🚀 go-pugleaf

[Disclaimer:  I have missed the beginning of this thread.  If I repeat
anything below which has been said before, I apologize in advance.]

In article <1992Dec16.144720.18569@husc3.harvard.edu> robison1@husc10.harvard.edu (Keith Robison) writes:
>Isn't just my signature convincing enough?  :-)
>
>	I have been asked to explain further my recent comment here
>that:
>
>	Any sort of heavy negative selection (large parasite load) can
>        cause speciation.  This has been shown in Tom Ray's Tierra stuff.
>        If some host's dodge one way, and some the other, it can
>        eventually lead to incompatibility.
>
>Tom Ray's work has been published in the Sante Fe Institute's 
>proceedings for their Artificial Life conferences.  I don't have
>the paper's handy, so I can't check them right now.  The work
>I am referring to may be unpublished (I saw it as a demo when
>I was an undergrad at Delaware)
>
>In Ray's system, the organisms are essentially asexual (although 
>recombination occasionally occurs).  If in his system you start with
>only a parasite and it's host, new forms of the host quickly arise,
>(which are immune) to be followed by new forms of the parasite.          
>(which can parasitize the new form).  Thus what I suggested probably
>holds true for asexual organisms -- strong negative selection can
>cause speciation.

What exactly does "negative" selection mean?  I would argue that
selection is selection:  some genotypes are more likely to reproduce
than others.

As for parasites leading to speciation, I suppose it depends on how
you define "species" in the case of asexual organisms.  The definition
above seems to be, "any new genotype."  Mind you, I am what W. D.
Hamilton has referred to as a "host-parasite coevolution freak"
myself, and I'm glad when anything can be explained in terms of
parasites :-), but it doesn't seem like any sort of new mechanism has
been offered here; since "incompatibility" means nothing in an asexual
population, there's no need for it here.

>What about for sexual species?  Well, the same should apply IF
>the "dodges" used by different variants affect the fitness of
>hybrids between those variants, such as changes in reproductive
>structures (macro or micro), alterations in the timing of reproduction,
>gross genetic rearrangements, etc.  

Here's where "incompatibility" is more likely to become important.  I
would argue, however, that any "revolutionary" strategy for avoiding
parasitism (or any other strong selective pressure) would spread
rapidly through a sexual population, and not cause speciation unless
there was significant population subdivision.  If population
subdivision was being assumed, I apologize for reinventing the wheel.

>Keith Robison
>robison@biosun.harvard.edu 
-- 
--Andy
     "God is a real estate developer / with offices around the nation
          They say one day he'll liquidate / his holdings on High
             I say it's all speculation." -- Michelle Shocked

adpeters@sunflower.bio.indiana.edu (Andy Peters) writes:

>[Disclaimer:  I have missed the beginning of this thread.  If I repeat
>anything below which has been said before, I apologize in advance.]

>In article <1992Dec16.144720.18569@husc3.harvard.edu> robison1@husc10.harvard.edu (Keith Robison) writes:
>>Isn't just my signature convincing enough?  :-)
>>
>>	I have been asked to explain further my recent comment here
>>that:
>>
>>	Any sort of heavy negative selection (large parasite load) can
>>        cause speciation.  This has been shown in Tom Ray's Tierra stuff.
>>        If some host's dodge one way, and some the other, it can
>>        eventually lead to incompatibility.
>>
>>Tom Ray's work has been published in the Sante Fe Institute's 
>>proceedings for their Artificial Life conferences.  I don't have
>>the paper's handy, so I can't check them right now.  The work
>>I am referring to may be unpublished (I saw it as a demo when
>>I was an undergrad at Delaware)
>>
>>In Ray's system, the organisms are essentially asexual (although 
>>recombination occasionally occurs).  If in his system you start with
>>only a parasite and it's host, new forms of the host quickly arise,
>>(which are immune) to be followed by new forms of the parasite.          
>>(which can parasitize the new form).  Thus what I suggested probably
>>holds true for asexual organisms -- strong negative selection can
>>cause speciation.

>What exactly does "negative" selection mean?  I would argue that
>selection is selection:  some genotypes are more likely to reproduce
>than others.

Agreed. If one genotype is experiencing negative selection, another
is being positively selected.  What I should have said is "extreme
selection".


>As for parasites leading to speciation, I suppose it depends on how
>you define "species" in the case of asexual organisms.  The definition
>above seems to be, "any new genotype."  Mind you, I am what W. D.
>Hamilton has referred to as a "host-parasite coevolution freak"
>myself, and I'm glad when anything can be explained in terms of
>parasites :-), but it doesn't seem like any sort of new mechanism has
>been offered here; since "incompatibility" means nothing in an asexual
>population, there's no need for it here.

Of course, many species in this world don't fall neatly into 
"sexual" and "asexual", and perhaps it's in that realm where these
effects might be most relevant.

>>What about for sexual species?  Well, the same should apply IF
>>the "dodges" used by different variants affect the fitness of
>>hybrids between those variants, such as changes in reproductive
>>structures (macro or micro), alterations in the timing of reproduction,
>>gross genetic rearrangements, etc.  

>Here's where "incompatibility" is more likely to become important.  I
>would argue, however, that any "revolutionary" strategy for avoiding
>parasitism (or any other strong selective pressure) would spread
>rapidly through a sexual population, and not cause speciation unless
>there was significant population subdivision.  If population
>subdivision was being assumed, I apologize for reinventing the wheel.

>>Keith Robison
>>robison@biosun.harvard.edu 
>-- 
>--Andy

Again, this phenomenon of selection-driven sympatric speciation would require
the following conditions:

	(1) Multiple possible beneficial adaptations ("dodges")
	(2) The different adaptations are either 
	    a. inherently incompatible
	    b. reduced fitness of hybrids

Imagine the following scenario:  suppose you have an organism (A) which
generally reproduces asexually but occasionally uses sexual reproduction.
Suppose further that another species (B) begins preying on the first,
using the sexual apparatus as it's route of attack.  A situation such
as this exists in E.coli, whose F (sex) pilus is also the receptor
for filamentous phages such as M13.  There might be multiple alterations
of A's sexual apparatus which might reduce predation by B.  However,
these alterations may also hinder the ability of an altered A to mate
with other A's bearing different modifications.  Alternatively, some
B may mutate so that they can prey on particular modifications of A, but
do this by trading off their ability to prey on the wild-type A.
Here is a summary:

			mates with    attacked by
wild-type	A	   A		B
		A1	   A1		B1
		A2	   A2	 	B2


If this were the case, then it would definitely be disadvantageous
to acquire an A1-A2 phenotype, because then both B1 and B2 would be
predators.  

Again, just throwing out thoughts.  Trash away!


Keith Robison
Harvard University
Department of Cellular & Developmental Biology
Department of Genetics / HHMI

robison@biosun.harvard.edu 

In article <robison1.724623436@husc10> robison1@husc10.harvard.edu (Keith Robison) writes:
>adpeters@sunflower.bio.indiana.edu (Andy Peters) writes:
>
>>As for parasites leading to speciation, I suppose it depends on how
>>you define "species" in the case of asexual organisms.  The definition
>>above seems to be, "any new genotype."  Mind you, I am what W. D.
>>Hamilton has referred to as a "host-parasite coevolution freak"
>>myself, and I'm glad when anything can be explained in terms of
>>parasites :-), but it doesn't seem like any sort of new mechanism has
>>been offered here; since "incompatibility" means nothing in an asexual
>>population, there's no need for it here.
>
>Of course, many species in this world don't fall neatly into 
>"sexual" and "asexual", and perhaps it's in that realm where these
>effects might be most relevant.

>>>What about for sexual species?  Well, the same should apply IF
>>>the "dodges" used by different variants affect the fitness of
>>>hybrids between those variants, such as changes in reproductive
>>>structures (macro or micro), alterations in the timing of reproduction,
>>>gross genetic rearrangements, etc.  
>
>>Here's where "incompatibility" is more likely to become important.  I
>>would argue, however, that any "revolutionary" strategy for avoiding
>>parasitism (or any other strong selective pressure) would spread
>>rapidly through a sexual population, and not cause speciation unless
>>there was significant population subdivision.  If population
>>subdivision was being assumed, I apologize for reinventing the wheel.
>
>>>Keith Robison
>>>robison@biosun.harvard.edu 
>>-- 
>>--Andy
>
>Again, this phenomenon of selection-driven sympatric speciation would require
>the following conditions:
>
>	(1) Multiple possible beneficial adaptations ("dodges")
>	(2) The different adaptations are either 
>	    a. inherently incompatible
>	    b. reduced fitness of hybrids
>
>Imagine the following scenario:  suppose you have an organism (A) which
>generally reproduces asexually but occasionally uses sexual reproduction.
>Suppose further that another species (B) begins preying on the first,
>using the sexual apparatus as it's route of attack.  A situation such
>as this exists in E.coli, whose F (sex) pilus is also the receptor
>for filamentous phages such as M13.  There might be multiple alterations
>of A's sexual apparatus which might reduce predation by B.  However,
>these alterations may also hinder the ability of an altered A to mate
>with other A's bearing different modifications.  Alternatively, some
>B may mutate so that they can prey on particular modifications of A, but
>do this by trading off their ability to prey on the wild-type A.
>Here is a summary:
>
>			mates with    attacked by
>wild-type	A	   A		B
>		A1	   A1		B1
>		A2	   A2	 	B2
>
>
>If this were the case, then it would definitely be disadvantageous
>to acquire an A1-A2 phenotype, because then both B1 and B2 would be
>predators.  

This mechanism would certainly work if all the assumptions were met.
Are there really multiple "dodges?"  Is it safe to assume that an
A1-A2 phenotype would be susceptible to both B1 and B2; if they're two
modifications of the same feature, mightn't crossing lead to a third
modification?  Most importantly, is there any reason to expect _both_
strategy A1 and strategy A2 to arise at about the same time?  If, say,
only A1 arises, it will spread through the population on its own,
completely replacing A; only if two novel strategies arise at the same
time can I see this mechanism working.  Don't get me wrong, the
mechanism seems viable, given the assumptions.  I just wonder how
often the conditions are met in the real world.

>Keith Robison
>robison@biosun.harvard.edu 


-- 
--Andy
     "God is a real estate developer / with offices around the nation
          They say one day he'll liquidate / his holdings on High
             I say it's all speculation." -- Michelle Shocked

adpeters@sunflower.bio.indiana.edu (Andy Peters) writes:

>[Disclaimer:  I have missed the beginning of this thread.  If I repeat
>anything below which has been said before, I apologize in advance.]

No disclaimer necessary. You're right on the ball.

>>	I have been asked to explain further my recent comment here
>>that:
>>
>>	Any sort of heavy negative selection (large parasite load) can
>>        cause speciation.  This has been shown in Tom Ray's Tierra stuff.
>>        If some host's dodge one way, and some the other, it can
>>        eventually lead to incompatibility.
>>
>>Tom Ray's work has been published in the Sante Fe Institute's 
>>proceedings for their Artificial Life conferences.  I don't have
>>the paper's handy, so I can't check them right now.  The work
>>I am referring to may be unpublished (I saw it as a demo when
>>I was an undergrad at Delaware)
>>
>>In Ray's system, the organisms are essentially asexual (although 
>>recombination occasionally occurs).  If in his system you start with
>>only a parasite and it's host, new forms of the host quickly arise,
>>(which are immune) to be followed by new forms of the parasite.          
>>(which can parasitize the new form).  Thus what I suggested probably
>>holds true for asexual organisms -- strong negative selection can
>>cause speciation.

>What exactly does "negative" selection mean?  I would argue that
>selection is selection:  some genotypes are more likely to reproduce
>than others.

I would interpret "negative selection", in the light of the rest of
Keith's post, as "disruptive selection". i.e. a bimodal distribution
of a trait is produced by selection against the mean of the original,
unimodal, distribution (see Mayr "Animal species and evolution").

>As for parasites leading to speciation, I suppose it depends on how
>you define "species" in the case of asexual organisms.  The definition
>above seems to be, "any new genotype."  Mind you, I am what W. D.
>Hamilton has referred to as a "host-parasite coevolution freak"
>myself, and I'm glad when anything can be explained in terms of
>parasites :-), but it doesn't seem like any sort of new mechanism has
>been offered here; since "incompatibility" means nothing in an asexual
>population, there's no need for it here.

I agree. Species concepts that are useful for sexual species are generally
difficult to apply to asexual species with any meaning.

>>What about for sexual species?  Well, the same should apply IF
>>the "dodges" used by different variants affect the fitness of
>>hybrids between those variants, such as changes in reproductive
>>structures (macro or micro), alterations in the timing of reproduction,
>>gross genetic rearrangements, etc.  

>Here's where "incompatibility" is more likely to become important.  I
>would argue, however, that any "revolutionary" strategy for avoiding
>parasitism (or any other strong selective pressure) would spread
>rapidly through a sexual population, and not cause speciation unless
>there was significant population subdivision.  If population
>subdivision was being assumed, I apologize for reinventing the wheel.

I agree here, too. Unless enough of the population "dodge" one way, and
enough "dodge" the other _all at once_ in a Goldschmidt "hopeful
monster" sort of way (highly unlikely IMHO), the genes for each strategy
would spread through the population, without causing speciation (unless
there was physical isolation of these subpopulations, as you say).
The most that you would get would be a polymorphism in the population
for the trait that deals with the parasite.

Simon.

-- 
Simon Blomberg                                   simonb@extro.ucc.su.OZ.AU
School of Biol. Sciences, A08
University of Sydney           We demand the right to stop to look even at
NSW   2006  Australia          lizards, who are nobody's property. -JBS Haldane