Small-Molecule Library Subset Screening as an Aid for Accelerating Lead Identification

Area of Science:

• Medicinal Chemistry
• Drug Discovery
• Computational Chemistry

Background:

• Large compound libraries are essential for drug discovery screening.
• Optimizing library subsets can improve screening efficiency and hit identification.
• Genentech established multiple focused library subsets to complement its main corporate library.

Purpose of the Study:

• To establish and validate diverse small-compound library subsets for drug discovery.
• To assess the utility of these subsets in guiding assay development and predicting hit rates.
• To identify ligand-efficient compounds using a property-restricted library.

Main Methods:

• Creation of validation, diversity, and property-restricted (in-between) library subsets.
• Selection of compounds based on representation of the main library's scaffold diversity.
• Screening of subsets using computational approaches and internal screening data analysis.
• Retrospective analysis of hit rates and scaffold recovery.

Main Results:

• Validation sets confirmed assay reproducibility and provided hit rate estimates.
• A diversity subset showed similar hit rates to the main library but recovered more unique hit scaffolds.
• The "in-between library" facilitated the identification of ligand-efficient compounds.

Conclusions:

• Purpose-focused, diversity-based library subsets accelerate lead discovery.
• Computational design and data analysis are crucial for optimizing library subsets.
• Strategic use of library subsets enhances the efficiency and success rate of drug discovery screening.