Philippe Gaborit; Oliver D. King
Theor. Comput. Sci., 2005, 334, 99-113
https://doi.org/10.1016/j.tcs.2004.11.004

Abstract

In this paper we translate in terms of coding theory constraints that are used in designing DNA codes for use in DNA computing or as bar-codes in chemical libraries. We propose new constructions for DNA codes satisfying either a reverse-complement constraint, a GC-content constraint, or both, that are derived from additive and linear codes over four-letter alphabets. We focus in particular on codes over , and we construct new DNA codes that are in many cases better (sometimes far better) than previously known codes. We provide updated tables up to length 20 that include these codes as well as new codes constructed using a combination of lexicographic techniques and stochastic search.