The genetic code: does what it says on the tin

There’s a really interesting piece of work in the latest Genome Research. Here’s the abstract and a link to the preliminary PDF.

The authors have analysed the universal genetic code in several interesting ways, including permuting it to obtain all possible alternative codes. They conclude that the code we have is almost optimal for carrying “parallel codes” – that is, information other than protein-coding information such as binding sites and splice sites. The ability to do this is related to the identity of stop codons and the minimisation of translation errors that result from frame-shifts.

It may take a while to get my head around all of the statistical methods employed, but it looks like a good read.