An improved method to predict the entropy term with the MM/PBSA approach

Area of Science:

• Computational Chemistry
• Structural Biology
• Drug Discovery

Background:

• Estimating protein-ligand binding affinities is crucial for drug discovery.
• The MM/PBSA approach is widely used but can be limited by entropy calculations.
• Conventional MM/PBSA methods involve truncating proteins, leading to geometric instability.

Purpose of the Study:

• To develop an improved MM/PBSA method for more accurate protein-ligand binding affinity predictions.
• To reduce the standard deviation in entropy calculations within the MM/PBSA framework.
• To enhance the precision of MM/PBSA predictions by stabilizing the entropy term.

Main Methods:

• A novel MM/PBSA approach was developed.
• A fixed buffer region of approximately 4 Å was introduced around the truncated protein.
• Molecular dynamics minimization was performed with the buffer region held constant.

Main Results:

• The standard deviation of the entropy term was reduced by a factor of 2-4.
• The improved method demonstrated more stable results across various test cases.
• Predictions of relative binding affinities were often improved compared to the conventional method.

Conclusions:

• The new MM/PBSA method offers enhanced stability and accuracy for binding affinity calculations.
• This refinement addresses limitations in entropy estimation, improving MM/PBSA predictive power.
• The approach shows promise for more reliable virtual screening and drug design.