What is evolution?

Tom Schneider toms at fcs260c2.ncifcrf.gov
Mon Sep 16 22:03:25 EST 1991

In article <1991Sep15.125452.8523 at gpu.utcs.utoronto.ca> lamoran at gpu.utcs.utoronto.ca
(L.A. Moran) writes:

>Please note that in spite of the fact that Tom referred to the standard
>definition as "idiocy" and "perverted" I have avoided flaming back....
>even though he obvioulsy doesn't know what he is talking about. (-:

No insult taken!  I used those words because the standard definition (wow
you really called it that!) irritates me.

>Tom, why do you believe that fixation of neutral mutations in cytochrome c
>is NOT evolution? Do you have another word to describe this process?
>What is your personal definition of evolution? 

For cytochrome c, consider the set of all variants of the protein which satisfy
the functional requirements of the organism.  As has surprised a number of
investigators in the last year or so, many variants of a protein can be made
with little effect on function.  Consider the sequence logo of this set
(PostScript example is in anonymous ftp to ncifcrf.gov in
pub/delila/globin.logo.Z; documentation is NAR 18: 6097, 1990.) If a species
starts with one particular sequence AND NO FUNCTIONAL REQUIREMENTS CHANGE (I
stipulate this merely to make discussion simpler), then the sequence logo does
not alter with time.  There is no evolution, even if a different sequence is
fixed, because the sequence logo made from the altered strain would be the same
as the sequence logo from the unaltered strain.

Two measures are important.  First, in the example above, the information
content of the set did not change.  Second, the PATTERN (which one sees by
using a sequence logo) did not change.  I would say evolution occurs when the
information content changes.  In the case of protein it would seem that one
should also count changes in the pattern.  (Ie, the logo would look
different.)  For nucleic acid binding sites, it is not so clear that changes in
the pattern have any functional significance, so long as the binding still
takes place.

Notice that it is the statistical population, corresponding to a (I think)
macrocanonical ensemble in statistical mechanics, which is the indicator of
evolution.  This is why changes that don't matter in a protein are equivalent
to rearrangements of water molecules in a glass of water.  Such rearrangements
don't affect the overall state of the system.

>Laurence A. Moran (Larry)
>Dept. of Biochemistry

  Tom Schneider
  National Cancer Institute
  Laboratory of Mathematical Biology
  Frederick, Maryland  21702-1201
  toms at ncifcrf.gov

More information about the Mol-evol mailing list