Nuisance Compounds, PAINS Filters, and Dark Chemical Matter in the GSK HTS Collection

Area of Science:

• Medicinal Chemistry
• Drug Discovery
• Computational Chemistry

Background:

• High-throughput screening (HTS) frequently yields hits with undesirable properties, necessitating effective filtering strategies.
• Pan-Assay Interference Compounds (PAINS) are a recognized class of problematic molecules, with guidelines from organizations like the American Chemical Society (ACS) sparking debate on filter application.
• Existing filters require empirical validation and refinement to optimize their utility in drug discovery pipelines.

Purpose of the Study:

• To comprehensively analyze the promiscuity profile of a large-scale HTS compound collection.
• To evaluate the effectiveness of established filters and identify novel nuisance compound classes.
• To provide empirical data to guide the design and optimization of HTS collections and hit-selection strategies.

Main Methods:

• Utilized an inhibitory frequency index to assess compound promiscuity across hundreds of screening assays.
• Analyzed the entire GlaxoSmithKline (GSK) HTS collection, comprising over 2 million unique compounds.
• Reviewed and analyzed previously published filters alongside newly identified nuisance structure classes.

Main Results:

• Detailed promiscuity profiles were generated for a vast HTS compound library.
• Empirical data on the performance of various filters and nuisance compound classes were obtained.
• Identified specific compound classes that frequently interfere with screening assays, posing challenges in hit identification.

Conclusions:

• The study provides a data-driven assessment of compound promiscuity within a large HTS collection.
• Findings offer valuable insights for refining hit triaging strategies and designing more effective HTS compound libraries.
• The comprehensive analysis serves as a resource for improving the efficiency and reliability of early-stage drug discovery.