Improving the design of sequences for DNA computing: A multiobjective evolutionary approach

• We propose a multiobjective evolutionary algorithm for DNA sequence design.
• The algorithm uses a matrix-based chromosome as encoding strategy.
• The algorithm uses specific recombination operators.
• We show that our approach may improve the performance for this problem compared to previous methods.

Designing oligonucleotide strands that selectively hybridize to reduce undesired reactions is a critical step for successful DNA computing. To accomplish this, DNA molecules must be restricted to a wide window of thermodynamical and logical conditions, which in turn facilitate and control the algorithmic processes implemented by chemical reactions. In this paper, we propose a multiobjective evolutionary algorithm for DNA sequence design that, unlike preceding evolutionary approaches, uses a matrix-based chromosome as encoding strategy. Computational results show that a matrix-based GA along with its specific genetic operators may improve the performance for DNA sequence optimization compared to previous methods.