Search

We present a fully atomistic Langevin dynamics approach as a method to simulate biopolymers under external forces. In the harmonic regime, this approach permits the computation of the long-term dynamics using only the eigenvalues and eigenvectors of the Hessian matrix of second derivatives. We apply this scheme to identify polymorphs of model proteins by their mechanical response fingerprint, and we relate the averaged dynamics of proteins to their biological functionality, with the ion channel gramicidin A, a phosphorylase, and neuropeptide Y as examples. In an environment akin to dilute solutions, even small proteins show relaxation times up to 50 ns. Atomically resolved Langevin dynamics computations have been performed for the stretched gramicidin A ion channel.

We address the problem of charge transfer (CT) between a nanosized inorganic system and a protein from a theoretical and numerical perspective. The CT process is described on an atomistic level by applying an electronic Hamiltonian that takes into account the chemical bond, vibronic coupling effects, and polarization degrees of freedom. As a structurally well-characterized example, we consider a complex of C60 and its antibody. For this system, we find a novel efficient protein CT mechanism; through-space superexchange is mediated by stacked pi orbital systems. The predicted rates are comparable to those obtained for short-range electron tunneling through covalent bonds, the fastest ground-state CT process known for proteins.