Screenit: Visual Analysis of Cellular Screens.

Area of Science:

• Biotechnology
• Cell Biology
• Bioinformatics

Background:

• High-throughput and high-content screening generate large, complex datasets from cell cultures exposed to various drugs.
• These datasets possess a hierarchical structure, with cell features derived from image data and higher levels representing experimental conditions.
• Analyzing these multivariate datasets efficiently is crucial for drug discovery and biological research.

Purpose of the Study:

• To introduce Screenit, a novel visual analysis approach for high-throughput and high-content screening data.
• To enable interactive navigation and analysis of multivariate data across multiple hierarchy levels and levels of detail.
• To integrate modeling of cell phenotypes and drug effects (hits), alongside quality control for anomaly detection.

Main Methods:

• Development of Screenit, a visual analysis tool in collaboration with screening experts.
• Integration of interactive modeling for cell physical states (phenotypes) and drug effects (hits).
• Implementation of quality control features for anomaly detection in screening data.

Main Results:

• Screenit facilitates exploration of multivariate data from large-scale cell-based screening experiments.
• The tool allows for interactive analysis of cell phenotypes and identification of potential drug hits.
• Demonstrated effectiveness on the CellMorph dataset, comprising 6 million cells across 20,000 cell cultures.

Conclusions:

• Screenit provides an effective visual analysis solution for complex high-content screening data.
• The approach supports efficient navigation, analysis, and quality control in drug discovery workflows.
• Screenit enhances the ability to model cellular responses and identify drug effects in large datasets.