DNA Base-Code Generation for Bio-molecular Computing by Using a Multiobjective Approach Based on SPEA2

The design of DNA strands suitable for bio-molecular computing involves several complex constraints which have to be fulfilled to ensure the reliability of operations. Two of the most important properties which have to be controlled to obtain reliable sequences are self-assembly and selfcomplementary hybridizations. These processes have to be restricted to avoid undesirable interactions which could produce incorrect computations. Our study is focused on six different design criteria that provide reliable and robust DNA sequences. We have tackled the problem as a multiobjective optimization problem in which there is not only an optimal solution, but a Pareto set of solutions. In this paper, we have used the Strength Pareto Evolutionary Algorithm 2 (SPEA2) to generate reliable DNA sequences for three different real datasets used in bio-molecular computation. Results indicate that our approach obtains satisfactory DNA libraries that are more reliable than other results previously published in the literature.