- Conference date: 1991
- Location: Obernai (France)
Two computational approaches to study plausible conformations of biological molecules and the transitions between them are presented and discussed. The first approach is a new search algorithm which enhances the sampling of alternative conformers using a mean field approximation. It is argued and demonstrated that the mean field approximation has a small effect on the location of the minima. The method is a combination of the LES protocol (Locally Enhanced Sampling) and simulated annealing. The LES method was used in the past to study the diffusion pathways of ligands from buried active sites in myoglobin and leghemoglobin to the exterior of the protein. The present formulation of LES and its implementation in a Molecular Dynamics program is described. An application for side chain placement in a tetrapeptide is presented. The computational effort associated with conformational searches using LES grows only linearly with the number of degrees of freedom, whereas in the exact case the computational effort grows exponentially. Such saving is of course associated with a mean field approximation. The second branch of studies pertains to the calculation of reaction paths in large and flexible biological systems.
An extensive mapping of minima and barriers for two different tetrapeptides is calculated from the known minima and barriers of alanine tetrapeptide which we calculated recently.1 The tetrapeptides are useful models for the formation of secondary structure elements since they are the shortest possible polymers of this type which can still form a complete helical turn. The tetrapeptides are isobutyryl‐val(χ1=60)‐ala‐ala and isobutyryl‐val(χ1=−60)‐ala‐ala. Properties of the hundreds of minima and of the hundreds intervening barriers are discussed. Estimates for thermal transition times between the many conformers (and times to explore the complete phase space) are calculated and compared. It is suggested that the most significant effect of the side chain size is on the number of the direct paths between the minima. The influence on the distribution of the barriers and the minima energies is less significant. Calculation of reaction paths in large molecular systems requires new computational techniques. We employed our newly developed reaction path algorithm (SPW) for the study of the B to Z transition in DNA. The SPW (Self Penalty Walk) algorithm is explained in detail. A complex reaction coordinate (the B to Z transition in DNA) is calculated and analyzed. The calculated reaction path is for six basepairs DNA (including all 376 atoms). The path consists of 180° flips of two basepairs from the B DNA conformation to the Z DNA conformation.
Data & Media loading...
Article metrics loading...