RNA secondary structure prediction with coincidence algorithm

Abstract

The main function of RNA is to translate genetic information of DNA into proteins. Understanding structure of RNA helps understanding the mechanism of actions within the cell. It is widely believed that the functions of bimolecular are dictated by its structure. This paper employs a RNA primary structure as an input and predicts its secondary structure by finding all possible helices in a RNA sequence. The problem is modelled as a combinatorial optimization problem. The objective is to find the combination of helices that forms the RNA secondary structure that has minimum free energy calculated via Individual Nearest Neighbor Hydrogen Bond model (INN-HB). This paper proposes an effective evolutionary algorithm, Coincidence Algorithm, to solve this problem. The ten known RNA sequences, a variety of lengths and various different organisms, are used to test the proposed method in terms of specificity, sensitivity and F-measure. The results are compared with others heuristic methods: RnaPredict and SARNA-Predict. The experimental results show that, on average, the proposed method outperforms RnaPredict and SARNA-Predict in all measures.