Received: by alpheratz.cpm.aca.mmu.ac.uk id PAA27851 (8.6.9/5.3[ref firstname.lastname@example.org] for cpm.aca.mmu.ac.uk from email@example.com); Mon, 13 Aug 2001 15:42:42 +0100 Message-ID: <3B77CDBA.D68872BF@bioinf.man.ac.uk> Date: Mon, 13 Aug 2001 13:53:14 +0100 From: Chris Taylor <Christopher.Taylor@man.ac.uk> Organization: University of Manchester X-Mailer: Mozilla 4.77 [en] (Windows NT 5.0; U) X-Accept-Language: en To: firstname.lastname@example.org Subject: Re: Teleology etc. References: <3B730788.19486.9186B1@localhost> <3B742097.DE53E8D@bioinf.man.ac.uk> <001f01c12292$329ffd40$ed87b2d1@teddace> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: email@example.com Precedence: bulk Reply-To: firstname.lastname@example.org
> And who says you can cite molecular biology as an authority on life? What
> has molecular biology ever explained about life? Description is not the
> same as explanation. We know all kinds of things that go on in our cells,
> but we don't know why any of it happens. We still can't answer the basic
> question of what distinguishes a living cell from a dead cell. Why doesn't
> the living cell just stop and begin disintegrating? Molecular biology has
> no answer to this question. The *science* of biology hasn't been invented
> yet. At least Sheldrake is actually trying to explain life on its own terms
> rather than calling it a machine and then trying to explain that instead.
This isn't that hard to explain, the problems with defining life etc.
revolve more around conference politics than anything - we all know
roughly what we mean (at least for organic life anyway). The 'animate'
state of a piece of (organic) matter arises from the (temporarily)
third-law-defying dynamics, which are those of a driven system far from
equilibrium. Dead stuff is on a steady path to thermodynamic
equilibrium, living stuff is not as long as it is alive.
It's not a binary thing either. If you shot me, and I was 'dead' then in
some bits of my body, genes would be being expressed for maybe half an
hour, other metabolic stuff would be going on too. There is no magic
there (that wasn't meant to be a pejorative pun btw), just a set of
reactions that do some stuff, and rely on each other in a closed ntwork
> > 2) Protein folding is rather complex -
> So complex in fact that we have absolutely no idea what's going on in there.
> You'd think by now we'd have some kind of model or map of how this process
> is carried out. After all, we have very powerful computers than can model
> extremely complex processes. But protein folding is so unimaginably complex
> that no one's ever been foolish enough to try to model it (though in theory
> the world's most powerful supercomputers could do the job if they worked at
> it continuously for the next 100 years.)
'In theory'? As in perfectly feasible given a decent computer (or the
> many chaperones help out,
> > different cellular compartments are involved, as are timing effects to
> > allow local folding. You need a concept of an energy landscape, which is
> > 'out there' in a sense(...), but you most emphatically do not need
> > mystery fields of force.
> It's all descriptive. We know these "chaperones" are involved, but they
> don't explain why it happens the way it does. It's just assumed that
> someday we'll have an explanation.
> The question is what controls this unbelievably complex process. If it's
> controlled from our genes, then our genes must be vastly more powerful than
> any supercomputer ever devised. The only other option is that it's somehow
Nothing controls it. Proteins fold to reduce the overall free energy
(from hydrogen bonds) in the system (that's what keeps the two strands
of DNA together too). This is no more complex than oil drops joining up
on the surface of water - the system seeks its lowest energy
configuration by *randomly* exploring the space of all configurational
states. Chaperones guide the folding away from the wrong low energy
configurations (like the BSE disease form of the prion protein) along
with other tricks (like piping into the ER, codon adaptation index
exploitation and so on), but proteins would fold into something anyway,
it's just a question of what.
As for evolving the things in the first place, again random search of
sequence space (with the implied search of protein configurational
space) does it just fine.
I snipped the complexity stuff about why ants are not like bees because
someone else took up the cudgels on that one (in case you thought I
> > > Memes not a product of genes, so must be from MR etc. etc.
> > Uh-uh - the whole point of this group is the study of culturally
> > heritable patterns - heritable as in copyable. No need for any ethereal
> > templates. And again, where do the first ones come from? Evolution by
> > natural selection operating on variation explains this diversification
> > for me, what does MR have to say about it (genuine question)?
> MR offers a model of evolution that gives organisms an active role in
> shaping themselves. We know, for instance, that camels begin developing
> calluses on their kneepads when they're still in the womb. This would
> suggest that camels who developed calluses as a result of kneeling in the
> desert passed this trait onto their offspring. Since behavior can't
> directly affect genes, the logical assumption is that the calluses are
> passed on non-genetically. Otherwise we must accept the colossal
> improbability that the genetic mutation for calluses on the kneepads just
> happened to appear right when the camels needed it. You'd think they'd have
> to have gone through a lot of useless mutations first, like calluses in
> other places, or the wrong alterations on kneepads before they'd hit on the
> right mutation. How many millions of years should it have taken for them to
> get the right mutation? Now consider the fact that this applies many times
> over for every species on earth, and you start to see just how high that
> mountain of improbability is. Sheldrake offers a more streamlined, elegant
> model of evolution.
Baldwin effect. Guy kept developing drosophila at a high temperature,
and found that a particular pattern of wing venation occurred (but only
at the high temp). However after breeding several generations at this
high temperature the new venation pattern became fixed, even when at low
temp. This is because you are selecting out from the existing variety
(heterozygosity drops like a stone when you do this sort of thing),
combinations of genes that do the thing best (bear in mind that the
venation pattern is just the visible change for a shift in genotype
arising from the usual selective mass killing). This would involve (in
the simplest explanation) fixation of null alleles (duds basically) that
systemically turn on a particular pathway.
Camels have selection pressure for really decent Doc Marten's kneepads
(apparently, who thinks of these examples?). But once the trait is fixed
in adults, why is it so hard to imagine that selection for building the
structures earlier in development could occur (timing changes are fairly
easy)? After all, many animals are born with their cryptic patterning in
place because they need it, many can run within minutes; lots happens as
selection finds ways to better prepare offsping for the future.
> There was indeed an experiment involving island monkeys and their ability to
> learn to wash sand off potatoes before eating them. The speed of learning
> proceeded at a slow pace at first, and then overnight all the monkeys seemed
> to have learned the method. Unfortunately, this experiment was not properly
> controlled, so Sheldrake refuses to invoke it in support of his theory. He
> prefers experiments, such as the water maze experiments with rats, that have
> never been disputed.
He does very well not to rely on an interpretation of one night's
observation of one group.
What's the rat thing (genuine question)? If it's a 'too good to be
learned' thing his controls had better be damn good...
Cheers Ted, Chris.
Chris Taylor (email@example.com)
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
This archive was generated by hypermail 2b29 : Mon Aug 13 2001 - 15:59:41 BST