In this talk we present a recently developed concept of a binary dichotomic algorithm (in short BDA) for analyzing the genetic code. The genetic code is based on four nucleotide bases that are combined into triples called codons. BDAs  decompose the set of 64 possible codons into two equally sized subsets, called dichotomic classes.  The concept arose as a generalization of the classical dichotomies, e.g. the biologically important Rumer´s dichotomy. Dichotomic classes allow to uncover many symmetry properties of the genetic code and mirror operations that have a factual biological function in the decoding center of the ribosome.

 

Using the concept of BDAs we will show how to construct new BDA-generated models of the genetic code by applying successively a sequence of BDAs to the set of codons. We explore different biologically motivated possibilities to construct such models. For instance, it turns out that one can find models that provide a unique classification of the codons as well as models that almost distinguish amino acids. The hope is that the analysis of the genetic code with BDAs may help to understand better the known biological mechanisms or even to uncover new mechanisms, e.g. error correction principles. Finally, some open questions will be discussed.