Geometry Optimization Using the Frozen Domain and Partial Dimer Approaches in the Fragment Molecular Orbital Method:...

Area of Science:

• Computational Chemistry
• Molecular Modeling
• Drug Discovery

Background:

• The Fragment Molecular Orbital (FMO) method with the frozen domain (FD) approximation offers efficiency for large system geometry optimization.
• Existing FD formulations, including frozen domain dimer (FDD), have been established for computational efficiency.

Purpose of the Study:

• To implement and evaluate a novel frozen domain and partial dimer (FDPD) method for molecular geometry optimization.
• To assess the FDPD method's performance in refining protein-ligand complexes for drug discovery applications.

Main Methods:

• Implementation of the frozen domain and partial dimer (FDPD) method, a variation of the frozen domain approximation.
• Application of FDPD/HF/6-31G* for geometry optimization of protein-ligand complexes, including a beta2-adrenergic receptor, human estrogen receptor, and influenza virus neuraminidase.
• FDPD/HF/6-31G* and FMO-MP2/6-31G* calculations for Serine-Threonine Kinase Pim1 and its inhibitors to analyze structure-activity relationships.

Main Results:

• The FDPD method achieved nearly 50% reduction in computational time for optimizing the active site of a large G-protein coupled receptor compared to conventional methods.
• FDPD optimization refined crystal structures by relaxing steric repulsion and optimizing hydrogen bonding in protein-ligand complexes.
• Expanding the optimized region in FDPD calculations significantly improved the correlation between pIC50 and ligand binding energy, enhancing structure-activity relationships.

Conclusions:

• The FDPD method provides a computationally efficient and accurate approach for molecular geometry optimization.
• This method shows significant promise for high-precision structure refinement in structure-based drug discovery.
• Improved structure-activity relationship analysis through FDPD optimization can accelerate the identification of potential drug candidates.