A JAYA algorithm based on normal clouds for DNA sequence optimization

DNA computing is one of the more popular computational methods currently studied, but the requirements for nucleic acid molecules in DNA sequences are high, and it is an important challenge to design reasonable and high-quality DNA sequences while satisfying various constraints. Evolutionary algorithms have good applications in DNA sequence optimization problems, but they still have some limitations. To this end, this paper proposes a JAYA algorithm based on normal clouds, referred to as IJAYA, which uses a combinatorial learning approach to update the optimal and worst positions, which is used to manipulate the subsequent merit search means, and then enhances the local search ability of individuals through the normal cloud model, and finally rejects the worst solutions through a harmony search algorithm to find more reasonable and high-quality solutions. The validity of IJAYA is verified in six benchmark functions, in comparison with multiple variants of JAYA and two statistical tests. In the DNA sequence design optimization problem, the average DNA metrics optimized by IJAYA are: 0 (Continuity), 0 (Hairpin), 59.43 (H-measure), 46.57 (Similarity) and 63.79 (Similarity). The feasibility and practicality of IJAYA was verified by comparing it with the solution algorithms proposed in recent years and ablation experiments.