Identification of Conserved Secondary Structures in RNA Virus Genomes

Principal Investigator
Peter Stadler
Co-Investigator:
Ivo Hofacker

Co-workers:
Susanne Rauscher, Christof Flamm, Martin Fekete, Roman Stocsits

Support:


Fonds zur Förderung der Wissenschatlichen Forschung, Proj.No. P12591-INF
Begin 1.1.1998 Ende 31.12.1999

Abstract

The task of three-dimensional structure prediction for biopolymers like RNAs and Proteins is much too difficult to be solved with current knowledge and methods, however there are promising lines of research. A simpler problem, namely the prediction of secondary structure, is tractable even for large molecules. Here, the prediction of functional secondary structures is of special interest. Functional secondary structures represent a qualitatively important description of the molecules, as documented by their extensive use for the interpretation of molecular evolution data. However, almost all RNA molecules form secondary structures. The presence of secondary structure in itself hence does not imply any functional significance. We propose to start our investigation with the genomic RNAs of Picornaviridae and Flaviviridae. RNA viruses are an ideal proving ground for developing novel approaches towards functional genome analysis for a variety of reasons:

  • Distant groups of RNA viruses have very little or no detectable sequence homology. Thus we can test our methods on essentially indenpendent data sets.
  • RNA viruses show an extremely high mutation rate. Due to this high mutation rate they form quasispecies A substantial fraction of the mutations are likely to be almost neutral, implying a very rapid evolution at sequence level.
  • he untranslated regions of RNA genoms are most likely functionally important since the high selection pressure acting on viral replication makes ``junk RNA'' very unlikely.
    • We search for functionally important structures in viral genomes and viral messenger RNAs using (i) thermodynamic folding procedures for structure prediction and (ii) structure based alignments for identifying those structural features that are conserved among a group of closely related viruses. The actual existence of such features is then verified by the presence of compensatory mutations.

    Thermodynamic folding algorithms are collected in the Vienna RNA Package. The large size of viral genomes requires the use of parallel computers. Implementations of folding programs for message passing systems was pursued in a collaboration with Paul Stolorz at the Artificial Intelligence Groups in Section 395 of JPL, NASA.

    We are also collaborating with Andreas Wagner, Department of Biology,University of New Mexico and the Albuquerque High Performance Computer Center on this subject.