Scalable, compressed phenotypic screening using pooled perturbations.

Area of Science:

• Biochemistry
• Cell Biology
• Genomics

Background:

• High-throughput phenotypic screening is essential for drug discovery but faces scalability limitations.
• Current methods require large sample sizes, increasing labor and costs.

Purpose of the Study:

• To develop and validate a compressed screening approach using perturbation pooling and computational deconvolution.
• To enhance the efficiency and reduce the cost of phenotypic screens.

Main Methods:

• Pooling of exogenous perturbations followed by computational deconvolution.
• Benchmarking against conventional screening with small-molecule libraries and high-content imaging.
• Application in pancreatic cancer organoid and peripheral blood mononuclear cell (PBMC) immune response studies.

Main Results:

• Demonstrated increased efficiency of compressed experimental designs compared to conventional methods.
• Mapped transcriptional responses in pancreatic cancer organoids to tumor microenvironment ligands, identifying clinically relevant phenotypic shifts.
• Identified pleiotropic effects of compounds on PBMC immune responses.

Conclusions:

• The developed compressed screening approach significantly enhances the scalability and cost-effectiveness of phenotypic screens.
• This method facilitates biological discovery and advances drug discovery efforts by enabling the use of information-rich readouts.
• The approach is applicable to diverse biological systems, including cancer organoids and immune cell responses.