Meeko: Molecule Parametrization and Software Interoperability for Docking and Beyond.

Area of Science:

• Computational chemistry
• Cheminformatics
• Molecular modeling

Background:

• Accurate molecule parametrization is crucial for reliable molecular docking and dynamics simulations.
• Challenges exist in handling diverse small molecules and complex biomacromolecules (proteins, nucleic acids) due to data format limitations and chemical diversity.
• Existing methods face difficulties in validating correctness and providing accurate parameters for a wide range of molecules.

Purpose of the Study:

• To develop a robust and customizable Python package for molecular parametrization.
• To address the limitations of existing tools in handling chemical accuracy and high-throughput processing.
• To provide an improved solution for preparing receptors and ligands in molecular modeling studies.

Main Methods:

• Developed Meeko, a Python package utilizing the RDKit cheminformatics library.
• Modeled small molecules as single RDKit molecules and biological macromolecules as multiple RDKit molecules per residue.
• Designed Meeko for high-throughput processing and customizability, enabling scripting for automated workflows.

Main Results:

• Meeko provides a chemically accurate molecular representation by leveraging RDKit.
• The package is highly customizable and scriptable, facilitating efficient high-throughput preparation of molecular structures.
• Meeko serves as a replacement for MGLTools in receptor and ligand preparation, enhancing computational efficiency.

Conclusions:

• Meeko offers a significant advancement in molecular parametrization for computational chemistry.
• The package effectively addresses challenges in handling diverse molecular structures and ensures accuracy in docking and dynamics.
• Meeko's design promotes efficient and accurate molecular modeling, supporting a wide range of applications in drug discovery and biochemical research.