| The structure of the genetic code has puzzled researchers since codon assignments were first elucidated, but recent genome and aptamer sequences allow quantitative testing of hypotheses about the code's origin, evolution, and subsequent diversification. My major findings can briefly be categorized by stage of evolution: (1) There is strong statistical support for two major theories about the evolution of the canonical code. The code is highly resistant to errors relative to random codes, and there is a strong statistical association between at least some codons and the RNA binding sites for their cognate amino acids. I found no support for specific coevolutionary models, although the code presumably evolved from a simpler form. (2) The exact time of code optimization remains unclear, but measures of amino acids that would have been more important early in evolution generally make the code appear more optimal. Thus codon assignments were perhaps determined early, in an RNA world. Aptamer data supports this idea, though equivocally. (3) Modern variant codes are probably recently-derived, nearly neutral mutants of the standard code. They may be adapted to the specific mitochondrial environments where most of them occur, although I did not find any supporting evidence. There is no support for the idea that variant codes are optimized for reduced tRNA number. There is strong support for the idea that code evolution proceeds through an ambiguous intermediate. The molecular basis for certain code changes can be elucidated: in ciliates, the mutations in the release factor eRF1 that prevent stop codon recognition can be recaptured by statistical techniques applied to phylogenies. (4) The genetic code is not just the pattern of codon assignments: it is also the frequency with which each codon is used. Surprisingly, the vast interspecific variation in codon and amino acid usage can largely be recaptured by a neutral model that assumes only purifying selection.; Despite these advances, many fundamental questions---such as the order in which amino acids were added to the code, and the extent to which stereochemistry and natural selection contributed to modern codon assignments---are still to be resolved. |
