Exploring Topological Pharmacophore Graphs for Scaffold Hopping.

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Drug Discovery

Background:

• Ligand-based virtual screening aims to find novel active compounds.
• Scaffold hopping is crucial for identifying compounds with new core structures.
• Existing methods require improvement for effective scaffold hopping.

Purpose of the Study:

• To develop and evaluate a new method for scaffold hopping.
• To represent compounds using pharmacophore graphs (PhGs) for topological analysis.
• To rank PhGs based on scaffold coverage for prioritizing potential drug candidates.

Main Methods:

• Utilized topological representations of pharmacophore features on chemical graphs.
• Developed the NScaffold method to rank PhGs based on the number of scaffolds covered.
• Applied NScaffold to a thrombin inhibitor dataset for validation.

Main Results:

• The NScaffold method demonstrated superior performance compared to conventional ranking methods.
• Highest-ranked PhGs identified by NScaffold were analyzed for protein-ligand interactions.
• NScaffold successfully retrieved three known important interactions in the thrombin inhibitor dataset.

Conclusions:

• NScaffold is an effective method for prioritizing pharmacophore graphs for scaffold hopping.
• The method shows potential for identifying novel scaffold-hopped compounds with interpretable pharmacophore graphs.
• This approach aids in discovering new drug candidates with unique chemical scaffolds.