Rewiring the keyboard: evolvability of the genetic code

Abstract

There are many variations in the genetic code, both in nuclear and mitochondrial genomes, but all are relatively recent modifications of the 'canonical' code found in the last common ancestor. The recent evolution of the genetic code relies on changes in components of the translation apparatus. In particular, many changes are due to base modifications that occur after transcription. Three theories have attempted to explain the range of changes to the genetic code: 1. Codon capture: biased mutations can eliminate certain codons from the genome, which are reassigned by neutral processes (that is, in the absence of selection). 2. Genome minimization: the code evolves to minimize the number of tRNAs required for translation. 3. Codon ambiguity: evolution of the genetic code is adaptive and occurs through an intermediate stage in which a codon can be translated into more than one amino acid. We find statistical support for the last two of these theories. Changes in the genetic code that are revealed by future studies will be used to test the specific predictions of each of these theories. Future work will also further clarify the roles of mutation and selection in producing new genetic codes.