Simplicity in RNA Secondary Structure Alignment: Towards biologically plausible alignments

Abstract

Ribonucleic acid (RNA) molecules contain the genetic information that regulates the functions of organisms. Given two different molecules, a preserved function corresponds to a preserved secondary RNA structure. Hence, RNA secondary-structure comparison is essential in predicting the functions of a newly discovered molecule. In this paper, we discuss our SPRC method for RNA structure comparison. In this work, we developed, a novel tree representation of RNA that reflects both its primary and secondary structure and a tree-alignment algorithm, which, given the tree representations of two RNA molecules, produces a sequence of mutations that could transform one RNA molecule to the other. Our SPRC algorithm extends the Zhang-Shasha tree-edit distance calculation algorithm in two ways: first, in addition to the distance, it reports all editing sequences with the same minimum edit cost, and second, it uses a biologically-inspired affine cost function. Furthermore, the SPRC method proposes set of heuristics designed to filter the produced solution set to recommend the simplest editing sequence, as corresponding to the most biologically correct alignment. Experiments on three 5S rRNA families: archaea, eubacteria, and eukaryota, show that SPRC is very effective in producing biologically meaningful RNA secondary structure alignments.