RNA secondary structure prediction using context-sensitive hidden Markov models

It has been believed for decades, that proteins are responsible for most of the genetically important functions in all cells. Due to this reason, most of the research in molecular biology was focused on identifying genes that encode proteins, and their roles in the genetic network. Recent studies indicate that noncoding RNAs play important roles in various processes. Such ncRNA genes cannot be effectively identified using traditional gene-finders that aim at protein-coding genes. Many ncRNAs conserve their secondary structures as well as their primary sequences, which have to be taken into account when looking for ncRNA genes. In this paper, we propose a new method based on context-sensitive HMMs, which can be used for predicting RNA secondary structure. It is demonstrated that the proposed model can predict the secondary structure very accurately, at a low computational cost.