Voyages in Protein Space

Principal Investigators
Peter Stadler
Ivo Hofacker
Co-workers:
Aderonke Babajide, Günther Weberndorfer

Support:


Abstract

Knowledge-Based potentials can be used to decide whether an amino acid sequence is likely to fold into a prescribed native protein structure. We propose to use this idea to survey the sequence-structure relations in protein space. While the problem of predicting protein structures from sequence remains largely unsolved, knowledge-base potentials can help solve the inverse problem of designing sequences that will adopt a predefined fold. Using inverse folding techniques we plan to study questions, such as:
  • Can the same fold be found for unrelated amino acid sequences?
  • Can we find different protein folds for very similar sequences?
  • Can biologically relevant structures be found all over sequence space?
  • Is it possible to build proteins from a small subset of amino acis?
  • How large a portion of sequence space has to be searched to find a desired structure?
    • We will employ several different potential functions to ensure that our results are generic. To further test the accuracy of knowledge-based potentials we plan to design a novel potential based on cycle-decomposition, which can be compared directly with the measured energy parameters for RNA secondary structures. Finally, we will perform simulations of protein evolution using our knowledge-based potentials to provide realistic fitness models on which we can observe the interplay between adaptation and neutral evolution.

    Peter F. Stadler <studla@tbi.univie.ac.at>
    Last modified: Fri Mar 5 12:25:58 CET 1999