Stepping Up Enhanced Rate Calculations with EATR-Flooding

Area of Science:

• Computational Chemistry
• Biophysics
• Molecular Dynamics

Background:

• Calculating rate constants for slow biomolecular processes is computationally challenging.
• Existing methods use time-dependent bias potentials, limiting their applicability.
• Previous exponential average time-dependent rate (EATR) method required time-varying bias.

Purpose of the Study:

• To generalize the EATR method for broader applicability in biomolecular simulations.
• To introduce a method that does not require a time-dependent bias potential.
• To improve the accuracy and efficiency of rate constant calculations using enhanced sampling techniques.

Main Methods:

• Developed the EATR-flooding approach, varying bias potential strength across simulation sets instead of time.
• Implemented EATR-flooding in an open-source Python library.
• Validated the method on coarse-grained protein and atomistic cavity-ligand systems.

Main Results:

• EATR-flooding accurately estimates rate constants without significant loss of efficiency compared to standard EATR.
• The method includes an internal check for overbiasing.
• A single, consistent gamma parameter is predicted for given collective variables (CVs).

Conclusions:

• EATR-flooding extends the utility of EATR to quasi-static biasing schemes like flooding.
• The approach is broadly applicable to various collective variable biasing enhanced sampling methods.
• This generalization enhances the accuracy and reliability of biomolecular simulation rate calculations.