Computational Solutions to the Protein Folding Problem,
Abstract
The protein folding problem attempts to predict the native, or folded, state of a protein in three-dimensional space, given its primary sequence of amino acids. One common approach for a solution is to treat each complex amino acid as a single sphere, or united atom, and to model each peptide linkage between residues by a virtual bond between spheres. Computational efforts being examined rely on two major assumptions: for any specific molecular conformation, a corresponding potential energy function can be computed, and the three-dimensional, folded state corresponds to the global minimum of this energy function. The optimization method being used to minimize the potential energy involves collecting a large number of conformers, each attained by finding a local minimum of the potential energy function from a random starting point. The information from these conformers is then used to form a convex quadratic global underestimating function for the potential energy of the known conformers. The minimum of this underestimator is used to predict the global minimum for the function, allowing a localized conformer search to be performed based on the predicted minimum. The new set of conformers generated by the localized search can serve as the basis for another quadratic underestimation. After several repetitions, the global minimum can be found with reasonable assurance. The conformer which lies at the global minimum represents the three-dimensional folded state of the molecule. Protein folding problem, Global underestimator, Conformer, Mathematical models
Document Details
- Document Type
- Technical Report
- Publication Date
- May 19, 1994
- Accession Number
- ADA284911
Entities
People
- Vann H. Walke
Organizations
- United States Naval Academy