noncoding genes

[BillS@xxxxxxxxxxxxx (Bill Shear)]

Let's see if I can get this right (corrections welcome!).  If we define
noncoding genes are those which do not affect the phenotype, then they
should not be subject to natural selection.  Any changes in them should be
due to random processes which might be expected to occur at a constant
rate, averaged over a long period of time. Since the changes are random,
identical changes can be taken to be homologous and would indicate descent
from a common ancestor.  The number of changes relative to another allele
would be proportional to the time since divergence of the two alleles.
Natural selection messes this up because varying intensities of selection
cause different rates of evolution at different times.  Indeed in molecular
phylogenies, use of sequences from rapidly evolving lineages can change the
position of that lineage in the cladogram, making it seem unaccountably
ancient.  This explains a fair amount of the disagreement in molecular
phylogenies.

This represents my understanding--it may be way off.

There is an older book by Bruce Wallace (title forgotten, something like
"Genes, Chromosomes and Evolution") that expounded this idea with regard to
changes in chromosome structure for Hawaiian Drosophila--much more clearly
and in more detail than the above paragraph.

I kind of regret have blown off our geneticist friend too harshly.  But,
gosh it bothers me when people want to disregard a whole legitimate field
of study because they have "the answer."  One is reminded of the so-called
CENTRAL DOGMA of molecular biology which turned out not to be true with the
discovery of reverse transcriptase.   Scientists should "hold their ideas
lightly" because, of course, the whole enterprise is about falsification of
hypotheses.


Best wishes, Bill
___________________
William A. Shear
Department of Biology
Hamden-Sydney College
Hampden-Sydney VA 23943 USA
phone (804) 223-6172
FAX (804) 223-6374