Predicting Binding Free Energy Change Caused by Point Mutations with Knowledge-Modified MM/PBSA Method.

Area of Science:

• Computational Biology
• Structural Bioinformatics
• Biophysics

Background:

• Predicting the impact of mutations on protein-protein binding free energy is crucial for understanding biological processes.
• Existing methods often face limitations in accuracy, speed, or the ability to integrate diverse data types.

Purpose of the Study:

• To develop and validate a novel computational methodology, Single Amino Acid Mutation based change in Binding free Energy (SAAMBE), for predicting mutation-induced changes in binding free energy.
• To leverage both sequence and structure-based information for enhanced prediction accuracy.

Main Methods:

• SAAMBE combines a Molecular Mechanics with the Poisson-Boltzmann and Surface Area (MM/PBSA) approach with statistical terms derived from physico-chemical properties of protein complexes.
• The method employs a rigid body approach, incorporating amino acid-specific dielectric constants to mimic the electrostatic effects of conformational changes.
• Utilizes 3D structures of protein-protein complexes as input.

Main Results:

• SAAMBE demonstrated significant improvement in prediction accuracy compared to experimental data for over 1300 mutations across 43 proteins.
• Achieved a high correlation coefficient of 0.624 against experimentally determined binding free energy changes.
• The computational approach is highly efficient, with an average prediction time of less than one minute per mutation, enabling large-scale calculations.

Conclusions:

• SAAMBE provides a robust and efficient method for predicting the effects of single amino acid mutations on protein-protein binding free energy.
• The integration of structural information and physico-chemical properties offers a powerful approach for computational drug design and protein engineering.
• The method's speed and accuracy make it suitable for high-throughput screening and analysis of mutation impacts.