Exploring Multidimensional Free Energy Landscapes Using Time-Dependent Biases on Collective Variables

Area of Science:

• Computational chemistry
• Molecular dynamics simulations
• Biophysics

Background:

• Enhanced sampling methods are crucial for calculating free energy profiles in complex molecular systems.
• Adaptive Biasing Force (ABF) and metadynamics are powerful techniques for overcoming energy barriers in simulations.
• Direct comparisons between ABF and metadynamics on identical systems are scarce.

Purpose of the Study:

• To introduce a new, versatile implementation of the adaptive biasing force (ABF) method.
• To compare the performance of ABF and metadynamics for calculating multidimensional free energy profiles.
• To provide a unified computational framework for enhanced sampling methods.

Main Methods:

• A novel implementation of the adaptive biasing force (ABF) method supporting diverse collective variables.
• Application of ABF and metadynamics to conformational changes in model peptides.
• Simulation of halide ion translocation across a lipid membrane through a peptide nanotube using both methods.

Main Results:

• Both ABF and metadynamics achieve enhanced sampling and yield quantitatively consistent free energy profiles.
• The study demonstrates the applicability of the new ABF implementation to multidimensional free energy calculations.
• Similarities and differences between ABF and metadynamics are elucidated through numerical tests.

Conclusions:

• The choice between ABF and metadynamics depends on system-specific factors like dimensionality and barrier heights.
• The new ABF implementation offers a flexible tool for molecular free energy calculations.
• Both methods are effective for enhanced sampling, providing reliable free energy landscapes.