Gating defines cell populations by drawing boundaries in parameter space. Essential for identifying cell subsets in flow cytometry and CyTOF data.
# R/Bioconductor
BiocManager::install(c('flowCore', 'flowWorkspace', 'openCyto', 'flowDensity'))Tell your AI agent what you want to do:
- "Gate lymphocytes from my flow data"
- "Apply an automated gating strategy"
- "Extract CD4+ T cells using a standard gating hierarchy"
"Create a polygon gate for lymphocytes based on FSC-A vs SSC-A" "Apply a rectangular gate for CD3+ cells" "Set up a quadrant gate for CD4 vs CD8"
"Use flowDensity to automatically gate the major populations" "Apply an openCyto gating template to my flowSet" "Find the optimal gate boundaries using density-based clustering"
"Apply a standard T cell gating strategy: singlets -> live -> CD3+ -> CD4/CD8" "Extract population statistics from my gating hierarchy" "Export gated populations as new FCS files"
- Load gating packages (flowWorkspace, openCyto, or flowDensity)
- Create GatingSet from flowSet
- Apply gates (manual or automated) in hierarchical order
- Visualize gating strategy
- Extract population statistics or gated cells
- Always gate singlets first (FSC-A vs FSC-H)
- Rectangular gates are fast and reproducible
- Polygon gates offer more flexibility for complex populations
- openCyto templates enable reproducible automated gating
- flowDensity uses density peaks for data-driven boundaries
| Type | Use Case |
|---|---|
| Rectangular | Simple threshold on 1-2 parameters |
| Polygon | Arbitrary shapes for complex populations |
| Quadrant | 4 populations from 2 parameters (e.g., CD4/CD8) |
| Boolean | Combine gates with AND, OR, NOT |
| Tool | Approach |
|---|---|
| flowDensity | Density-based automatic gating |
| openCyto | Template-based reproducible gating |
| flowClust | Model-based (mixture models) |
- flowWorkspace: doi:10.1186/s12859-018-2425-9
- openCyto: doi:10.1371/journal.pcbi.1003806