An improved algorithm for solving helix generation of RNA secondary structure prediction

This paper presents an efficient O(n²) time algorithm for solving the helix generation problem of RNA to predict the predict the secondary structure of that RNA. It encodes the RNA secondary structures as an integer permutation of helices. The helices are pre-computed by the helix generation algorithm and each integer corresponds to a candidate helix. In this paper, a helix is formed only when three or more adjacent base pairs are formed and the loop connecting the helix must be at least three nucleotides in length. From this algorithm we find all possible helices that can form in a structure. After that we predict secondary structure of RNA by SARNA-Predict based on Simulated Annealing (SA). SARNA-Predict use a permutation-based representation to the RNA secondary structure and percentage swap translocating mutation operator to find a solution with a lower free energy [1]. Calculating the minimum free energy, we find the stable secondary structure of the RNA.