Application of Biased Metropolis Algorithms: From protons to proteins

Area of Science:

• Computational physics
• Statistical mechanics
• Biomolecular simulations

Background:

• The heatbath algorithm is efficient for many sampling tasks.
• However, its applicability is limited when an efficient algorithm does not exist.
• Biased sampling methods offer potential alternatives.

Purpose of the Study:

• To establish the equivalence between biased Metropolis sampling and the heatbath algorithm.
• To demonstrate the utility of the biased Metropolis method in scenarios where heatbath is not feasible.
• To introduce and illustrate the Rugged Metropolis method for exploring complex energy landscapes.

Main Methods:

• Biased Metropolis sampling scheme
• Heatbath algorithm comparison
• Lattice gauge theory simulations
• Rugged Metropolis method application

Main Results:

• Biased Metropolis sampling is shown to be equivalent to the heatbath algorithm.
• The biased Metropolis method is successfully applied to lattice gauge theory simulations.
• The Rugged Metropolis method is demonstrated for locating configurations in rugged free energy landscapes.

Conclusions:

• The biased Metropolis method provides a viable alternative to the heatbath algorithm, particularly when efficient heatbath implementations are unavailable.
• The Rugged Metropolis method offers a novel approach for simulations in complex systems like biomolecules, aiding in the discovery of most likely configurations.