Fast protein structure prediction using Monte Carlo simulations with modal moves

J Am Chem Soc. 2003 Nov 26;125(47):14244-5. doi: 10.1021/ja036647b.

Abstract

Using normal modes to generate torsion space moves in Monte Carlo simulations of peptides and proteins is not a new idea; nevertheless, despite its power it has not received widespread application. We show that such a "Modal Monte Carlo" approach is an efficient tool for ab initio predictions of small-protein structures. We apply this method to the Trp cage, a 20-residue polypeptide designed to fold rapidly into a structure that includes tertiary contacts, despite its short length. We achieve a high-quality ab initio structure prediction in about 2 orders of magnitude less computation time than state of the art molecular dynamics techniques.

MeSH terms

  • Computer Simulation
  • Models, Molecular*
  • Monte Carlo Method*
  • Nuclear Magnetic Resonance, Biomolecular
  • Peptides / chemistry*
  • Protein Folding
  • Recombinant Proteins / chemistry*

Substances

  • Peptides
  • Recombinant Proteins
  • TC5b protein, synthetic