Research On RNA Secondary Structure Prediction Algorithm Including Pseudoknots

In recent years,with the deepening of RNA molecular research,it is found that RNA is not only the information carrier between DNA and protein,but also plays an important role in other biochemical processes such as catalysis,immunity and development.Gradually,RNA obtained the same important position with DNA and protein,has become a hot research.As the function of RNA molecules is closely related to its structure,the study of RNA structure is also increasingly important.The use of biological experiments such as X-ray and nuclear magnetic resonance(NMR)to obtain RNA structure,not only time-consuming and labor intensive,and high research costs.Furthermore,the use of computers through bioinformatics means to solve the structural prediction problem.RNA secondary structure prediction methods have been rapid development,there have been many structural prediction methods,softwares,energy models and structure databases.Pseudoknot is a special structural motif,although not a real topology,but it can intersect with many chains and form a complex 3D structure that makes the RNA structure diverse.The Pseudoknot is related to a series of different biological processes,including catalysis,ribosome mapping and regulatory translation.The prediction of the pseudoknot is the difficulty and focus of the current structure prediction.At present,the RNA secondary structure prediction algorithm with pseudoknot can be divided into three main categories: the first is the comparative sequence analysis,and the large number of homologous sequences are needed;the second is the dynamic programming algorithm based on the minimum free energy,and the structure with the optimal energy is obtained;the third is the heuristic method,the use of local optimal solution instead of the global optimal solution.These methods contain different theories and use different methods to achieve,but these algorithms are still inadequate,to be improved.At this stage,the prediction of the pseudoknot has not yet reached the accuracy of application,but also need to continue in-depth study,with more molecular dynamics,thermodynamics and other factors,in order to more accurately simulate RNA fold and improve the structure prediction accuracy.In this paper,a new method of secondary structure prediction based on RNA folding simulation model is proposed.Combining with the current thermodynamics and dynamics theory,we can carry out the structure prediction method without pseudoknot and with pseudoknot.In this method,the process of RNA folding from a random coil to a complete structure is phased,and the formation of the pseudoknot is also formed by the local motif pair with other motifs.We propose a maximum free energy difference method,select the most urgent helical region in all possible helical regions,finish the structure prediction without pseudoknot.At this time,the energy model used for the nearest neighbor model selected thermodynamic parameters by Turner 1999.Based on the structure of the above method,the single-chain structure of the hairpin loop is paired with other single-chain bases in the structure,and the helical region is adjusted in part to form the set of candidate pseudo-helical regions,calculate the free energy and use the maximum free energy difference method to find the most urgent one.Ultimately,complete the secondary structure prediction with pseudoknot.On the basis of this,the D & P energy model is used to extend the partial R & E model,and the pseudoknot is predicted on the structure without pseudoknot and merged into the existing structure to form the final secondary structure.The method is characterized by narrowing the candidate spiral region search domain,considering using the local optimal solution instead of the global optimal solution.The method can be divided into two parts: without pseudoknot and including pseudoknot,and use two different test sets,testing the algorithm,Med Fold's accuracy has improved than other RNA secondary structure prediction methods in the prediction.