From: Technotranscendence (neptune@mars.superlink.net)
Date: Sun Feb 04 2001 - 10:11:44 MST
On Saturday, February 03, 2001 9:09 AM John Marlow johnmarlow@gmx.net wrote:
> > However, the surprising thing should not be the amount of information,
but
> > the amount of noise in living systems. I'm sure we are all aware of DNA
> > changes -- to stick with DNA; the focus should not be on DNA alone --
many
> > random changes in DNA have no or little impact on anything else. Part
of
> > this is because DNA and proteins made from it -- indirectly, of
course --
> > have active sites and passive sites. Most of DNA and most of protein is
> > passive -- just there for the ride and as long as it doesn't affect
anything
> > else, it doesn't make a difference. An analogy might prove helpful
here. A
> > protein's active site might be compared to the tires on a car or any
other
> > vital component. The license plate cover on a car, however, is more
like a
> > passive site. It has no or little effect on performance. It might on
> > margin, but, for the most part, it does not.
> >
> > (Now, imagine if lots of random change accumulates in passive areas,
then an
> > active area changes or an area changes that makes formerly passive areas
> > active. You might wind up with a huge shift in DNA or protein
behavior.)
>
> **I believe what he's saying is that all, or at least many, eventualities
were coded
> for at or near inception. That "useless excess" DNA is not in fact
useless--but
> waiting to be triggered into usefulness by changing conditions.
That remains to be proved. Actually, the opposite seems to be the case.
See, e.g., the work of Motoo Kimura. The thing is a lot of this random
change fits mathematical models of being -- you guessed it! -- random. Now,
if one distinguishes between passive and active sites on DNA and protein and
one realizes, as I state above, that most of any strand of DNA or protein is
passive, it's not surprising that lots of random changes would have no
effect whatsoever on the phenotype -- and hence would not be inputs for
environmental [boundary] selection.
> > Genomes, in general, are very resilient to some disruptions. Even
though,
> > e.g., humans get cancer, even most cancer victims live long enough to
> > reproduce, raise their children, and so on. I would look at the problem
the
> > other way. Why is there so little information in any genome of any
> > relevance to the organism?
>
> **See above point. Little of relevance to the organism _in the
> present environment._
Well, if you're positing some forwarding-tracking mechanism -- some means by
which any genome can predict future environments -- then please elaborate.
I do not believe in such mechanisms. At least, I haven't seen a convincing
one yet. That said, there's no reason to discount orthogenetic processes,
such as S. A. Kauffman (in _The Origins of Order_ and elsewhere) has
detailed. But these processes do not seem to track future environments.
Instead, they are more akin to complex systems having certain inner
directions -- states toward which they tend.
Sometimes, I believe, the system (gene, genome, organism, population,
species, ecosystem, etc.) can head in only a few directions AND
environmental selection couples with it. This makes for rapid change which
might seem outwardly directed -- as if by some mind. A lot of this is
detailed in _Evolution as Entropy_ by Brooks and Wiley.
Cheers!
Daniel Ust
http://uweb.superlink.net/neptune/
Film recommendation: "The Color of Paradise."
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:05:35 MST