Global Langevin model of multidimensional biomolecular dynamics

Area of Science:

• Computational Biophysics
• Molecular Dynamics Simulations
• Statistical Mechanics

Background:

• Biomolecular dynamics are often simplified to motion on a free energy landscape.
• Zwanzig's system-bath theory provides a theoretical framework using Langevin equations for this.
• Extending this theory to realistic biomolecular systems requires practical methods.

Purpose of the Study:

• To develop a practical method for constructing an analytically defined global model of structural dynamics.
• To extend Zwanzig's system-bath formulation to data-based biomolecular systems.
• To create a model that can predict the effects of parameter changes and mutations.

Main Methods:

• Utilized molecular dynamics simulations and collective coordinates.
• Employed an empirical valence bond-type model to represent free energy landscapes and friction fields.
• Derived a memory-free Langevin equation describing system motion.

Main Results:

• The proposed Langevin model successfully reproduced results from all-atom simulations for alanine dipeptide and heptaalanine.
• The model satisfies the theoretical assumptions of Zwanzig's formulation, including delta-correlated Gaussian noise.
• The global model serves as an empirical realization of the theoretical framework.

Conclusions:

• The developed method provides a valid and practical approach to modeling biomolecular dynamics on free energy landscapes.
• This empirical Langevin model can be used for investigating system parameter dependencies.
• The model enables predictions regarding the impact of site-selective mutations on molecular dynamics.