Parts | Days | Selection | Search | Downloads | Help

MO: Molekülphysik

MO 31: Poster: Molecular Dynamics

MO 31.2: Poster

Tuesday, March 14, 2006, 16:30–18:30, Labsaal

Nonlinear time series analysis of peptide dynamics — •Rainer Hegger, Alexandros Altis, and Gerhard Stock — J.W. Goethe Universität, Institut f. Physikalische und Theoretische Chemie, Marie-Curie Str. 11, 60439 Frankfurt

During the last few years the computer simulation of peptide dynamics has become feasible, by using an all-atom force field and an explicit or implicit representation of the solvent. Since the simulations produce a huge amount of data, it is of great importance to develop and improve suitable tools to analyze and interpret these data. As a new approach to this well-known problem, we suggest to apply the ideas originating from the theory of nonlinear dynamical systems to analyze molecular dynamics data. The goal is to develop mathematical models which, on one hand, are simple enough to be analyzed in great detail, and on the other hand, are realistic enough to reproduce the essential features of the dynamics.

In this work we present results obtained for small peptides (<=10 amino acids). By performing a PCA rotation in the space spanned by the dihedral angles of the peptide, we get rid of the dominant linear correlations between the peptide degrees of freedom. In the new coordinate system we are able to reveal the leading nonlinear properties.

By means of an analysis of these nonlinear properties we are able to describe the chaoticity of the system as a function of molecular properties, such as the length, the sequence and the types of residues of the peptide and thereby we obtain novel insight into the complexity and cooperativity of peptide folding.