IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

Evolution Simulator

Elpeleg On Elpeleg at On.to
Fri Jan 19 13:02:45 EST 2001

"Arlin Stoltzfus" <arlin at carb.nist.gov> wrote in message
news:93vuc5$4dv$1 at mercury.hgmp.mrc.ac.uk...
> Manfred Knorr wrote:
> >
> > Does anyone know of a program that will simulate a living organism from
> > early in earth's history that will interact and evolve with it's
> > environment? I don't mean something like a bunch of smiley faces on a
> > computer screen. Rather a simple single celled animal that , given
> > program run time will evolve into a multi celled organism and perhaps
> > further, developing a mouth, light sensing organs etc.
> >
> > Of course it would need to be sped up somewhat from real time. I don't
> > to sit in front of my computer for a billion years just to see my life
> > become 2 celled.

Check Richard Dawkins book: The Blind Watch  Maker, he wrote about a program
he used there.

On Elpeleg

> You are asking too much.  Multicellular organisms became prominent about
> 2 to 3 billion years after life first evolved.  At just one generation
> per month, this interval would correspond to more than 10^10 generations.
> I've done a few population simulations that go on for millions of
> generations, but this was only possible because the model was simple
> and it skipped over most of the generations while waiting for mutations
> to happen.  Billions of generations are totally out of the question.
> By a "simple" model I mean that there were a few numbers to represent
> genotypic values, a few possible genotypes that could be reached by
> mutation, constant fitness values, and no explicit model of development,
> environment, or phenotypes.
> To use computer models to study evolution, you have to make difficult
> and highly problematic choices that amount to hypothesizing about
> the nature of the evolutionary process.  If  you didn't go through
> this process of abstraction, but were somehow able to recreate all of
> the complexity of life in your model, then it wouldn't be a "model"
> and, after finishing your simulations, you wouldn't have learned
> anything.  It is this need for abstraction that accounts for the
> use of theories, hypothesis-testing, and smiley faces on computer
> screens.
> Arlin
> ---


More information about the Mol-evol mailing list

Send comments to us at biosci-help [At] net.bio.net