Subject: Re: genetic code & mutation rates
Also sprach Steve LaBonne
Date: 29 Jul 1995 00:08:16 GMT
Message-ID: <3vbu5g$bc at rebecca.albany.edu>
Protein synthesis _is_ a chemical process and the chemistry cannot be
safely neglected. "Changing the code" requires only changing the
sequence of an anticodon, generally by just a single nucleotide. Only
the obvious selection pressure resulting from screwing up many
proteins simulataneously prevents much larger and more frequent
deviations from the "universal" code. Furthermore, it looks as though
the anticodon stem-loop evolved _after_ the acceptor-TUC minihelix,
which is probably the older part of the tRNA and originally had
nothing at all to do with "coding" (Schimmel and Ribas de Pouplana
(1995), Cell 81: 983-86). In that case, I think the burden of proof
is on anyone who claims that the assignment of amino acids to codons
is anything more than a "frozen accident". Even if one of the many
(contradictory) schemes for rationally associating amino acids with
base triplets should turn out to be true, that again would involve a
_chemical_ interaction, inacessible to understanding by "coding
Finally, the attempt to think about "coding" in purely abstract terms
should be deja vu to anyone who knows the history of molecular biology
(and Yockey must know this at first hand): remember the wonderful
theoretical constructs like Crick's "comma-free" code, that turned out
to be entirely unhelpful because nature just wasn't as clever as
Francis Crick? The moral is that too high a level of abstraction
produces thinking that just doesn't connect with the everyday reality
of experimental biology.
Also sprach Julius Caesar: "Yon Cassius hath a lean and hungry look. He
thinks too much. Such men are dangerous."
Let us therefor not think too much!
Also sprach Hubert Yockey:
In the first place Francis Crick's "comma free" code was not his finest
hour nor were his codes clever. Codes do not have commas. The moral is
that denisons of the groves of academe are not on speaking terms with each
other. Mathematicians and communication engineers had developed coding
theory and practice to a high state of sophistication. Why were they not
called upon? What do professors talk about at the faculty club?
The "Frozen accident" proposal for the origin of the genetic code is a
cop-out that explains nothing. How did it happen that eight amino acids
are effectively coded by a doublet code?
The stereochemical proposal fell on its face when it was discovered that
the standard genetic code was not universal and that there were a number
of non-standard genetic codes. It was a nice try, however. So much for
Steve's remarks about the central role of chemistry in the development of
To save Steve's valuable time here are the high points of Chapter 7:
1) Page 183 There are 1.40 x10^70 ways to arrange the present complexion
of the genetic code into blocks of six, four, three, two and one. It takes
a great deal of faith to believe that the modern genetic standard and
non-standard codes were found among these codes in a matter of ten million
years Lazcano and Miller J. Mol. Evol 39: 546-554 (1994) Say was
available. Furthermore, there are enormously more codes that are not
arranged in the same complexion, six, four, three, two and one as the
modern genetic code.
2) Let us follow Thomas Jukes' suggestion that the early genetic code was
doublet and had a vocabulary of fewer than 16 amino acids, say 8 amino
acids. They might have been the 8 amino acids that today are effectively
coded by a doublet code.
3) Let us suppose that the chemistry of the doublet code was essentially
the same as today and was based on a triplet in which the third
nucleotide was silent.
4) There are only 12870 doublet codes for 8 amino acids. Suppose we
consider the doublet codes that have 28 possible mutations to non-sense
doublet codons. That is mutations to codons that did not have an amino
acid assignment and therefore constituted noise. There are 3792 such
doublet codes. Table 7.1 shows Markov states through which the doublet
codes must evolve as the vocabulary is augmented.
5) The doublet codes follow an evolutionary path that leads to a
bottleneck when the vocabulary contains14 or 15 amino acids with 48 to 16
6) As the code emerges from the bottleneck, the vocabulary can only be
augmented by adding specificity to the third nucleotide.
7) This proposal predicts that more one surviving doublet code will
emerge from the bottleneck to become triplet codes. This will be
especially evident in the assignment of specificity to the third
All the numbers given above were calculated from standard means of
calculation of the number of ways several things can be arranged in
specified ways. Now you must go to the discussion in Chapter 7 for my
Wovon man nicht berechen kann, darueber man kann nicht begreifen.
Adapted from Ludwig Wittgenstein.
Best regards, Hubert