Poster abstracts

Poster number 56 submitted by Anton Petrov

Constructing a library of RNA 3D motifs for motif structure prediction

Anton I. Petrov (Department of Biological Sciences, Bowling Green State University), Jesse Stombaugh (Department of Chemistry and Biochemistry, University of Colorado at Boulder), Craig L. Zirbel (Department of Mathematics and Statistics, Bowling Green State University), Neocles B. Leontis (Department of Chemistry, Bowling Green State University)

Abstract:
Structured RNA molecules fold hierarchically to form complex, protein-like 3D structures. Their function depends on forming the correct 3D structure. Unlike DNA, RNA molecules are single-stranded. The secondary structure of an RNA consists, therefore, of short double-helices that form when the RNA strand folds back on itself so that Watson-Crick complementary regions can base-pair. Once the sequence of an RNA is determined, programs such as Mfold (Zuker, 2003) are used to predict its secondary structure. Typically, about 2/3 of the bases of an RNA form Watson-Crick helices. The rest of the bases appear in secondary structures as unpaired “loops” – hairpin loops, which occur at the end of a helix, internal loops, between two helices, or junction loops, connecting more than two helices. However, X-ray crystallography and NMR spectroscopy have shown that most of these RNA “loops” are actually well-structured 3D motifs, consisting of one or more non-Watson-Crick basepairs (Leontis & Westhof, 2001) and responsible for important biological functions. Some of these 3D motifs mediate tertiary interactions that stabilize the 3D fold of the RNA. Other motifs bind proteins or small molecules, including drugs and ions. Others interact with other RNAs. Many of these “loop” motifs are recurrent – they are found in many different, unrelated RNAs, where they form very similar 3D structures and play similar functional roles. Thus, it is desirable to predict the 3D structures of hairpin, internal, and junction motifs based on their sequences. Different instances of the same 3D motif can differ in sequence and in number of nucleotides, due to substitutions, insertions or deletions of bases at specific positions. To predict instances of a motif from sequences, it is necessary to understand the possible sequence variations of each motif – this is what we call the sequence signature of the motif. To define the sequence signature of each motif, we search a reduced-redundancy set of atomic-resolution RNA structure files from the Protein Data Bank (PDB) using the “Find RNA 3D” (FR3D) software package (Sarver, Zirbel, Stombaugh, Mokdad, & Leontis, 2008). Motifs that share the same structural features as query motifs are stored in a library and are used to build probabilistic models based on stochastic context-free grammars.

References:
1. Leontis, N. B., & Westhof, E. (2001). Geometric nomenclature and classification of RNA base pairs. RNA, 7(4), 499-512.
2. Sarver, M., Zirbel, C. L., Stombaugh, J., Mokdad, A., & Leontis, N. B. (2008). FR3D: finding local and composite recurrent structural motifs in RNA 3D structures. J Math Biol, 56(1-2), 215-252.
3. Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res, 31(13), 3406-3415.

Keywords: RNA 3D Motifs, RNA structure, Find RNA 3D program