Biased Diversity for Effective Virtual Screening.

Area of Science:

• Computational chemistry and cheminformatics
• Drug discovery and medicinal chemistry

Background:

• Virtual screening (VS) has evolved beyond simple activity prediction.
• Current drug discovery necessitates balancing multiple compound properties and chemical diversity.

Purpose of the Study:

• To present Novartis's distinctive virtual screening practices.
• To highlight methods for biasing compound selection towards desired properties while ensuring chemical space diversity.

Main Methods:

• Integration of multiple computational tools for VS.
• Consideration of factors like triaging, compound quality, and accuracy.
• Application of activity prediction, multitask modeling, and virtual profiling.
• Implementation of automation, multiproperty bias, and diversity-focused designs.

Main Results:

• Successful integration of VS with multiproperty bias and diversity.
• Identification of quality compounds that inform the drug discovery pipeline.
• Development of biased-diversity designs for efficient compound selection.

Conclusions:

• Novartis's integrated VS approach effectively balances property optimization and chemical space exploration.
• This methodology enhances the identification of promising drug candidates.
• The practices discussed provide a framework for successful virtual screening in drug discovery.