A computational method for the similarity analysis of RNA secondary structures and its application

In this contribution, we present a novel method for analyzing the similarity of RNA secondary structures. This method is based on stochastic process model: Each RNA secondary structure is mapped into stochastic process model and described by a transition probability matrix, which characterizes the RNA secondary structure numerically by incorporating the information on the primary structure (the base and the order of the bases) and base pairing. This method avoids dealing with complex RNA secondary structures. Its application to the similarity analysis of a set of mixed secondary structures from 5S rRNAs, 16S rRNAs and RNase P RNAs has been demonstrated at the end of our paper. The result indicates that our method can analyze the similarity relationship reasonably, especially fast as a result of the fact that the whole process requires no complicated algorithms and is easy to operate.