Research On The Algorithms Of Predicting The Secondary Structure Of DNA Based On Nearest-Neighbor Thermodynamics And Its Planar Embedding

Effective prediction of DNA secondary structure is an important research field in bioinformatics. Based on the structure stability of the secondary structure of DNA molecules and thermodynamic stability, an algorithm to predict the secondary structure of DNA is proposed. The proposed algorithm improves the Nussinov algorithm, and can satisfy the rapid largest base pair matching all of the secondary structure, and then the neighbor thermodynamic model is used to calculate the free energy of all structure, and free energy in the structure of threshold value range is the secondary structure of DNA molecules. With the development of visualization analysis in bioinformatics, more and more biological researchers will analysis DNA secondary structure drawing on the plane. ST-number algorithm is a common plane embedding algorithm. This algorithm can ensure that DNA secondary structure plan do not appear cross overlap, but the DNA secondary structure plan and the actual structure differs very big, not suitable for biological researchers analyzing. To solve this defect, iterative flat embedding algorithm is proposed in this paper. The proposed algorithm uses the characteristics of the adjacent die body with a base pairs to realize iterative recursion embedded in the die body and to make the floor plan similar to the secondary structure of DNA, but when the sequence of the DNA molecule is too long or fold, cross overlap parts is easy appear. So this article finally achieved the combination of the two algorithms. Firstly, ST-number algorithm is used to realize the plane embedded to get the initial plan. Secondly, iterative flat embedding algorithm is used to analyze the initial plan layout adjustment to solve the defect of the cross overlap, and is similar to the actual structure. The experimental results compared with RNAStructure software show that the prediction accuracy of the algorithm proposed in this paper has higher and coverage and analysis of the planar embedding of graph realization more standardized, easy biological researchers.