ShinySOM

ShinySOM performs clustering, visualization, and statistical dissection of high-dimensional flow and mass cytometry data to enable reproducible identification of cell populations.

Key Features:

• Self-Organizing Maps (SOMs): Uses self-organizing maps to drive clustering and visualization of high-dimensional cytometry datasets.
• FlowSOM-based workflow: Builds upon the FlowSOM workflow with algorithmic and workflow-level enhancements.
• High-dimensional flow and mass cytometry support: Processes high-dimensional flow cytometry and mass cytometry (CyTOF) data for high-throughput analyses.
• Enhanced computational performance: Implements performance optimizations that improve computational efficiency relative to traditional FlowSOM implementations.
• Advanced visualizations: Produces visualization outputs to enhance interpretability of complex, high-dimensional results.
• Data dissection and sample statistics: Provides detailed dissection of samples with precise statistical information for each dissected sample.
• R-compatible metadata: Generates metadata compatible with R to support batch processing and downstream analyses within the R ecosystem.

Scientific Applications:

• Identification of novel cell populations: Enables discovery and characterization of previously unrecognized cell subsets from cytometry data.
• Immune profiling: Supports analysis of immune responses by resolving cellular phenotypes and frequencies in immunology studies.
• Cancer research: Facilitates characterization of tumor and tumor-infiltrating cell populations in cancer studies using cytometry datasets.
• Single-cell disease mechanism characterization: Aids in characterizing disease mechanisms at single-cell resolution through detailed cellular analysis.

Methodology:

Implements self-organizing maps within an extended FlowSOM workflow, includes visualization generation, sample-wise statistical dissection, and produces R-compatible metadata with performance-oriented optimizations.