Application of Free-Wilson selectivity analysis for combinatorial library design.

Area of Science:

• Computational chemistry
• Medicinal chemistry
• Drug discovery

Background:

• Quantitative structure-activity relationship (QSAR) models are crucial for predicting compound activity.
• Virtual libraries offer a vast resource for identifying novel drug candidates.
• Understanding target selectivity is key to developing effective and safe therapeutics.

Purpose of the Study:

• To develop a novel ligand-based computational approach for generating virtual compound libraries.
• To apply the Free-Wilson methodology for extracting predictive rules from experimental data.
• To construct R-group selectivity profiles for kinase and phosphodiesterase (PDE) targets.

Main Methods:

• Application of in silico QSAR models.
• Utilizing the Free-Wilson methodology to derive activity rules.
• Enumeration of virtual libraries and prediction of compound activities.
• Construction of R-group selectivity profiles using QSAR-derived activity contributions.
• Integration of protein structural data (X-ray) for enhanced analysis.

Main Results:

• Successful extraction of predictive rules from kinase and PDE datasets.
• Generation and prediction of activities for virtual compound libraries.
• Development of R-group selectivity profiles revealing target-specific contributions.
• Demonstration of improved understanding of selectivity relationships between kinase and PDE families.

Conclusions:

• The presented computational approach enables efficient generation of virtual libraries.
• QSAR models and selectivity profiles provide valuable insights into ligand-target interactions.
• This methodology can aid in the rational design of selective inhibitors for kinase and PDE targets.