Scalable data harmonization for single-cell image-based profiling with CytoTable

Area of Science:

• Computational Biology
• Bioinformatics
• Cellular Imaging

Background:

• High-content imaging (HCI) automates cell phenotype analysis from microscopy.
• Large-scale screening generates terabytes of complex, high-dimensional single-cell data.
• Current data curation methods present challenges in reproducibility and scalability.

Purpose of the Study:

• To introduce CytoTable, a software package for harmonizing single-cell image-based profiling data.
• To address analytical bottlenecks in curating high-dimensional datasets from HCI.
• To enhance data integration and reproducibility in the Cytomining ecosystem.

Main Methods:

• Development of CytoTable, a modular and portable software package.
• Implementation of a robust engine for harmonizing single-cell readouts.
• Facilitation of cross-language data integration for image analysis tools.

Main Results:

• CytoTable provides a scalable solution for curating diverse single-cell profiling data.
• The software ensures reproducible data integration across multiple image analysis platforms.
• Enables seamless preparation of data for downstream analysis with tools like Pycytominer.

Conclusions:

• CytoTable effectively resolves inconsistencies in single-cell data schemas and formats.
• The package promotes reproducibility and accelerates progress in image-based profiling research.
• CytoTable is a key enabler for advanced computational analysis in cell biology.