Computational Identification of Three-Way Junctions in Folded RNAs: A Case Study in Arabidopsis

摘要

Three-way junctions in folded RNAs have been investigated both experimentally and computationally. Theninterest in their analysis stems from the fact that they have significantly been found to possess a functional role. In recentnwork, three-way junctions have been categorized into families depending on the relative lengths of the segments linking thenthree helices. Here, based on ideas originating from computational geometry, an algorithm is proposed for detecting three-waynjunctions in data sets of genes that are related to a metabolic pathway of interest. In its current implementation, the algorithmnrelies on a moving window that performs energy minimization folding predictions, and is demonstrated on a set of genes thatnare involved in purine metabolism in plants. The pattern matching algorithm can be extended to other organisms and othernmetabolic cycles of interest in which three-way junctions have been or will be discovered to play an important role. In the testncase presented here with, the computational prediction of a three-way junction in Arabidopsis that was speculated to have anninteresting functional role is verified experimentally.