Novel approach to structure-based pharmacophore search using computational geometry and shape matching techniques

Area of Science:

• Computational chemistry
• Drug discovery
• Structural biology

Background:

• Structure-based virtual screening (SBVS) methods accelerate drug discovery by analyzing molecular structures.
• Existing SBVS methods often rely on detailed molecular docking, which can be computationally intensive.
• Rapid overlay of chemical structures (ROCS) is a shape-based method using ligand structures as queries.

Purpose of the Study:

• To develop and validate a novel structure-based pharmacophore search method (Shape4) for efficient virtual screening.
• To adapt shape-based screening to utilize empty crystal structures, expanding its applicability.
• To improve hit identification and diversity in drug discovery pipelines.

Main Methods:

• Shape4 employs a variant of ROCS shape technology with empty crystal structures.
• A computational geometry method and geometric casting algorithm generate a pseudoligand (negative image) of the target binding site.
• An efficient shape comparison algorithm from the OE SHAPE Toolkit matches molecules to the pseudoligand.

Main Results:

• Shape4 was computationally validated using known biologically active compounds from the WOMBAT database.
• Virtual screening experiments on five targets showed Shape4 achieved similar or better enrichment ratios compared to other methods.
• The method often produced greater diversity among top-ranking computational hits.

Conclusions:

• Shape4 offers an efficient and effective structure-based virtual screening approach.
• The method's ability to use empty crystal structures broadens its utility in drug discovery.
• Shape4 enhances hit identification and diversity, contributing to more robust lead generation.