molecular clocks

Tom Holroyd tomh at BAMBI.CCS.FAU.EDU
Fri Sep 6 14:25:44 EST 1991


Thanks to all the folks who sent replies; there were so many it will
take me a day or two to respond to all of them.  Here's one that was
posted, so it gets replied to with a post:

>From: donnel at helix.nih.gov (Donald A. Lehn)

>It is probably safe to assume that the mutational rate of DNA is
>fairly constant ( 5e-9 nucleotide substitutions per base per year).

Why?  I'm not sure that we even know all the possible mechanisms of
mutation, let alone the time constants, if any.  I can think of cosmic
rays, replication errors, and any number of carcinogens and other
chemicals that bind to DNA, all of which produce point mutations.
These processes must all happen with different rates, and most are
not simple Poisson processes.  Environmental influences and so on.

>However all mutations are not created equal.  The vast majority of
>mutations in mammals are in regions of 'junk' DNA and are silent.

I bet if I removed all your 'junk' DNA, you'd die horribly of 50
different things, all at the same time.  How do we know if a DNA
region is junk?  It might be simply structural, and the length is
more important than the codons..  There could be regulatory mechanisms
we know nothing about.  True, though, some mutations will have less
effect than others.

>For instance, histones H3 and
>H4 are virtually identical in all eukaryotes. ...
>This doesn't imply that the H3/H4 DNA mutates
>less frequently only that mutations there confer a strong selective
>disadvantage.  As this illustrates,  the rate of evolution is multivarient.
>It is not only a function of the rate of mutation of DNA with time
>(which is virtually constant) but also the rate at which these mutations
>can be propagated in future generations.  Thus, the determining factor
>is more often the selective pressures in the environment.

Yes, the histone example is a good one.  Different proteins show different
mutation rates.  And yes, the changes in environment are going to affect
the mutation rate.  The molecular clock people try to find proteins or
DNA that are somehow immune to these confounding influences of rate.
I have a paper at home (I'll post the ref. later) that shows variations
in mutation rate in constant (laboratory) environments.  So maybe there
are no molecules that are immune to rate changes.  This could imply
a non-constant base rate.

Someone pointed out the effect of population size on genetic drift.
Even disadvantageous alleles can become fixed if the population is
small.  Pop. size will have a large effect on the rate of evolution.

So we agree that even if the base rate WERE constant, it wouldn't always
be observable (or maybe never observable).  So the issue of constant
mutation rate is moot, and molecular clocks are still poor timekeepers.

Tom Holroyd
Center for Complex Systems
Florida Atlantic University
tomh at bambi.ccs.fau.edu




More information about the Mol-evol mailing list