Multiple free energies from a single simulation: extending enveloping distribution sampling to nonoverlapping...

Area of Science:

• Computational Chemistry
• Statistical Mechanics
• Molecular Dynamics

Background:

• Calculating free energy differences is crucial for understanding molecular systems.
• Existing methods can be computationally expensive, especially for multiple states or non-overlapping configurations.

Purpose of the Study:

• To extend the enveloping distribution sampling (EDS) method for free energy calculations.
• To enable accurate free energy determination from a single simulation, even when end-state configurations do not overlap.

Main Methods:

• Utilized a reference state Hamiltonian within a molecular dynamics framework.
• Expanded the enveloping distribution sampling (EDS) method to handle non-overlapping configuration space densities.
• Investigated the impact of parameter choices on accuracy and precision.

Main Results:

• Accurate free energy differences (deviations ≤ 0.6 kJ/mol) were obtained for various two-state and multiple-state test systems.
• The method successfully calculated free energy differences even for systems with non-overlapping end-state configurations.
• Convergence times were longer for systems with non-overlapping configurations compared to those with overlap.

Conclusions:

• The enhanced EDS method provides a robust approach for calculating multiple free energy differences from a single simulation.
• Accurate estimations are achievable regardless of the dissimilarity between end states.
• Parameter optimization is key for achieving high accuracy and precision.