DNA Coding Research Based On Improved IWO Algorithm

With the development of electronic technology, electronic computer has become increasingly unable to meet human needs, at the same time, biological technology is rapidly developing. DNA computing which is the new computation model emerges as the times require. DNA molecules and some related biological enzymes are the basic materials of the DNA computing, and biochemical reactions are important means to achieve the calculation. Its appearance has caused the extensive attention of scholars all over the world. Many researchers have used its advantages of high degree of parallelism, mass storage, low consumption of energy and rich resource to solve the NP hard problems successfully, such as the Hamilton circuit problem, the maximum clique problem.DNA coding plays an important role in DNA computing. The coding quality and quantity directly affect the calculation accuracy and efficiency. In order to make the DNA computing more reliable, a lot of research work is focused on DNA coding. The paper focuses on improving the coding quality to avoid the occurrence of nonspecific hybridization in the process of calculation. Main work is as follow:(1) A detailed analysis of the factors affecting the DNA coding is made by referring to the relevant references, and constraints which are mainly considered in the existing research are summarized: continuity, hairpin, similarity, H-measure, GC content and Tm. A multi-objective evaluation system based on the above constraints is established to evaluate the quality of DNA coding, and the concrete function realization is given.(2) IWO algorithm based on niche crowding for optimization design of DNA coding sequence(NCIWO) is put forward. In the paper, the invasive weed optimization is adopted and the niche crowding mechanism is applied into the algorithm to solve the multi-objective optimization problem. Meanwhile, in the spatial diffusion stage, offspring are generated by Cauchy distribution instead of Gauss distribution of the traditional IWO algorithm. The specific implementation of NCIWO is designed, and the results verify the feasibility and effectiveness of the algorithm.(3) A multi-objective IWO intelligence approach for solving DNA coding optimization(MA_IWO) is proposed. The non-dominated sorting and the invasive weed optimization algorithm are combined to form a multi-objective IWO algorithm. Moreover, adaptive mechanism is introduced into the reproduction phase of the IWO algorithm, so that the standard deviation of each individual in each generation can be changed adaptively according to the fitness value. The concrete realization of MA_IWO is designed, and the qualities of the generated DNA sequences are significantly superior to the other previously published results.