Why nature chose A, C, G and U/T: an error-coding perspective of nucleotide alphabet composition.

The question of whether the size and make-up of the natural nucleotide alphabet is a consequence of selection pressure, or simply a frozen accident, is one of the fundamental questions of biology. Nucleotide replication is essentially an information transmission phenomenon, and so it seems reasonable to explore the issue from the perspective of theoretical computer science, and of error-coding theory in particular. In this analysis it is shown that the essential recognition features of nucleotides may be naturally expressed as 4-digit binary numbers, capturing the hydrogen acceptor/donor patterns (3-bits) and the purine/pyrimidine feature (1-bit). Optimal alphabets consist of nucleotides in which the purine/pyrimidine feature is related to the acceptor/donor pattern as a parity bit. Numerically interpreted, such alphabets correspond to parity check codes, simple but effective error-resistant structures. The natural alphabet appears to be an adaptation of one of two optimal solutions, constrained to its present size and composition by a combination of chemical and coding-theory factors.