Re: Conservation of information (long)

Tim Ikeda (
Sat, 21 Aug 1999 00:47:21 -0400

I just got Spenter's book _Not by chance_. The first I heard
about it was from Art Chadwick, who mentioned the book in
this group some months ago.

As I suspected from his web articles and from descriptions
of others who read his book, Lee's descriptions of what
constitutes biological information bounces around quite
a bit.

In chapter 3, Lee talks a little bit about the ability
of selection to add information to the genome.

"When a mutation occurs, selection can choose only
between the mutant and the rest of the population.
It can choose the better from the good, the more adaptive
from the less adaptive. IN one step, selection can can no
more than one bit of information. That's because is
makes only a binary choice between YES and NO, no matter
how complex the two options." pp 71.

He later goes on to suggest reasons why the information
generated may be less than one bit because the options
may be biased one way or another.

Reading on I find him again trying to link to selection,
but in a different way.

"Changing an amino acid in a protein very often affects
the way the protein functions. An organism is generally
well adapted to its niche**. Its proteins are well suited
to carrying out their functions. A change in one of its
proteins is then likely to degrade the organism in some
way. In particular, when an organism becomes resistant to
a drug through a change in one of its proteins, it is
likely to become less fit in some other way. Of course,
so long as the drug is present, the organism has to
be resistant to survive, even at the price of being
less fit in another way. But when the drug is removed,
the nonresistant type is again more adaptive." pp 143-144

** A truism. Note also that the proteins in a bacterium living in a
broth of 25 ug/ml streptomycin without the benefit of streptomycin
resistance are not well suited for carrying out their functions.

Anyway, from the above we see Spetner trying to quantitate
information in terms of "normal function", where anything
deviating from the norm or not performing better in all possible
circumstances is a loss of information. This is interesting
because I can't think of any change we can make to an organism's
genome (including adding new genes which Spetner does agree is
an increase of information), that: A) Won't disrupt the
normal functions of a cell, and B) Wouldn't make the
organism less fit in some other environment.

Later, in chapter 5, he discusses information in terms
of the sequence or functional specificity of enzymes.
In one passage (which I had only heard second-hand and
thought that Lee could not have possibly said), Spetner
claims that in the cases where some bacteria developed
streptomycin resistance by losing the ability to
bind the antibiotic, one should classify this as a
loss of information.

But here's the rub: If we take Lee's arguments at face value,
where selection can add information to the genome and losing
the ability to bind a lethal antibiotic is taken as a loss,
what's the net effect? Are the "currencies" of information
exchangeable -- Do the units match?

Later in the chapter Lee claims that a mutation which permits
an enzyme to recognize other substrate and thus survive on
other carbohydrate sources must again count as a loss of
information. This time, he says it is because the enzyme has
lost specificity for particular substrate.

Again, this statement is a bit difficult to reconcile with his
earlier metrics. In the case of strep resistance, a loss of
information was claimed because the ribosome _lost_ the ability
to bind a molecule. But in the latter case with carbohydrate
metabolism, information loss is claimed because the enzyme
_gained_ the ability to interact with additional molecules.

Overall, I've found few surprises in Lee Spetner's book --
This is a good thing because it means that most of what
is accessible on the net (for example, Art's earlier
descriptions in this group) is indeed a fair account of
what Lee says in his book. Since we've discussed most of
the stuff about Lee's peculiar and unmatched ideas about
biological information previously, I don't see much more
to add here. (The book is pretty light reading; I
suspect my copy will go into the lending
library pretty quickly).

What you might not get from the web are Lee's positions
on other aspects of the creation/evolution debate. He
appears to be an old-earther with a strong leaning toward
common descent. He is very interested in directional
mutations and his ideas seem similar to Mike Behe's
notion that life was created "pre packed" with the ability
to evolve. Evolving is this case means to activate and
actively re configure already embedded structures in
the genome in response to environmental pressures or
cues. I did not see him invoke intrusive divine intervention
in evolution (his term: Non-random evolution) beyond the
creation of life (or possibly a few forms of life). Thus
I can see why creationist web sites have tended to pull
ideas out of Lee's book to bash evolution but quickly skip
through his ideas in other passages.

Tim Ikeda