MetalloGen: Automated 3D Conformer Generation for Diverse Coordination Complexes.

Area of Science:

• Computational Chemistry
• Inorganic Chemistry
• Organometallic Chemistry

Background:

• Conformer generation is vital for computational chemistry, yet challenging for coordination complexes due to their complexity.
• Existing methods lack the flexibility and reliability for diverse coordination geometries, ligands, and stereochemistry.

Purpose of the Study:

• Introduce MetalloGen, a novel algorithm for automated 3D conformer generation of mononuclear coordination complexes.
• Address limitations in current tools for handling polyhapto ligands and complex stereochemistry.

Main Methods:

• MetalloGen accepts SMILES strings or molecular graphs as input.
• Algorithm benchmarked on curated experimental data (Cambridge Structural Database), MOR41 benchmark set, and catalytic reactions.
• Performance assessed for geometric accuracy, fidelity, and stereochemical control.

Main Results:

• MetalloGen successfully reproduced appropriate geometries with high fidelity across all test sets.
• Demonstrated robust stereochemical control, including for challenging cases with multiple polyhapto ligands.
• Consistently outperformed conventional conformer generators for complex systems.

Conclusions:

• MetalloGen provides a versatile and reliable solution for coordination complex conformer generation.
• Enhances accuracy and efficiency in computational investigations of inorganic and organometallic chemistry.
• Facilitates advanced structure-based modeling and property prediction for challenging chemical systems.