Surely that depends on the size of the advantage resulting from mixing.
People generally find it unlikely that allelic variation in the
population can produce such large selective differences as 50%, at least
over the short run, and that asexuals would become fixed before any of
the evolutionary superiority of sex could assert itself. Now, you can
rescue that explanation by resorting to species selection, i.e. the
populations in which cheaters never arise have longer-term survival
potential and thus are selected *as populations*. Is that what you
wanted to say?

The grey squirrel is squeezing out the red in the UK. They are both
adaptable sexual organisms. An 'arms race' may prolong the struggle,
but the red is losing. But two points: the grey comes from outside the
red's gene pool, and can match it move for move. This is vastly
different to Xeroxing a single exemplar from within a gene pool, when
the maternal pool has both the variation and the recombinant means to
deal with it.

So it's irrelevant to any point you may have. And even clonal
populations have genetic variation and can respond to selection. This is
another confusion between the supposed countervailing advantages of sex
and the absence of any 2:1 advantage in fecundity.

This isn't a radical position, surely?

Merely a very, very confused one.

This is not a killer mechanism for females to subvert males.

Who said it was? What does that even mean?

My case is that, apparently contrary to most views, sex does not need
an undiscovered advantage in order to explain why females do not defect
and gain the short-term benefits. It is supposed a mystery why sex
persists, but I don't see a mystery, if females don't have a clear and
rapid short-term mechanism ("a killer mechanism") to reap gains by
defecting. I'm trying to see the candidate which everyone can see but
me, and failing.

Obviously. And I don't understand why a 2:1 immediate advantage in
reproductive success is not a "killer mechlanism". Every single allele
in the female's genome gets twice as many copies if it reproduces
asexually as if it reproduces sexually. Right?

Males have daughters too; has everyone forgotten that? Where do their
genes come from?


Half from the father, half from the mother. So the mother's genes could
be reproduced twice as much by dispensing with the father. What does the
fact that males have daughters contribute to this?

Sexual females produce gametes. Not male or female gametes. Males
produce gametes which, for the vast majority of their DNA, are
essentially the same as female. The machinery for female sexual
reproduction is represented fully in all those gametes.

Any *individual* maternal allele can be doubled by going asexual, yes
(but see above - NOT the one which defects).

I don't understand the parenthetical remark.

Two forms of the two-for-one cost at allele level are being chucked at
me.

The incidental one - every allele but the defector allele - genuinely
does seem trivial to me. If that makes me look stupid then I can't
think differently just to fit in with the crowd.

The defector allele itself does not gain because it does not get into
the asexual population.

You need to learn to write less elliptically. I still have no idea what
you're talking about. You just raise new questions. Of course the
"defector allele" (terrible terminology, by the way) gets double the
success of an allele at the same locus in a sexual population. I suppose
you are assuming homozygosity of a single alternative allele in your
counter-case? Silly and loaded assumption, if so.

But she does not halve her
output of "sexually reproducing female" genes by liaising with, or
producing, males. In fact, inasmuch as males exploit females, male
offspring are an effective means of females getting in on that act also.

Sorry, but any sense that may have made did not come through. Again,
it's simple math. Every allele in the asexual female will produce twice
the copies as that same allele in an otherwise identical sexual female.

I'm just going to have to say 'trivial' again, I'm afraid. In a real
sense this is a speciation event, with reproductive isolation. We don't
keep comparing homologous allele frequencies between well-diverged
lines in separate species, even though base sequences, phenotypic
effect and selective advantage may be almost identical, and the
characters may have an effect on competition and differential selection
between integrated populations.

The problem is that for most purposes these two "new" populations act as
the same species. They compete for exactly the same resources. Unless
they diverge in resource use (competitive character displacement), one
of them will cause the extinction of the other. If you prefer to call
this interspecies competition rather than natural selection, feel free.
But it's the same process in either case, with the same end result.

As for the defector allele, what I am arguing is that there are simply
two phenotypes, Sexual and Asexual, expressed in females but carried
equally by the sexes. An indistinguishable variant of the Sexual
'polygene' is passed to 100% of the offspring of every male and every
female in the sexual population. When any mutation causes a defection
to the asexual genotype, that genotype is passed to 100% of offspring.
The actual source of the Asexual change could be a consequence of any
mutation in any gene, whether involved in sexual reproduction or not.
The original gene is irrelevant (it is not passed on, even though an
Asexual-promoting derivative is). 'Sexual' cannot be subverted because
it is not located in any one place, passed on to females only.

Once again, the words you string together are making little sense to me,
so it's hard to respond. If you have a point, you need to explain it
more carefully.

.