CATALYST

CATALYST preprocesses and analyzes high-dimensional mass cytometry and flow cytometry (CyTOF/HDCyto) data by performing spillover compensation, normalization, clustering, and statistical modeling of cellular populations.

Key Features:

• Spillover Compensation: Implements bead-based spillover compensation to model and correct signal interference between CyTOF channels.
• Comprehensive Preprocessing Pipeline: Supports file concatenation, bead-based normalization, single-cell deconvolution, doublet and debris removal, live-cell gating, and batch correction.
• SingleCellExperiment Integration: Organizes cytometry datasets within the SingleCellExperiment data structure for consistent downstream analysis.
• Cell Population Identification: Applies FlowSOM clustering to identify cellular populations with optional manual merging of clusters.
• Differential Cytometry Analysis: Performs statistical modeling using generalized linear mixed models (GLMMs) and linear mixed models (LMMs) to analyze cell population abundance and marker expression changes.
• Quality Control Assessment: Provides modules to evaluate machine sensitivity, staining performance, and sample-level technical variation.

Scientific Applications:

• Mass Cytometry Data Analysis: Processes and interprets CyTOF and high-dimensional cytometry datasets for single-cell studies.
• Cell Population Profiling: Identifies and characterizes immune and cellular subpopulations in complex biological samples.
• Differential Cytometry Studies: Detects changes in cellular abundance and signaling marker expression across experimental conditions.
• Clinical and Systems Immunology Research: Supports large-scale cytometry analyses in heterogeneous tissues and clinical cohorts.

Methodology:

CATALYST performs bead-based spillover compensation, normalization, and quality control on cytometry data, organizes processed data within a SingleCellExperiment object, identifies cell populations using FlowSOM clustering, and applies generalized linear mixed models (GLMMs) or linear mixed models (LMMs) for differential analysis of population abundance and marker expression.