RNAsnoop: efficient target prediction for H/ACA snoRNAs

Bioinformatics. 2010 Mar 1;26(5):610-6. doi: 10.1093/bioinformatics/btp680. Epub 2009 Dec 16.

Abstract

Motivation: Small nucleolar RNAs are an abundant class of non-coding RNAs that guide chemical modifications of rRNAs, snRNAs and some mRNAs. In the case of many 'orphan' snoRNAs, the targeted nucleotides remain unknown, however. The box H/ACA subclass determines uridine residues that are to be converted into pseudouridines via specific complementary binding in a well-defined secondary structure configuration that is outside the scope of common RNA (co-)folding algorithms.

Results: RNAsnoop implements a dynamic programming algorithm that computes thermodynamically optimal H/ACA-RNA interactions in an efficient scanning variant. Complemented by an support vector machine (SVM)-based machine learning approach to distinguish true binding sites from spurious solutions and a system to evaluate comparative information, it presents an efficient and reliable tool for the prediction of H/ACA snoRNA target sites. We apply RNAsnoop to identify the snoRNAs that are responsible for several of the remaining 'orphan' pseudouridine modifications in human rRNAs, and we assign a target to one of the five orphan H/ACA snoRNAs in Drosophila.

Availability: The C source code of RNAsnoop is freely available at http://www.tbi.univie.ac.at/ -htafer/RNAsnoop

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Binding Sites
  • Genomics / methods*
  • Molecular Sequence Data
  • Nucleic Acid Conformation
  • RNA, Ribosomal / chemistry
  • RNA, Small Nucleolar / chemistry*
  • Sequence Analysis, RNA
  • Software*

Substances

  • RNA, Ribosomal
  • RNA, Small Nucleolar