COMPUTING SCIENCE
Ode to the Code
Brian Hayes
The genetic code was cracked 40 years ago, and yet we still don't
fully understand it. We know enough to read individual messages,
translating from the language of nucleotide bases in DNA or RNA into
the language of amino acids in a protein molecule. The RNA language
is written in an alphabet of four letters (A, C, G, U), grouped into
words three letters long, called triplets or codons. Each of the 64
codons specifies one of 20 amino acids or else serves as a
punctuation mark signaling the end of a message. That's all there is
to the code. But a nagging question has never been put to rest: Why
this particular code, rather than some other? Given 64 codons and 20
amino acids plus a punctuation mark, there are 1083
possible genetic codes. What's so special about the one code
that—with a few minor variations—rules all life on
Planet Earth?
The canonical nonanswer to this question came from Francis Crick,
who argued that the code need not be special at all; it could be
nothing more than a "frozen accident." The assignment of
codons to amino acids might have been subject to reshuffling and
refinement in the earliest era of evolution, but further change
became impossible because the code was embedded so deeply in the
core machinery of life. A mutation that altered the codon table
would also alter the structure of every protein molecule, and thus
would almost surely be lethal. In other words, the genetic code is
the qwerty keyboard of biology—not necessarily the best
solution, but too deeply ingrained to be replaced or improved.
There has always been resistance to the frozen-accident theory. Who
wants to believe that the key to life is so arbitrary and ad
hoc? And there is evidence that the accident is not quite
frozen. Certain protozoa, bacteria and intracellular organelles
employ genetic codes slightly different from the standard one,
hinting that changes to codon assignments are not impossible after
all. And if the code is subject to change, then it must also be
subject to natural selection, which in turn suggests the possibility
of ongoing improvement. Perhaps ours is not the very best of all
possible codes, but after four billion years of evolution it ought
to be a pretty darn good one.
The urge to find something singular and superlative about the code
was already evident even before it was deciphered. For several years
before experiments began to reveal the true structure of the genetic
code, theorists were at liberty to dream up codes of their own. Some
of the proposals were so ingenious that the real code seemed a bit
disappointing. An earlier column in this series (January-February
1998) described that era of imaginary genetic engineering. But the
creative thinking did not end with the publication of the codon
table; indeed speculation seems to have been inhibited very little
by the constraints of mere fact. This sequel is meant to bring the
story up to date, covering both the biological mainstream and a few
ideas from wilder shores.
» Post Comment