GateMeClass: Gate Mining and Classification of cytometry data.

Area of Science:

• Biotechnology
• Computational Biology
• Immunology

Background:

• Cytometry techniques analyze cell heterogeneity via protein marker expression, crucial for oncoimmunology and tumor microenvironment studies.
• Current cytometry data annotation tools are either reference-based supervised methods or semisupervised methods requiring manual marker table definition.
• A gap exists for methods combining both classification approaches while preserving biological interpretability via marker tables.

Purpose of the Study:

• To introduce GateMeClass, a novel tool for cytometry data classification.
• To enable both supervised and semisupervised annotation using a flexible marker table definition.
• To address the limitations of existing tools in cytometry data analysis.

Main Methods:

• GateMeClass (Gate Mining and Classification) was developed for annotating cytometry datasets.
• The tool supports marker tables that can be manually defined or extracted from external annotated datasets.
• Accuracy was evaluated on benchmark mass and flow cytometry datasets.

Main Results:

• GateMeClass demonstrated performance comparable to existing reference-based and marker table-based methods.
• The tool offers enhanced flexibility and rapid execution times for cytometry data analysis.
• Successful annotation of multiple benchmark datasets was achieved.

Conclusions:

• GateMeClass provides a versatile solution for cytometry data annotation, accommodating both supervised and semisupervised approaches.
• The tool's ability to utilize manually defined or extracted marker tables enhances its applicability.
• GateMeClass offers a valuable advancement for oncoimmunology research and tumor microenvironment analysis.