Reverse fingerprinting and mutual information-based activity labeling and scoring (MIBALS)

Area of Science:

• Computational chemistry
• Cheminformatics
• Drug discovery

Background:

• Virtual high-throughput screening (vHTS) is crucial for identifying potential drug candidates.
• Traditional methods for analyzing molecular activity and identifying pharmacophores have limitations.

Purpose of the Study:

• To introduce and evaluate a novel mutual information based activity labeling and scoring (MIBALS) approach for reverse fingerprint analysis.
• To demonstrate MIBALS's capability in ranking compounds during vHTS and identifying critical pharmacophore fragments.
• To compare MIBALS performance against established similarity search methods.

Main Methods:

• Development of the MIBALS algorithm for activity labeling and scoring.
• Application of MIBALS for whole molecule ranking in vHTS experiments.
• Utilizing MIBALS fragment scores to identify pharmacophore moieties and analyze structure-activity relationships.

Main Results:

• MIBALS successfully ranked compounds in vHTS experiments across 40 biological targets.
• Fragment scores from MIBALS accurately identified pharmacophore groups and highlighted beneficial/detrimental molecular features.
• MIBALS demonstrated superior or comparable retrieval rates compared to traditional group fusion similarity search methods.

Conclusions:

• MIBALS is a robust method for reverse fingerprint analysis in drug discovery.
• The approach effectively supports virtual screening and the elucidation of structure-activity relationships.
• MIBALS provides valuable insights into molecular interactions and pharmacophore identification.