A BPSON Algorithm Applied to DNA Codes Design

Abstract

DNA computing proposed by Adelman is new biotechnology which provides a new way to solve NP-hard problem. It has a promising future and successful results of various applications. DNA codes design is a significant step in DNA computing. Therefore, reliable DNA codes design not only can avoid non-specific hybridization between a code and its Watson-Crick complement but also can improve the efficiency of DNA computing. In this paper, a new algorithm is proposed to design reliable DNA codes. This algorithm combines the Bat algorithm and PSO algorithm. Fast nondominated sorting is used to assign a rank for codes. Thus, it is called BPSON for short. A bat algorithm is used to overcome PSO fall into the local optimal solution and enhance global search ability. In addition, for the purpose of verifying the effectiveness of our algorithm, the performance of BPSON is compared with the previous works. The experimental results show that codes obtained by our algorithm can avoid the appearance of secondary structure, which is beneficial to the specific hybridization among codes and has better thermodynamic properties. The results show that our algorithm can provide optimal codes for DNA computing.