Research On Prediction Algorithm For RNA Secondary Structures Based On Multimodal Multi-objective Genetic Algorithm

RNA plays an important role in biological cells,and the structure of RNA has a major impact on its function.Among them,the secondary structure of RNA is the central hub that carries the primary structure and tertiary structure.Predicting the secondary structure of RNA is an important way to understand the biochemical function of RNA.RNA secondary structure with pseudoknots is a complex secondary structure,which is more difficult to predict and has been proved to be an NP-complete problem.Pseudoknots are involved in many important life activities in biological cells,and more and more researchers pay attention to the research of RNA with pseudoknots.Therefore,it is of far-reaching significance to design a reasonable and efficient algorithm to predict RNA secondary structure with pseudoknots.In order to obtain the most stable RNA secondary structure,this paper proposes an RNA secondary structure prediction algorithm based on multi-objective genetic algorithm for the two optimization goals of the maximum complementary base pair and the minimum grouping.This algorithm makes the population iteratively evolve through the crossover and mutation operators in the genetic algorithm,and through non-dominated sorting and crowding distance sorting to ensure that the population continues to converge to the Pareto front,until the algorithm iterates to the maximum number of times to output the result set,and the final prediction result is selected by the size of free energy.In order to further improve the prediction accuracy,for RNA secondary structure,there is a multimodal solution set with close free energy but different complementary base matching,this paper proposes a multimodal multi-objective genetic algorithm to predict RNA secondary structure.The algorithm can effectively improve the diversity of the population,obtain multiple sets of equivalent multi-modal global optimal solutions,and prevent the algorithm from falling into the local optimal.This algorithm predicts the secondary structure of the RNA sequence through the dual optimization of the objective space and the decision space,and evaluates the pros and cons of the population through the dual scoring mechanism of the evaluation function of the decision space and the fitness function of the objective space,so as to select the better score individuals multiply until the termination conditions are met.The multi-modal multi-objective optimization method proposed in this paper can predict more possible secondary structures,thereby further improving the diversity and accuracy of the prediction results.Finally,the experimental results show that compared with the current state-of-the-art algorithms in the case of predicting RNA sequences,the algorithm in this paper has a significant improvement in the prediction accuracy of the existing algorithms.