Elucidation of structure-activity relationship pathways in biological screening data

Area of Science:

• Computational chemistry and cheminformatics
• Drug discovery and development

Background:

• High-throughput screening (HTS) generates large datasets of potential drug candidates.
• Understanding local structure-activity relationships (SARs) within HTS data is crucial for hit prioritization.
• Existing methods may not systematically explore diverse SAR environments within screening hits.

Purpose of the Study:

• To develop a computational method for organizing HTS hits based on local SARs.
• To identify and rank SAR pathways within screening data to guide hit selection.
• To systematically explore and prioritize hits by leveraging SAR information content.

Main Methods:

• Computational molecular network analysis of HTS data, including inhibition assays and cell-based screens.
• Development of a scoring function to identify SAR pathways based on similarity and potency relationships.
• Ranking of alternative SAR pathways by their information content within hit clusters.

Main Results:

• Network representations enable focused analysis of different local SAR environments.
• Identified SAR pathways lead from active compounds to key compounds forming activity cliffs.
• Systematic exploration and ranking of pathways allow for consistent hit prioritization.

Conclusions:

• Computational network analysis provides a robust framework for organizing and interpreting HTS data.
• The developed scoring function effectively identifies and ranks SAR pathways for hit selection.
• This approach enhances the efficiency and consistency of prioritizing compounds in drug discovery.