Validated ligand geometries for macromolecular refinement restraints and molecular-mechanics force fields

Area of Science:

• Structural biology
• Computational chemistry
• Biochemistry

Background:

• Macromolecular structure refinement relies on restraints derived from a priori information due to limited data.
• Accurate geometries of chemical components are crucial for generating reliable restraints.
• There is a need for precise restraints for both known and novel ligand entities in structural studies.

Purpose of the Study:

• To create a comprehensive library of minimized small molecule geometries and associated restraints.
• To enhance the accuracy of macromolecular structure refinement, particularly for ligand components.
• To provide validated resources for crystallography, cryo-EM, and molecular dynamics simulations.

Main Methods:

• Utilized density-functional quantum mechanics (QM) to minimize the energy and optimize the geometries of approximately 37,500 small molecules.
• Extracted small molecules from the Protein Data Bank's Chemical Component Dictionary.
• Validated minimized geometries against the Cambridge Structural Database.

Main Results:

• Developed a library containing ~37,500 minimized small molecule geometries.
• Generated restraint files compatible with crystallography, cryo-EM, and molecular dynamics simulations.
• Established procedures for generating new, accurate restraints for chemical entities.

Conclusions:

• The QM-minimized small molecule library and associated restraints improve the accuracy of macromolecular structure determination.
• The validated geometries and restraints support reliable modeling of ligands in biological structures.
• This resource facilitates the generation of high-quality structural models in various biophysical techniques.