GateMeClass

GateMeClass enables both semi-supervised and supervised annotation of cytometry data based on a marker table that can be defined manually or extracted from an external annotated dataset.

Key Features:

• Dual annotation modes: Supports both semi-supervised and supervised classification approaches for cytometry data.
• Marker table flexibility: Allows marker tables to be defined manually or extracted automatically from an external annotated dataset.
• Rapid execution: Provides fast computational performance compared to existing classification methods.
• Biological interpretability: Maintains high biological interpretability through marker table-based annotation analogous to traditional manual gating.

Scientific Applications:

• Oncoimmunology research: Identifies cell populations within the tumor microenvironment through analysis of protein marker expression.
• Mass cytometry annotation: Annotates benchmark mass cytometry datasets with accuracy comparable to reference-based and marker table-based methods.
• Flow cytometry annotation: Classifies cell populations in flow cytometry datasets.
• Cell heterogeneity analysis: Analyzes cell heterogeneity in biological samples by examining protein marker expression patterns.

Methodology:

GateMeClass implements gate mining and classification algorithms in R, combining marker table-based annotation with automated extraction from annotated reference datasets.