MovableType Software for Fast Free Energy-Based Virtual Screening: Protocol Development, Deployment, Validation, and...

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular modeling

Background:

• Protein-ligand binding free energy estimation is crucial for drug discovery but computationally expensive.
• Traditional methods struggle with complex energy surfaces and require significant computational resources.
• The Movable Type (MT) methodology was developed to address these computational challenges.

Purpose of the Study:

• To introduce a software package, the DivCon Discovery Suite with the MovableType module, for efficient free energy estimation.
• To detail the computational procedures, improvements, and software settings for the MT method.
• To validate the MT method's robustness and predictability across various protein-ligand structural cases.

Main Methods:

• Utilizing the Movable Type (MT) sampling methodology based on numerical integration.
• Calculating independent atomic partition functions from initial molecular conformations.
• Estimating molecular free energy using ensembles of atomic partition functions.
• Implementing the MT protocol in a fast, fully encapsulated software package (DivCon Discovery Suite).

Main Results:

• The MT method significantly reduces computational expense for energy state sampling.
• The software package performs MT free energy estimation efficiently.
• Validation benchmarks demonstrate comparable predictability using both high-quality X-ray structures and lower-quality docked models.
• The method shows robustness across a broad range of protein-ligand structural cases.

Conclusions:

• The DivCon Discovery Suite with the MovableType module provides a fast and reliable tool for free energy estimation in drug discovery.
• The MT method offers comparable predictability regardless of input structure quality, broadening its applicability.
• The developed software and validation benchmarks facilitate routine free energy calculations in industrial settings.