Markov State Models to Elucidate Ligand Binding Mechanism.

Area of Science:

• Computational chemistry
• Biophysics
• Molecular modeling

Background:

• Microsecond molecular dynamics simulations now allow direct observation of ligand association.
• Markov State Models (MSMs) are a powerful framework for analyzing simulation data to understand biomolecular mechanisms.
• Accurate modeling of ligand binding kinetics and pathways requires careful construction of MSMs.

Purpose of the Study:

• To describe methods and best practices for constructing MSMs of ligand binding from unbiased simulation data.
• To demonstrate the utility of time-lagged independent component analysis (tICA) for building informative MSMs.
• To provide examples of MSMs applied to ligand binding events.

Main Methods:

• Utilizing microsecond molecular dynamics simulations.
• Constructing Markov State Models (MSMs) from simulation trajectories.
• Applying time-lagged independent component analysis (tICA) for model building.
• Estimating ligand association rates from MSMs.

Main Results:

• Developed and validated methods for constructing accurate MSMs of ligand binding.
• Demonstrated the application of tICA in building informative MSMs.
• Provided examples of MSMs elucidating binding mechanisms of phenylalanine to phenylalanine hydroxylase and peptides to MDM2.

Conclusions:

• MSMs are crucial for analyzing microsecond molecular dynamics data to understand ligand binding mechanisms and kinetics.
• Careful construction and analysis of MSMs, aided by tICA, are essential for reliable insights.
• MSMs offer a versatile approach to study diverse biomolecular interactions, including protein-ligand and protein-peptide binding.