A chemoinformatics analysis of hit lists obtained from high-throughput affinity-selection screening.

Area of Science:

• Biochemistry
• Medicinal Chemistry
• Drug Discovery

Background:

• High-throughput screening (HTS) is crucial for identifying novel drug candidates.
• Label-free, homogeneous screening technologies with mass spectrometry readout offer advantages in efficiency and data quality.
• Novartis Institutes for BioMedical Research developed SpeedScreen, an affinity-based HTS platform.

Purpose of the Study:

• To analyze the molecular properties of hits identified by the SpeedScreen platform.
• To assess the representativeness of SpeedScreen hits in relation to known drug scaffolds.
• To evaluate strategies for optimizing SpeedScreen compound sets for limited protein availability.

Main Methods:

• SpeedScreen utilizes affinity-selection, size-exclusion chromatography, and liquid chromatography/mass spectrometry for compound screening.
• Analysis of molecular properties of validated SpeedScreen hits across 26 targets.
• Application of chemoinformatics methods, including genetic algorithms and K-means clustering, for compound set optimization.

Main Results:

• SpeedScreen is a robust HTS technology that avoids frequent hitters and covalent binders.
• Identified hits are representative of common drug scaffold classes.
• Validated hits are enriched for more lipophilic and higher molecular weight compounds compared to the general library.
• Chemoinformatics methods can be used to create reduced compound sets for screening with limited protein.

Conclusions:

• SpeedScreen is a valuable tool for drug discovery, yielding drug-like hits.
• The characterized molecular properties of hits inform future drug design and library selection.
• Optimized screening sets can enhance the efficiency of SpeedScreen, even with limited protein quantities.