In this paper we revisit the idea of relating the genetic code to a Gray code. This idea was originally fostered by Rosemary Swanson as a “unifying concept for the amino acid code”. We show that cyclic arrangements of the codons which are different from the one proposed by Swanson, but which are still based on the principle of minimal change, yield a better grouping of the amino acids by similarity. To this end we model the problem of finding generalized Gray codes (codon arrangements) as a Traveling Salesman Problem. Solutions of this problem lead to several Gray code arrangements of the codons. We discuss the possible implications of the results from evolutionary and physico-chemical point of views.
Dragan Bosnacki, Huub M. M. ten Eikelder, Peter A.