Protein Structure and Function: Applications of Bioinformatics Methods - John Rigden 2014

Protein Dynamics: From Structure to Function
Molecular Dynamics Calculations

Markus B. Kubitzki, Bert L. de Groot, Daniel Seeliger

Understanding protein function requires detailed insight into protein dynamics, i.e., the various conformational states the system can explore. Despite significant advances in experimental techniques, computational Methods such as Structure/8.html">Molecular Dynamics (MD) simulations currently remain the only generally accepted tool to access dynamic information at atomic detail on time scales ranging from nanoseconds to microseconds. Even with state-of-the-art computing power, enhanced sampling methods beyond standard MD are necessary to improve the sampling of large Proteins and their assemblies. In this context, The Use of collective coordinates has emerged as a promising approach, either as a means of analysis or as novel sampling algorithms. This chapter reviews several enhanced sampling algorithms for biomolecular simulations, building upon the foundations of MD calculations. Numerous Examples are provided to demonstrate how examining the dynamic Properties of Proteins sheds light on their biological function.

Over the past decades, experimental techniques have achieved remarkable success in determining the three-dimensional structure of proteins, most notably X-ray crystallography, nuclear magnetic Resonance (NMR) spectroscopy, and cryo-Electron Cell/15.html">Microscopy. At the same time, moving beyond a static snapshot of a single Protein Structure has proven more challenging. Although various methods such as NMR relaxation, fluorescence spectroscopy, or time-resolved X-ray crystallography (Kempf and Loria 2003; Weiss 1999; Moffat 2003; Schotte et al. 2003) can provide insight into the internal conformational mobility of proteins, they have limitations. Despite this wide array of techniques, methods capable of resolving both spatial and temporal dynamics at the nanometer and nano- to microsecond scales are not universally accessible to the average researcher. Consequently, our picture of the conformational space accessible to a protein in vivo often remains incomplete. For instance, the exact transition pathways between different known Conformations usually remain elusive, even though they are frequently crucial for protein function. In such cases, computational methods offer a compelling opportunity to obtain atomic-resolution insights into protein dynamics on nanosecond-to-microsecond time scales. Among all approaches for simulating protein motion (Adcock and McCammon 2006), molecular dynamics (MD) simulations are by far the most widespread. Ever since the first MD simulation of a protein was reported over 30 years ago (McCammon et al. 1977), these calculations have become a standard tool in biomolecular research. Like all computational sciences, the field of MD simulations is in a state of continuous evolution, driven by exponentially increasing computer performance. This progress is further fueled by methodological advancements that have yielded a vast array of algorithms tailored to diverse Applications such as cellular transport, signal Transduction, allostery, Molecular recognition, molecular docking, atomic force microscopy modeling, and Enzymatic Catalysis.

Marcus B. Kubitzki, Bert L. de Groot, and Daniel Seeliger

Computational Biomolecular Dynamics Group,

Max Planck Institute for Biophysical Chemistry,

Am Fassberg 11, 37077, Goettingen, Germany

*e-mail:mkubitz@gwdg.de