ScreenUpset - Visualize cell type intersections from screened Seurat object

This function creates an UpSet plot to visualize intersections between different screening methods (e.g., scissor, scAB, scPAS, scPP) in a Seurat object metadata. It calculates all possible combinations of screening types and displays the number of cells positive for each combination.

Usage

ScreenUpset(
  screened_seurat,
  screen_type = NULL,
  show_plot = TRUE,
  n_intersections = 20,
  x_lab = "Screen Set Intersections",
  y_lab = "Number of Cells",
  title = "Cell Counts Across Screen Set Intersections",
  bar_color = "#4E79A7",
  combmatrix_point_color = "black",
  ...
)

Arguments

screened_seurat

A Seurat object containing screening results in metadata. Must contain columns with screening types marked as "Positive".

screen_type

Character vector of screening types to analyze. Default: NULL, indicating that a self-search pattern will be used.

show_plot

Whether to show the upset plot. Default: TRUE.

n_intersections

Number of intersections to display in the plot. Default: 20.

x_lab

Label for the x-axis. Default: "Screen Set Intersections".

y_lab

Label for the y-axis. Default: "Number of Cells".

title

Plot title. Default: "Cell Counts Across Screen Set Intersections".

bar_color

Color for the bars in the plot. Default: "#4E79A7".

combmatrix_point_color

Color for points in the combination matrix. Default: "black".

...

Additional arguments passed to ggplot2::theme() for customizing the plot appearance.

Value

A list containing two elements:

• plot: A ggplot object displaying the UpSet plot

• stats: A data frame with intersection statistics including:

  • intersection: Name of the intersection

  • sets: List of screening types in the intersection

  • count: Number of cells positive for all screening types in the intersection

Details

The function performs the following steps:

1. Validates input parameters and checks if specified screening types exist in metadata

2. Generates all possible combinations of screening types (from 1 to the total number of types)

3. Creates a logical matrix of positive cells for efficient computation

4. Calculates cell counts for each intersection using vectorized operations

5. Creates an UpSet plot visualization with customizable appearance

Examples

if (FALSE) { # \dontrun{
# Basic usage with default parameters
result <- ScreenUpset(screened_seurat = my_seurat_obj)
print(result$plot)

# Customize screening types and appearance
result <- ScreenUpset(
  screened_seurat = my_seurat_obj,
  screen_type = c("scissor", "scAB"),
  n_intersections = 15,
  title = "Custom Title",
  bar_color = "darkred"
)
} # }