Ligand affinities predicted with the MM/PBSA method: dependence on the simulation method and the force field

Area of Science:

• Computational chemistry
• Biophysics

Background:

• Avidin-biotin interactions are crucial in biological systems and drug development.
• Accurate calculation of binding free energies is essential for predicting molecular interactions.

Purpose of the Study:

• To evaluate the impact of force fields and geometry generation methods on MM/PBSA calculations for avidin-biotin analogues.
• To determine optimal simulation parameters for accurate binding free energy predictions.

Main Methods:

• Molecular mechanics Poisson-Boltzmann surface area (MM/PBSA) calculations.
• Comparison of four different molecular mechanics force fields.
• Analysis of geometry generation techniques and simulation system sizes.
• Inclusion of explicit water molecules in molecular dynamics simulations.

Main Results:

• No significant differences in binding free energies were observed across the four tested force fields.
• Mixing force fields between geometry generation and energy calculation steps is not recommended.
• Explicit water molecules are necessary for accurate simulations.
• Nonperiodic simulations with a fixed protein and a small system size (18 Å) are effective.
• Mean absolute error of 9-19 kJ/mol and standard error of 5-15 kJ/mol were achieved, primarily due to entropic contributions.

Conclusions:

• The MM/PBSA method can reliably predict binding free energies for avidin-biotin analogues with careful parameter selection.
• Simulation system size and boundary conditions have a minor impact compared to the use of explicit water.
• Further refinement is needed to address entropic contributions for improved accuracy.