Recent Developments in Linear Interaction Energy Based Binding Free Energy Calculations.

Area of Science:

• Computational chemistry
• Molecular modeling
• Biophysics

Background:

• The linear interaction energy (LIE) approach is an efficient end-point method for calculating binding affinities.
• Accurate calculation of protein-ligand binding free energy (ΔG) is crucial in drug discovery.
• Flexible proteins, like Cytochrome P450s, present computational challenges for ΔG prediction.

Purpose of the Study:

• To summarize recent advancements in using molecular simulation and LIE models for accurate and efficient ΔG calculations.
• To address challenges in computing ΔG for flexible proteins.
• To review strategies for defining applicability domains (AD) for LIE models.

Main Methods:

• Utilizing explicit conformational sampling for protein-bound and unbound ligand states.
• Employing statistically weighted LIE schemes for improved accuracy.
• Developing simulation-based strategies for LIE model applicability domain (AD) assessment.

Main Results:

• Demonstrated accurate and efficient calculation of ΔG for diverse compounds binding to flexible proteins.
• Introduced a statistically weighted LIE scheme to handle protein flexibility.
• Developed methods for AD assessment based solely on simulation and training data.

Conclusions:

• The LIE approach, enhanced with statistically weighted schemes and AD assessment, provides accurate and efficient ΔG calculations.
• The developed AD assessment strategies increase confidence in predictions for novel compounds.
• Available tools facilitate automated LIE computation and AD evaluation for broader application.