Essential considerations for using protein-ligand structures in drug discovery

Area of Science:

• Computational chemistry
• Structural biology
• Drug discovery

Background:

• Protein-ligand structures are fundamental to structure-based drug design (SBDD).
• Accurate assessment of protein-ligand structure quality is vital for developing reliable SBDD tools.
• Existing datasets used for validating molecular docking software may contain significant errors.

Purpose of the Study:

• To present methods for assessing protein-ligand structure quality.
• To introduce new criteria for identifying high-quality protein-ligand structures.
• To create a trustworthy database of protein-ligand structures for SBDD tool development.

Main Methods:

• Development of novel criteria for assessing protein-ligand structure quality.
• Application of these criteria to a dataset of 728 protein-ligand structures.
• Re-refinement of structures to ensure internal consistency for quality assessment.

Main Results:

• Only 17% of the 728 evaluated protein-ligand structures met the new quality criteria.
• A refined set of structures was established based on rigorous quality assessment.
• The creation of Iridium, a high-quality protein-ligand structure database.

Conclusions:

• A significant portion of existing protein-ligand structural data is of insufficient quality for SBDD.
• The developed quality assessment criteria are effective in identifying reliable structures.
• The Iridium database provides a robust resource for the development and validation of SBDD tools.