🚀 go-pugleaf

I've just run across something of relevance to Ken's ideas about
bandwidth limitations in rate of DNA transcription.  Some stretches of
DNA are repetitive, adding another layer of parallelism to the
"computations" involved.  As a specific example (until today I didn't
know it could be this involved) _Xenopus laevis_ (the south african
clawed frog) has 450 copies of the genes coding for 18S and 28S rRNA,
and 24,000 copies of the gene for 5S rRNA.  These RNA's code for the
machinery that creates proteins, and is not actually translated into
proteins itself.  One might infer from the numbers that if "bandwidth"
limitations got to be a problem, then more channels could be opened by
the addition of more copies of any particular gene.  I got the numbers
from _Evolutionary Biology_ by D. Futuyma.

Does anybody know if all of these copies are *really* identical?
Given that (in my understanding anyways) new genes arise largely from
copies of previously existing genes, how is it that the sequences of
the copies of the rRNA genes in _X. laevis_ don't mutate
independently?  (I'm still close to the beginning of Futuyma's book,
so please feel free to respond by telling me to keep reading :-)

Mickey Rowe     (rowe@pender.ee.upenn.edu)