Using weighted entropy to rank chemicals in quantitative high-throughput screening experiments

Area of Science:

• Drug discovery and chemical genomics
• Computational toxicology and cheminformatics

Background:

• Quantitative high-throughput screening (qHTS) enables simultaneous testing of thousands of chemicals.
• Traditional analysis often relies on sigmoidal models (e.g., Hill equation), which may not fit all concentration-response profiles.
• Prioritizing chemicals for further investigation is challenging due to model uncertainties.

Purpose of the Study:

• To introduce and evaluate a novel method for ranking chemicals based on their concentration-response profiles.
• To address the limitations of sigmoidal models in qHTS data analysis.
• To provide a robust strategy for prioritizing compounds for follow-up studies.

Main Methods:

• Utilized weighted Shannon entropy to derive profile-specific statistics from response probability distributions.
• Applied the method to simulated data from the Hill equation model.
• Validated the approach on a chemical genomics dataset for androgen receptor agonist activity.

Main Results:

• The weighted Shannon entropy approach successfully ranked compounds based on their concentration-response profiles.
• The method demonstrated effectiveness in prioritizing chemicals, even when sigmoidal models were inadequate.
• The strategy proved applicable to real-world chemical genomics data.

Conclusions:

• Weighted Shannon entropy offers a flexible and robust alternative for analyzing qHTS data.
• This approach enhances the prioritization of candidate compounds for drug discovery and toxicological studies.
• The method complements existing activity call algorithms by providing a model-agnostic ranking system.