Assessing consistency across functional screening datasets in cancer cells

Area of Science:

• Genomics
• Bioinformatics
• Cancer Research

Background:

• High-throughput screening in cancer cell lines is crucial for identifying therapeutic agents and targets.
• Previous consistency assessments were limited, examining only two datasets and selected features, potentially overlooking global data concordance issues.

Purpose of the Study:

• To develop and present novel measures for assessing the global consistency of functional genomics and compound screening data in cancer cell lines.
• To create an accessible web application for users to explore and visualize data consistency.

Main Methods:

• Assembled nine compound screening datasets and three functional genomics datasets.
• Derived direct and indirect measures of data consistency, including associations with copy number-adjusted gene expression data.
• Integrated consistency measures into an interactive web application, the Functional Data Consistency Explorer (FDCE).

Main Results:

• Identified significant inconsistencies across multiple cancer cell line screening datasets.
• Developed a comprehensive approach to assess global data consistency beyond selected features.
• The Functional Data Consistency Explorer (FDCE) enables interactive querying and visualization of feature-specific data consistency.

Conclusions:

• Existing cancer cell line screening data exhibit considerable inconsistencies, impacting the reliability of therapeutic agent and target identification.
• The developed functional data consistency measures and the FDCE web tool provide a valuable resource for researchers to evaluate data reliability.
• Promotes more rigorous assessment of high-throughput screening data for improved translational research in cancer therapy.