Identifying the Most Probable Transition Path with Constant Advance Replicas.

Area of Science:

• Computational Chemistry
• Biophysics
• Molecular Dynamics

Background:

• Understanding biomolecular dynamics requires locating transition paths and sampling rare events.
• Current methods face challenges in accurately identifying these pathways.

Purpose of the Study:

• To develop a novel constraint-based constant advance replicas (CAR) formalism for identifying the most probable transition path (MPTP) between two states.
• To establish a robust and extensible platform for enhanced sampling in biomolecular simulations.

Main Methods:

• Derivation of temporal-integrated effective dynamics under holonomic CAR path constraints.
• Optimization of the MPTP by minimizing an upper bound of the CAR action functional using a variational expectation-maximization framework.
• Retrieval of thermodynamic and kinetic observables via molecular dynamics integration on the CAR MPTP with adaptive reflecting boundary conditions.

Main Results:

• Successfully identified the MPTP for the Müller potential, alanine dipeptide isomerization, and DNA base pairing transitions (Watson-Crick to Hoogsteen) in explicit solvent.
• Demonstrated the efficiency and accuracy of the CAR formalism in locating transition paths.
• Validated the method's ability to retrieve essential thermodynamic and kinetic information.

Conclusions:

• The CAR formalism provides a robust and extensible platform for identifying transition paths and enhancing sampling in biomolecular simulations.
• This method facilitates more flexible and reliable simulations of complex molecular events.
• The approach has broad applicability in computational biophysics and chemistry.