pyInfinityFlow

pyInfinityFlow performs imputation of cell surface protein markers from flow cytometry Infinity Flow experiments using XGBoost regression to enable large-scale cellular characterization and discovery.

Key Features:

• Imputation of Cell Surface Protein Markers: Uses XGBoost regression to impute hundreds of cell surface protein markers from flow cytometry data, extending marker coverage beyond the typical few dozen measurable by conventional flow cytometry.
• End-to-End Workflow Integration: Provides an end-to-end analysis workflow for Infinity Flow data and interoperates with established Python packages used in single-cell genomics analysis.
• Handling Large Datasets Efficiently: Processes millions of cells without down-sampling, enabling detection of both common and rare cell populations.
• Optimized Performance: Implements optimizations focused on increased speed and memory efficiency for large datasets.
• Novel Marker Identification: Nominates novel markers to support the design of new flow cytometry gating strategies for predicted cell populations.
• Flexibility and Adaptability: Adapts to various Infinity Flow experimental designs for diverse cell discovery analyses.

Scientific Applications:

• Cellular characterization and rare population detection: Enables detailed cellular characterization and identification of rare cell populations from flow cytometry data.
• Immunology and oncology studies: Supports marker imputation and cell population discovery in immunology and oncology research.
• Developmental biology and complementary single-cell genomics: Facilitates studies in developmental biology and complements single-cell genomics by extending surface marker information.

Methodology:

pyInfinityFlow uses XGBoost regression models to predict missing cell surface protein markers from Infinity Flow flow cytometry data and supports processing of millions of cells without down-sampling.