A Perspective on Water Site Prediction Methods for Structure Based Drug Design

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular modeling

Background:

• Computational methods evaluating water molecule thermodynamics at the solute-solvent interface have emerged for protein-ligand binding assessment.
• Existing tools sometimes provide obvious insights or are most effective when applied retrospectively.

Purpose of the Study:

• To explore the utility of computational water methods for understanding non-intuitive drug discovery results.
• To demonstrate how these methods can generate novel chemistry ideas and guide prospective drug design.

Main Methods:

• Review and perspective on computational water methods for assessing thermodynamic properties.
• Analysis of case studies from GSK and other institutions.
• Focus on retrospective explanation of structure-activity relationships and prospective application in chemistry design.

Main Results:

• Computational water methods can retrospectively explain non-intuitive structure-activity relationships in drug discovery.
• These methods have been prospectively applied to generate innovative chemistry designs.
• The interpretation and application of results from water methods are crucial for their effective use.

Conclusions:

• Computational water methods offer valuable insights into protein-ligand interactions, aiding in understanding complex binding phenomena.
• Strategic application of these methods can accelerate drug discovery by informing both retrospective analysis and prospective design.
• Further development and integration of water methods hold significant potential for maximizing impact in drug discovery research.