Skip to contents

A per-cell record table from a simulated 12-donor immune profiling experiment, shaped for dittoViz::freqPlot(). Each donor (sample) contributes 150 cells and maps to exactly one condition and one batch, which is the nesting freqPlot() requires to compare per-sample cell-type frequencies across groups. batch is crossed with condition (three donors each), so it works as a color.by without confounding the comparison.

Usage

example_composition

Format

A data frame with 1800 rows and 7 columns:

cell_id

Unique cell identifier (character)

sample

Donor identifier, P01-P12 (factor); 150 cells each

condition

Disease state, Healthy or Disease (factor); six donors each

batch

Processing batch, B1 or B2 (factor); crossed with condition

cell_type

Annotated cell type (factor), the variable whose per-sample frequency freqPlot() tabulates

n_genes

Number of genes detected in the cell (integer)

percent_mito

Percentage of mitochondrial reads (numeric)

Source

Simulated. Per-donor compositions are Dirichlet draws around condition-specific means, with cell counts drawn multinomially. See data-raw/generate_example_data.R.

Details

Composition differs between the two conditions: the Disease donors show an expanded monocyte compartment and depleted CD4 T cells relative to Healthy.

Author

Jared Andrews