Andrzej Kolinski Research Group

Coarse-grained protein modeling

Modeling Software & Servers

Biomolecules — dynamics & interactions

0
-1
-2
-3

Application of reduced models to protein structure prediction

Source:

Theoretical and Computational Chemistry: Computational Molecular Biology, Elsevier , Amsterdam, 8:397–440, 1999

Abstract

This chapter describes the state of the art of contemporary approaches to protein tertiary structure prediction, and focuses on reduced models. Any successful tertiary structure prediction algorithm must address two intertwined issues: first, it is required to have an energy or fitness function that distinguishes the native conformation from the sea of alternative structures. Second, it must have a conformational search protocol that can find the native conformation among the possible alternative structures. A key issue that is faced when embarking on a program of protein structure prediction is deciding on the level of detail of protein representation. The advantage of a lattice is purely computational. Because the protein is confined to a set of grid points, many geometric and energetic properties can be precalculated in advance. Thus, a well-designed lattice model is about a factor of 10 to 100 times faster than the corresponding continuous space model.