Abstract: Vibrational spectroscopies at mid-infrared frequencies provide excellent probes to characterize functional groups and their immediate chemical environment. However, from a thermodynamic and dynamic point of view, only the ground state of these vibrations is significantly populated. Most of the “jiggling and wiggling” of atoms and molecules (referred to in the famous quote by Feynman) happens at lower frequencies in the far-infrared, where vibrations are easily excited by thermal collisions.

The Heyden research group develops methods to characterize and extract information from far-infrared vibrations in molecular dynamics simulations of biomolecular systems: 1) We develop methods that eliminate the need for harmonic and quasi-harmonic approximations in the analysis of collective vibrational modes and their contributions to vibrational spectra. 2) We use intermolecular vibrations of the water-hydrogen bond network to generate detailed 3D maps of protein hydration free energies and water-mediated interactions. 3) We use such 3D maps to generate implicit solvation models that enable realistic simulations of large biomolecular systems including many interacting proteins.

Each of these methods has its own applications, but combined they allow us to develop computationally efficient models of complex biomolecular environments such as the cytoplasm. We developed a multi-conformation Monte Carlo algorithm that uses input from existing molecular dynamics simulation trajectories to generate new simulation models of systems containing 100’s of interacting flexible proteins at very low computational cost. This allows us to analyze consequences of biomolecular crowding in a multitude of scenarios, which remain inaccessible to direct molecular dynamics simulations.