Kin Selection contradiction?

in article [email hidden], Brian Berns at
[email hidden] wrote on 6/2/04 1:39 PM:

Quoted message said:

I have read many explanations of kin selection that are
based on the "fact" that I share 1/2 of my genes with each
of my parents and siblings, 1/8 of my genes with each of
my first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve this problem for me? Thanks.

The first set of numbers refer to gene copies that are
"identical by decent" (IBD). For example, 1/2 of your genes
are identical with the copies in your sibling because you
inherited copies of the same molecules from your parents.
Most of your other genes are also identical with your
siblings just because there is not an infinite amount of
genetic variation in the population. In other words, these
other genes are identical for reasons other than common
decent from your parents. While it is true that your DNA is
probably >99% the same as any randomly selected human, W. D.
Hamilton focused specifically on those copies that are
identical by decent because he was interested in how
selection might cause a new mutation for altruistic behavior
to increase in frequency when close relatives have frequent
interactions. The probability that a pair of individuals
shares a rare mutation is much better approximated by the
IBD calculation than the overall level of genetic
similarity, which would be more appropriate for moderately
common polymorphisms.

Guy

Brian Berns said:

I have read many explanations of kin selection that are
based on the "fact" that I share 1/2 of my genes with each
of my parents and siblings, 1/8 of my genes with each of
my first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve this problem for me? Thanks.

-- Brian

Others will no doubt do better, but here is my take.

"Gene" in evolution means more than just DNA. It means
*differences* in DNA. Humans might have, for example, ~99%
of all their *DNA* in common, but of the remainder, we each
have 50% of the differences from each parent.

"Relatedness" means sharing in those genes that *vary*
(called alleles, from the Greek for "other"😉 in a
population. If there are N alleles in a population for the
same gene, then your parents may each pass on one or two of
them only. This means you can be grouped in the population
as being either AA or Aa for that allele-pair, even if there
are also a range of other alleles.

Your average share in the differences in genes will be 50%
or near it from each parent. Since genes (with exceptions)
split into halves at

genes according to the Mendelian ratio:

1/4 AA: 1/2 Aa: 1/4 aa

for each gene/allele.

If there are around 35,000 expressed (protein-making) genes,
a 1% variation (or polytypy, as it is called) covers 3,500
genes. This difference can be considerable. If it went to
5%, of course we are dealing with around 17,500 genes. To
make matters more complicated, genes control the expression
of other genes. So a small number of alleles can have
*major* effects in the development of a phenotype (the
structure of the organism's body over its lifetime).
--
John S Wilkins PhD - www.wilkins.id.au a little emptier, a
little spent as always by that quiver in the self,
subjugated, yes, and obedient. -- Seamus Heaney

"Brian Berns" <[email hidden]> wrote in
message > I have read many explanations of kin selection
that are based on

Quoted message said:

the "fact" that I share 1/2 of my genes with each of my
parents and siblings, 1/8 of my genes with each of my
first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve > this

problem for me? Thanks.

Quoted message said:

The technical term used is "identical by descent", which
maybe doesn't help much.

Imagine we have a new mutation which has been going for only
five or six generations and is thus still very rare. The
chance of this new mutation being in a relative is obviously
given by the 1/2, 1/8 metric and not the 99% one.

The point is that every new allele starts off as just such a
rare mutation, so we use the more restricted definition of
"related" when calculating whether altruism is adaptive.

Brian Berns <[email hidden]> wrote or quoted:

Quoted message said:

I have read many explanations of kin selection that are
based on the "fact" that I share 1/2 of my genes with each
of my parents and siblings, 1/8 of my genes with each of
my first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve this problem for me? Thanks.

The second claim is false.

Humans share about 99% of their nucleotides at corresponding
loci with each other.

Genes are different from nucleotides - they are long strings
of nucleotides.

In a similar manner, the two strings "old wig bat" and "one
big hat" share the majority of their letters - but none of
the words are the same.
--
__________
|im |yler timtyler.orgtimtyler.orgOpen ↗ [email hidden] Remove
lock to reply.

Brian Berns said:

I have read many explanations of kin selection that are
based on the "fact" that I share 1/2 of my genes with each
of my parents and siblings, 1/8 of my genes with each of
my first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve this problem for me? Thanks.

See:

'Twelve Misunderstandings of Kin Selection' Zeitschrift
fur Tierpsychologie 51, 184-200 Richard Dawkins

I think this one might have been no. 7 if memory serves.

PR

[email hidden] (Brian Berns) wrote in message news:<[email hidden]>...

Quoted message said:

I have read many explanations of kin selection that are
based on the "fact" that I share 1/2 of my genes with each
of my parents and siblings, 1/8 of my genes with each of
my first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve this problem for me? Thanks.

-- Brian

Brian,

You can go to www.google.com/groups and do a search in
sci.bio.evolution on the words: share alike.

I cut and pasted the results of my search below. Once you
get used to using this search engine you should have no
trouble finding these threads.

[moderator's note: I'd caution Brian and everyone else that
discussion in s.b.e. is hardly authoritative. While we have
our share of sober-minded evolution professionals, we also
have our share of eager tyros, and perhaps more than our
share of just plain contrariness. It may be a useful source
of references in the professional literature, however, and
in many cases, the talk.origins FAQ documents are also
useful, at www.talkorigins.org. Good luck. - JAH]

Related groups: sci.bio.evolution

Share and share alike? ... Now ask yourself this question:
was he talking about genes being shared (recieved from [the
active connotation of share]) or was he talking about genes
being ... sci.bio.evolution - Dec 30, 1999 by jimmcginn -
View Thread (1 article)

Re: Share and share not alike. As I indicated the problem is
that you (neoDarwinists) are using the two different
meanings of the word share in a tag team manner. ...
sci.bio.evolution - Mar 31, 2000 by jimmcginn - View Thread
(10 articles)

Re: The 99.9% vs 0% paradox ... on deja.com in
sci.bio.evolution (be sure to limit the search to
sci.bio.evolution) and search using the following words:
jimmcginn share alike Regards, Jim sci.bio.evolution - Oct
14, 2001 by Jim McGinn - View Thread (99 articles)

Re: Haeckel on the Net ... we have more similarity with
other primates than a bundle of bananas but at the
evolutionary level we share a continuum with primates and
bananas alike. ... sci.bio.evolution - Feb 12, 2003 by
Michael Ragland - View Thread (12 articles)

Re: Kin selection question ... Many people, scientists and
lay person alike, believe just as you do that IBD represents
the degree that you (share) have-in-common a certain allele
with any ... sci.bio.evolution - Dec 1, 1999 by jimmcginn -
View Thread (152 articles)

Re: Kin selection question [possible REPOST] ... It's the
genes you share that happen to be more or less fit ... If
you have genes that monitor how many other genes are alike
so you can choose who to help out, why ... sci.bio.evolution
- Jan 8, 2000 by Jonah Thomas - View Thread (80 articles)

"Malcolm" <[email hidden]> wrote in message news:<[email hidden]>...

<snip>

Quoted message said:

The technical term used is "identical by descent", which
maybe doesn't help much.

Imagine we have a new mutation which has been going for
only five or six generations and is thus still very
rare. The chance of this new mutation being in a
relative is obviously given by the 1/2, 1/8 metric and
not the 99% one.

The point is that every new allele starts off as just such
a rare mutation, so we use the more restricted definition
of "related" when calculating whether altruism is
adaptive.

I don't see your point. The implication seems to be that the
relatedness appropriate to Hamilton's rule will increase as
the allele becomes more frequent. That's clearly not the
case. Hamilton's rule makes no assumption about allele
frequencies, except for pq>0.

Also, as has been discussed extensively in sbe, relatedness
in Hamilton's rule is not a probability of identity by
descent measure. Hamilton's 'green beard' example makes this
quite clear.

Tim Tyler <[email hidden]> wrote

Quoted message said:

Humans share about 99% of their nucleotides at
corresponding loci with each other.

Genes are different from nucleotides - they are long
strings of nucleotides.

There is no difference between genes and nucleotides.

[moderator's note: Warning: this is not anything like
standard usage. Please search on Jim's previous articles to
get context for this claim. - JAH]

Jim

[email hidden] (Jim McGinn) wrote

Quoted message said:

[moderator's note: I'd caution Brian and everyone else
that discussion in s.b.e. is hardly authoritative. While
we have our share of sober-minded evolution professionals,
we also have our share of eager tyros, and perhaps more
than our share of just plain contrariness. It may be a
useful source of references in the professional
literature, however, and in many cases, the talk.origins
FAQ documents are also useful, at www.talkorigins.org.
Good luck. - JAH]

As I explain previously, the logic of Hamilton's kin
selection pivots off of nothing but an erroneous application
of english (specifically it involves misusage of the word
share which involves two different meanings of the word
being used interchangeably). When this misusage is corrected
Hamilton's kin selection falls like a house of cards. This
is true regardless of whether the person investigating it is
or is not a "sober-minded evolution professional."

[moderator's note: I agree that one should examine Jim's
previous claims on this matter. - JAH]

Jim

[email hidden] (Brian Berns) wrote in message news:<[email hidden]>...

Quoted message said:

I have read many explanations of kin selection that are
based on the "fact" that I share 1/2 of my genes with each
of my parents and siblings, 1/8 of my genes with each of
my first cousins, etc.

I have also seen it stated many times that each human
shares a very large fraction of his/her genes (90%? 95%?
99%? -- I forget the exact number) with every other human.

These two claims are obviously contradictory. Can anyone
resolve this problem for me? Thanks.

-- Brian

Brian,

You have received a confusing mix of responses from both
experts and "tyro"s. As another tyro, though I hope an
orthodox one, let me try to summarize:

1. The meaning of "shared" in your two cases is
different. "Shared" is a bad word in any case - if you
and I drive indistinguishable cars, does this mean
that we "share" cars?

2. As someone correctly points out, the number you get (90%,
95%, 99%) for genetic similarity between persons will
depend on the granularity you choose. You and I could be
99.99% identical at the base pair level, though only 95%
of our genes are absolutely identical. (The genes where
we differ are 99% identical at the base pair level, but
we are saying that that 1% difference in the gene
sequence makes the whole gene different.)

3. In any case, if you want to know how many of your genes
are identical with "every other human", the number is
probably low, less than 20%. For any particular gene, it
is very likely that at least one of the 6 billion of us
is a mutant.

4. These numbers are estimates anyways. No one has ever
completely sequenced two different people and compared
the sequences, let alone sequencing a large sample of
everybody.

5. Regarding kin selection, the numbers you quote are
correct, and the method of calculating them is sometimes
called IBD, or identity by descent. It turns out that
when inbreeding is taken into account, the IBD method is
slightly wrong, and a slightly different method is
called for.

6. But the important thing is that, whatever number happens
to be "right" for genes shared between two random people
- say it is 99% - the number of genes shared with your
sibling will be higher - 99.5%, say. That is, you and
your sibling "share" half of the genes that make you
different from other people. And that extra bit of
identity between you is enough to provide an incentive
for altruism.

However, the proof of this last statement is subtle, and it
is not just the "tyro"s who misunderstand it.

Jim

Jim Menegay <[email hidden]> wrote or quoted:

Quoted message said:

6. But the important thing is that, whatever number
happens to be "right" for genes shared between two
random people - say it is 99% - the number of genes
shared with your sibling will be higher - 99.5%, say.
That is, you and your sibling "share" half of the
genes that make you different from other people. And
that extra bit of identity between you is enough to
provide an incentive for altruism.

However, the proof of this last statement is subtle, and
it is not just the "tyro"s who misunderstand it.

A couple of "thought experiments" might throw some light on
this issue:

Firstly, it should be clear that the extent that altruistic
acts are directed towards relatives rather than strangers
depends not on their relatedness to one another - but to
their *percieved* relatedness to one another.

To see why this might matter, consider two populations:

One population used to use scent to identify relatives.
Pheremones were coded for my polymorphic genes - and you
could tell how closely someone is related to you by
sniffing them.

In the other population, relatedness was judged by whether
an organisms fed you as a child, grew up in the same nest as
you, or was one that you gave birth to, or perhaps that your
mate gave birth to.

Now imagine that both populations have recently been through
a bottleneck that destroyed most of the genetic diversity in
their gene pool, and left their genomes almost identical to
one another.

Genetic relatedness between randomly-selected individuals
within each population would be high.

However in one population, individuals would act as
though they were all related to one another - while in
the other, they would act as though "conventional"
relatedness rules applied.

So - we can see that any argument that rare genes that
promote altruism between relatives will be favoured
equally no matter how genetically homogeneous the
population is in need of some adjustment - since a whole
bunch of mechainsms that are normally available for
judging relatedness can get thrown out of joint by
population bottlenecks that destroy genetic diversity -
and it is percieved relatedness that matters.

Secondly, a theoretical doubt:

While the notion of a population persisting for very long
in a genetically homogenous state is probably not
terribly realistic, I /suppose/ it could happen in
principle - perhaps in a population located in a confined
space for an extended period, and also subjected to
strong selective forces that cause the effective
population size to be very small.

Would such a population exhibit greater altruism between
"unrelated" members of the same species than normal?

Aside from the argument about percieved relatedness being
greater in genetically uniform populations, there's another
reason for thinking that such populations might behave more
"altruistically" - i.e. more like a single big clonal
organism - than normal populations would: high-level
selection.

A usual argument (which I basically accept) for species
level selection being a weak and ineffective force invokes
the facts that species reproduction times are very long
compared to individual reproductive times - so the effects
of individual-level selection are correspondingly magnified
- and there's a lot of scope for individual level selection
destroying much of the variation in populations that species
level selection might work on.

However, a genetically homogeneous population would severely
hamper the progress individual-level selection can make -
and would be likely to be effective at reducing the
resulting rate of adaptive evolution.

This could tip the scales in favour of selection at higher
levels. Would it be enough to make the members of the
homogeneous species act together as a team - in their
competition with other species? Probably not - but it seems
to be a quantitative issue - and depends on the degree of
homogeneity and the relative reproduction rates in question.
--
__________
|im |yler timtyler.orgtimtyler.orgOpen ↗ [email hidden] Remove
lock to reply.

[email hidden] (Jim McGinn) wrote in message news:<[email hidden]>...

Quoted message said:

[email hidden] (Jim McGinn) wrote

Quoted message said:

[moderator's note: I'd caution Brian and everyone else
that discussion in s.b.e. is hardly authoritative. While
we have our share of sober-minded evolution
professionals, we also have our share of eager tyros,
and perhaps more than our share of just plain
contrariness. It may be a useful source of references in
the professional literature, however, and in many cases,
the talk.origins FAQ documents are also useful, at
www.talkorigins.org. Good luck. - JAH]

As I explain previously, the logic of Hamilton's kin
selection pivots off of nothing but an erroneous
application of english (specifically it involves misusage
of the word share which involves two different meanings of
the word being used interchangeably). When this misusage
is corrected Hamilton's kin selection falls like a house
of cards. This is true regardless of whether the person
investigating it is or is not a "sober-minded evolution
professional."

[moderator's note: I agree that one should examine Jim's
previous claims on this matter. - JAH]

Yes, Josh. Why don't you take the challenge that I offered
to Guy back when, you know where I asked him to choose one
of the two definitions of share and see if he could use it
throughout the argument. You won't do this though. You'll do
just as Guy did and simply refuse to answer the question.

It's really funny the degree that you professional
pretenders rely on the fact that few people actually think
any of this through.

Jim

"Name And Address Supplied"

Quoted message said:

Quoted message said:

The point is that every new allele starts off as just
such a rare mutation, so we use the more restricted
definition of "related" when calculating whether
altruism is adaptive.

I don't see your point. The implication seems to be that
the relatedness appropriate to Hamilton's rule will
increase as the allele becomes more frequent. That's
clearly not the case. Hamilton's rule makes no assumption
about allele frequencies, except for pq>0.

Hamilton's rule refers to the behaviour of the whole
organism, and we assume

partners. In a clonal species, pq is zero (there are no
alternative alleles) and the coefficient of relatedness is
one. We would predict perfect co-operation between clones,
but we would also predict periodic mutations for
freeloading, which then enjoy greater success. So the stable
situation is to treat only identifiable relatives as perfect
relations, to whom one shows absolute altruism.

Quoted message said:

Also, as has been discussed extensively in sbe,
relatedness in Hamilton's rule is not a probability of
identity by descent measure. Hamilton's 'green beard'
example makes this quite clear.

The "green beard" effect ( a gene has two effects, to cause
a man to grow a green eard and to be nice to other men with
green beards ) is an example where the rule breaks down,
because we are no longer using the coefficient of
relatedness to determine whether a person shares our genes,
but another identifier.

[email hidden] (Jim McGinn) wrote in message news:<[email hidden]>...

Quoted message said:

[email hidden] (Jim McGinn) wrote

Quoted message said:

[moderator's note: I'd caution Brian and everyone else
that discussion in s.b.e. is hardly authoritative. While
we have our share of sober-minded evolution
professionals, we also have our share of eager tyros,
and perhaps more than our share of just plain
contrariness. It may be a useful source of references in
the professional literature, however, and in many cases,
the talk.origins FAQ documents are also useful, at
www.talkorigins.org. Good luck. - JAH]

As I explain previously, the logic of Hamilton's kin
selection pivots off of nothing but an erroneous
application of english (specifically it involves misusage
of the word share which involves two different meanings of
the word being used interchangeably). When this misusage
is corrected Hamilton's kin selection falls like a house
of cards. This is true regardless of whether the person
investigating it is or is not a "sober-minded evolution
professional."

[moderator's note: I agree that one should examine Jim's
previous claims on this matter. - JAH]

I'm certainly interested to hear why Jim thinks that "the
logic of Hamilton's kin selection pivots off of nothing but
an erroneous application of english".

But this would require either

i) I do alot of work sifting through past posts to this
newsgroup to pick through threads and conversations to
try to glean exactly why Jim thinks this way

or

ii) Jim does alot of work re-typing out his argument, which
I would read, and determine whether he has anything
interesting to say about kin selection.

Neither seems entirely satisfactory, since if Jim is intent
on challenging the foundations of kin selection theory the
onus is really on him to bring the theory to us in an
intelligible form. But then, it would be unfair on Jim to
make him do all this typing each time someone enquires as to
his rationale for such bold claims.

I believe I have an answer. Couldn't Jim prepare an article
to be posted on the SBE website? In my opinion, a scientific
journal would be a far more appropriate context for such
discussion (I previously suggested the Journal of
Theoretical Biology, which after all is where Hamilton
published his 1964 papers) but it seems that Jim is
uncomfortable with this. So perhaps an article on the
webpage would make a suitable compromise?

What do you think, Jim?

in article [email hidden], Name And Address Supplied at
[email hidden] wrote on 6/5/04 10:43 PM:

Quoted message said:

"Malcolm" <[email hidden]> wrote in
message news:<[email hidden]>...

<snip>

Quoted message said:

The technical term used is "identical by descent", which
maybe doesn't help much.

Imagine we have a new mutation which has been going for
only five or six generations and is thus still very
rare. The chance of this new mutation being in a
relative is obviously given by the 1/2, 1/8 metric and
not the 99% one.

The point is that every new allele starts off as just
such a rare mutation, so we use the more restricted
definition of "related" when calculating whether altruism
is adaptive.

I don't see your point. The implication seems to be that
the relatedness appropriate to Hamilton's rule will
increase as the allele becomes more frequent. That's
clearly not the case. Hamilton's rule makes no assumption
about allele frequencies, except for pq>0.

This is not correct. Hamilton's Rule makes plenty of cryptic
assumptions, which is a primary source of confusion. Let me
give examples from both ends of the frequency spectrum
showing why Hamilton's Rule assumes a limited window of
frequency for the altruism mutation. The Rule assumes that
both the cost on benefit of the altruistic act affect the
frequency of the altruism allele in a deterministic and
invariant way. When the mutation is present in only one
copy, only the cost of the altruistic behavior affects the
fate of the allele; so Hamilton's Rule is an invalid and overly-
optimistic model in this case. When the altruism allele is
very common, then the coefficient of relatedness (r) is a
very poor predictor of the presence of the altruism allele
in a behavioral partner; thus the benefit of altruism does
not affect the frequency of the altruism allele with
probability "r" under these conditions either, as assumed by
Hamilton's Rule. For any given social structure and
phenotypic expression of an "altruism mutation", there would
be an optimum frequency of the altruism allele corresponding
to the maximum effect of kin selection; but I have never
seen this calculated for a given situation. It would also be
helpful to see an analysis of the rate at which the
effectiveness of kin selection diminishes as you move away
from this optimum.

Cheers,

Guy

In article <[email hidden]>,

Jim McGinn said:

As I explain previously, the logic of Hamilton's kin
selection pivots off of nothing but an erroneous
application of english (specifically it involves misusage
of the word share which involves two different meanings of
the word being used interchangeably). When this misusage is
corrected Hamilton's kin selection falls like a house of
cards. This is true regardless of whether the person
investigating it is or is not a "sober-minded evolution
professional."

Oh here we go again. There have been many debates on this
newsgroup about the logic of kin selection, between McGinn
and many others. McGinn is convinced he has refuted the
logic of Hamilton's rules for kin selection. No one else
seems to agree that he has. Before we rehash all this again,
interested persons should reread some of these previous
threads. In particular, I call attention to the thread "Re:
Part2 (Kin Selection)" which you can find by using that
phrase in Google Groups search, or by using McGinn
Felsenstein "kin selection" in Google Groups.

In a posting of 26 October 2002 I put forward a list of
assumptions which I hoped to use to demonstrate that in a
simple model, Hamilton's rule could be derived. McGinn
immediately (27 October 2002) objected to the assumptions,
saying that this approach "lacks causal validity" and
"involves the wholescale inclusion of whole sets of
unexamined assumptions". He would not put forward any simple
model of his own that could be shown to come to any
different conclusion than Hamilton's rule. He objected to
all simple models as oversimplified.

He describes the situation in dramatic terms ("falls like a
house of cards", "the confusion that ensues when such care
is not taken", "your continued belief in the validity of a
Hamilton's equation even though you are unable to
demonstrate it's validity", and "Hamilton's equations has
never amounted to anything but a rather vague analogy",
"Hamilton's fantasy", and "the establishment accepted
Hamilton's Rule and has since continued pretending that it
makes sense"😉.

Given our inability to get a model-based argument out of
McGinn, I asked in frustration (11 November 2002):

"Perhaps we should see whether he has convinced anyone at
all. Is there anyone else out there who thinks McGinn has
shown that Hamilton's result is invalid? If so, do they have
some model situation that could help us understand the logic
of that objection?"

Silence. No one agreed, even tentatively, with McGinn's
assertions. I suggest that his argument convinced no one
else. If so, this puts McGinn's dramatic descriptions into
perspective.

I hope that before a lengthy repetition of this argument,
that people reread some of this old debate, judge for
themselves whether they agree with McGinn. If they do, I
hope that they will explain here why they do.

--
Joe Felsenstein [email hidden]
Department of Genome Sciences and Department of
Biology, University of Washington, Box 357730, Seattle,
WA 98195-7730 USA

Tim Tyler <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

Firstly, it should be clear that the extent that
altruistic acts are directed towards relatives rather than
strangers depends not on their relatedness to one another
- but to their *percieved* relatedness to one another.

Actually, there are two possible reasons why altruistic acts
might be directed toward relatives more frequently than
towards non-relatives. One is what you have called
perception. The other is that you interact most frequently
with neighbors, and your neighbors may be a tiny fraction of
the population as a whole. But if most of your relatives
live in the neighborhood, then you can have a fair-sized "r"
value even if you are incapable of perceiving.

Quoted message said:

[snip] So - we can see that any argument that rare genes
that promote altruism between relatives will be favoured
equally no matter how genetically homogeneous the
population is in need of some adjustment - ...

No one has made such an argument. The claim was that even
common genes that promote altruism will be favored if rb>c.
The frequency of the altruistic gene is the only one that
matters, and I claim that it doesn't matter either!

"Guy Hoelzer" <[email hidden]> wrote in message
"]news:[email hidden]...

Quoted message said:

in article [email hidden], Name And
Address Supplied at [email hidden]
wrote on 6/5/04 10:43 PM:

Quoted message said:

"Malcolm" <[email hidden]> wrote in
message news:<[email hidden]>...

<snip>

Quoted message said:

The technical term used is "identical by descent",
which maybe doesn't

help

Quoted message said:

Quoted message said:

Quoted message said:

much.

Imagine we have a new mutation which has been going for
only five or

six

Quoted message said:

Quoted message said:

Quoted message said:

generations and is thus still very rare. The chance of
this new

mutation

Quoted message said:

Quoted message said:

Quoted message said:

being in a relative is obviously given by the 1/2, 1/8
metric and not

the

Quoted message said:

Quoted message said:

Quoted message said:

99% one.

The point is that every new allele starts off as just
such a rare

mutation,

Quoted message said:

Quoted message said:

Quoted message said:

so we use the more restricted definition of "related"
when calculating whether altruism is adaptive.

I don't see your point. The implication seems to be that
the relatedness appropriate to Hamilton's rule will
increase as the allele becomes more frequent. That's
clearly not the case. Hamilton's rule makes no
assumption about allele frequencies, except for pq>0.

This is not correct. Hamilton's Rule makes plenty of
cryptic assumptions, which is a primary source of
confusion. Let me give examples from both

ends

Quoted message said:

of the frequency spectrum showing why Hamilton's Rule
assumes a limited window of frequency for the altruism
mutation. The Rule assumes that both the cost on benefit
of the altruistic act affect the frequency of the altruism
allele in a deterministic and invariant way. When the
mutation

is

Quoted message said:

present in only one copy, only the cost of the altruistic
behavior affects the fate of the allele; so Hamilton's
Rule is an invalid and overly-optimistic model in this
case. When the altruism allele is very common, then the
coefficient of relatedness (r) is a very poor predictor

of

Quoted message said:

the presence of the altruism allele in a behavioral
partner; thus the benefit of altruism does not affect the
frequency of the altruism allele with probability "r"
under these conditions either, as assumed by

Hamilton's

Quoted message said:

Rule. For any given social structure and phenotypic
expression of an "altruism mutation", there would be an
optimum frequency of the altruism allele corresponding to
the maximum effect of kin selection; but I have never seen
this calculated for a given situation. It would also be

helpful

Quoted message said:

to see an analysis of the rate at which the effectiveness
of kin selection diminishes as you move away from this
optimum.

You are wrong about Hamilton's rule being frequency
sensitive in its applicability. The rule applies equally
well at all frequencies. If rb > c, then it is advantageous
to have the "altruistic allele". The frequency of the allele
in the population makes absolutely no difference in whether
it is advantageous. Where the frequency DOES make a
difference is in just how big of an advantage it is.
Hamilton [1964] talks about a "dilution factor" in
discussing this. That is, "r" is still the correct factor in
the rule, but there is a positive frequency dependent
dilution factor "d" such that the selection coefficient will
be proportional to d(rb-c).

At the high end of the frequency range, you can see your
error by asking whether the selfish allele can invade a
population of altruists. Notice that this simply multiplies
"b" and "c" by -1. The rule still applies and predicts that
the selfish allele will be disadvantageous.

At the low end, your error is more subtle. There is still an
advantage to carrying the only (dominant) gene for sibling
altruism in the population. This advantage is realized in
your own fitness, assuming that you compute fitness in the
way that Edser and Hamilton [1964] recommend - that is by
counting the number of offspring that survive to maturity.
Your offspring are more likely to survive, because they will
(often) have altruistic siblings. And even if you define
fitness as birth-to-birth, altruism genes are no more
problematic than genes for parental care. For cross-
generation effects, you really need to have a way of
defining overall fitness that includes the expected personal
fitnesses of your offspring.

"Malcolm" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

"Name And Address Supplied"

Quoted message said:

Quoted message said:

The point is that every new allele starts off as just
such a rare mutation, so we use the more restricted
definition of "related" when calculating whether
altruism is adaptive.

I don't see your point. The implication seems to be that
the relatedness appropriate to Hamilton's rule will
increase as the allele becomes more frequent. That's
clearly not the case. Hamilton's rule makes no
assumption about allele frequencies, except for pq>0.

Hamilton's rule refers to the behaviour of the whole
organism, and we assume

Quoted message said:

partners. In a clonal species, pq is zero (there are no
alternative alleles)

This is only true when every individual is genetically
identical. That doesn't seem very realistic, or relevant.

Quoted message said:

and the coefficient of relatedness is one.

The coefficient of relatedness is actually undefined in
this context.

Quoted message said:

We would predict perfect co-operation between clones, but
we would also predict periodic mutations for freeloading,
which then enjoy greater success.

And because the free-loaders have higher success, the
cooperators are out-competed, so we wouldn't expect
cooperation at all!

Quoted message said:

So the stable situation is to treat only identifiable
relatives as perfect relations, to whom one shows absolute
altruism.

Quoted message said:

Also, as has been discussed extensively in sbe,
relatedness in Hamilton's rule is not a probability of
identity by descent measure. Hamilton's 'green beard'
example makes this quite clear.

The "green beard" effect ( a gene has two effects, to
cause a man to grow a green eard and to be nice to other
men with green beards )

Actually, the greenbeard is much more general than that, in
Hamilton's discussion in 1964. Note that Dawkins coined the
term 'green beard' later, and the name has stuck.

Quoted message said:

is an example where the rule breaks down, because we are no
longer using the coefficient of relatedness to determine
whether a person shares our genes, but another identifier.

I believe you are confusing the coefficient of relatedness
with a measure of genealogical relationship. In some
circumstances it will not matter to do so, but as far back
as 1963 Hamilton (and others) were aware that this is not
what r in Hamilton's rule actually is.