A Method of RNA Secondary Structure Prediction Based on Hidden Markov Model

The effective prediction of RNA secondary structure is an important research field of bioinformatics. We propose a new method based on hidden Markov model to predict the RNA secondary structure. We appliy the matching algorithm of prefix and suffix to find all the possible (including the pseudo-knot) stem zones quickly, establish the RNA-HMM, find the optimal method of the combination of stem zones, and obtain the RNA secondary structure including the pseudo-knot. The experiment results show that this method can reduce the computational complexity and improve the specificity and sensitivity of prediction with high accuracy, and can also predict the pseudo-knot structure.