Life on earth follows certain rules of how to translate information stored in nucleic acids into the sequences of amino acids that make up most of the cellular machinery. The instruction how to translate the genetic information is called the genetic code. There are a few exceptions to the universal code but the standard genetic code applies to nearly all living organisms. The code consists of 64 codons, made up by units of 3 nucleic acids, which code for 22 amino acids or 3 stop signals. In the process of transcription the genetic information stored in the DNA is transferred into the sequence of mRNA which carries the message to the ribosome where the process of translation assembles amino acids into peptide/protein-sequences according to the instructions stored in the nucleic acids. In this process the codons on the mRNA bind amino acid-charged tRNA via their anticodons. Their has been a lot of debate on how the genetic code could have evolved out of nothing. Whilst the genetic code was deciphered mainly by Nirenberg, Khorana, Matthaei, Leder and colleagues, Francis Crick's proposal of the code as a "frozen accident" has been hugely influential. However, it seems that Wong's theory of coevolution of the genetic code whereby the availability of amino acids and their chemical properties shaped the genetic code as it is today is the most satisfactory explanation of how the code came to be the way it is today. Bresch and Hausmann pioneered the circular graph of the genetic code (Figure 1) and I decided to play sudoku with it late one night in the lab, after wondering if methionine's role as start amino acid in protein synthesis could be linked to the role of S-adenosylmethionine (AdoMet) in transmethylation (as well as transsulfuration and polyamine synthesis), trying to see if there was a way to get all the codons for leucine, serine, arginine and stop to cluster together. This was possible by placing the second codon base at the centre of the graph, followed by bases one and three, by changing the repetitive motif from UCAG to AGCU (or UCGA) and to allow for some deviation from this repetitive unit (Figure 2). Whilst the code may not be easy to read this way, it does highlight the interdependency of nucleic acids and amino acids in shaping the code with structurally/biosynthetically-related amino acids occupying related codons. It looks shockingly logical and makes it obvious that simple physical and chemical forces are sufficient to explain the evolution of the code. In addition, it may offer clues to the way in which the interaction between codon and anticodon takes place. Rather than being a simple 1-2-3 process, the tRNA may bind in a 2-1-2-3 way, simultaneously [123], or in a 1-2-2-3 fashion. Anyway, here is the remix (also published in the Journal of Theoretical Biology):