Plots Ig usage between two groups using Fisher's Exact Test

Plot the Ig usage from two groups and perform Fisher's Exact Test to find significant differences.

Usage

TS_plot_IG_usage(
  input_file,
  compare_groups,
  compare_what = "IGHV",
  grouping_variable = "named_clusters",
  export_pdf = F,
  return_only_data = F,
  return_only_sig_data = F,
  file_name_suffix = NULL,
  custom_gene_levels = NULL,
  file_type = "IMGT",
  ylim_max = NULL,
  p.adj.method = "hochberg",
  fill_colours = NULL,
  border_colour = "black",
  ggplot_theme = ggplot2::theme_classic(),
  plot_width = 15,
  plot_height = 5,
  remove_genes_present_in_one_dataset = F,
  plot_zeroes = T,
  plot_subtitle = "% of max",
  custom_plot_title = NULL,
  legend_title = "",
  plot_n_sequences = F,
  add_pattern = F,
  pattern = grouping_variable,
  pattern_colour = "black",
  pattern_fill = "black",
  pattern_density = 0.1,
  add_significancies = T,
  plot_only_significant_comparisons = T,
  sig_distance_above_bar = 1,
  sig_distance_between_significancies = 0.8,
  sig_line_thickness = 1,
  sig_label_distance_from_line = 0.2,
  sig_label_text_size = 6,
  return_usage_proportions = F
)

Arguments

input_file: A dataframe (e.g. from 10X Genomics Cellranger VDJ output) that includes a grouping column.
compare_groups: Character vector. Names of group values to compare.
compare_what: Genes to compare (e.g. "IGHV", "IGKV"). Defaults to "IGHV".
grouping_variable: Column used to group data. Defaults to "named_clusters"
export_pdf: Logical. If TRUE, exports the plot to a PDF. Defaults to FALSE
return_only_data: Logical. If TRUE, returns only test results. Defaults to FALSE.
return_only_sig_data: Logical. If TRUE, returns only significant test results. Defaults to FALSE.
file_name_suffix: Character string. Optional suffix to be added to the exported pdf.
custom_gene_levels: Character vector. Optional, vector used to set gene list levels.
file_type: Set type of file/annotation input. Defaults to "cellranger". Options are "IMGT" and "AIRR".
ylim_max: Numeric. Optional upper limit for y-axis. ggplot2 will adjust the y-axis scale to fit the data.
p.adj.method: Method for adjust for multiple tests. Defaults to 'Hochberg'. Alternative is 'FDR'.
fill_colours: A character vector of fill colours. Defaults to greyscale.
border_colour: Colour of bar borders. Defaults to "black".
ggplot_theme: Optional ggplot2 theme to apply. Defaults to "theme_classic".
plot_width: Numeric. Width of exported plot in inches. Used if 'export_pdf = TRUE'. Defaults to 15.
plot_height: Numeric. Height of exported plot in inches. Used if 'export_pdf = TRUE'. Defaults to 5.
remove_genes_present_in_one_dataset: Logical. If TRUE, removes genes not shared by both groups. Defaults to FALSE
plot_zeroes: Logical. If FALSE, genes with 0 counts in one group are not plotted, resulting in a wider bar for the other group.
plot_subtitle: Character string. Plot subtitle. Defaults to "% of max".
legend_title: Character string. Optional. Title of the legend.
plot_n_sequences: Logical. Should the number of sequences used for each group be plotted in the legend? Defaults to FALSE.
add_pattern: Logical. Should the bars be plotted with pattern? Defaults to FALSE. Requires ggpattern
pattern: What to base the pattern on. Defaults to the grouping_variable.
pattern_colour: String. Colour of the pattern. Defaults to black.
pattern_fill: String. Fill colour of the pattern. Defaults to black.
pattern_density: Numeric. Density of the pattern. Defaults to 0.1
add_significancies: Whether to calculate and plot test results. Defaults to TRUE.
plot_only_significant_comparisons: Logical. Should only significant test results be plotted? Defaults to TRUE.
sig_distance_above_bar: Numeric. What distance above the tallest bar the stats results should be plotted. Defaults to 1.
sig_distance_between_significancies: Numeric. What distance should be between each test result? Defaults to 0.8.
sig_line_thickness: Numeric. Thickness of the comparison line. Defaults to 1.
sig_label_distance_from_line: Numeric. Distance of asterisk from the comparison line. Defaults to 0.2.
sig_label_text_size: Numeric. Size of asterisks. Defaults to 6.
return_usage_proportions: Logical. Whether to return the proportions of usage per group? Defaults to FALSE.

Value

Returns a ggplot object, or a data.frame containing test results.

Examples

# TS_plot_IG_usage(input_file = VDJ_data, compare_groups = c("Group1", "Group2", "Group3"), compare_what = "IGHV", export_pdf = TRUE, file_type = "IMGT")