Retrospective analysis of an experimental high-throughput screening data set by recursive partitioning

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Biostatistics

Background:

• Combinatorial chemistry enables the generation of large compound libraries.
• High-throughput screening (HTS) is crucial for identifying bioactive compounds.
• Efficiently designing and screening these libraries presents a significant challenge.

Purpose of the Study:

• To demonstrate the design of targeted compound libraries using nonparametric statistical methods.
• To apply recursive partitioning to large datasets of compounds and biological data.
• To improve hit rates and screening efficiency in HTS.

Main Methods:

• Utilized nonparametric statistical methods, specifically recursive partitioning.
• Applied the method to large datasets containing thousands of compounds and their biological data.
• Validated the approach on an experimental high-throughput screening dataset.

Main Results:

• Recursive partitioning enables the design of targeted compound libraries.
• The method improved the hit rate of primary screens by approximately 4- to 5-fold.
• Screening efficiency was increased, identifying about 75% of active compounds by screening less than one-fifth of the library.

Conclusions:

• Nonparametric statistical methods, like recursive partitioning, are effective for designing targeted compound libraries.
• This approach significantly enhances the efficiency and success rate of high-throughput screening.
• The findings suggest a more focused and productive strategy for drug discovery efforts.