Prepare significance annotation data for a ggplot — prepare_sig

Facet-aware version. If p_table contains facet columns matching the plot, annotations are placed only in the matching panels. If not, annotations are repeated across all panels.

Usage

prepare_sig_annotations(
  plot,
  p_table,
  y_npc,
  group1_col = "group1",
  group2_col = "group2",
  label_col = "significance",
  x_levels = NULL,
  facet_cols = NULL,
  draw_brackets = TRUE,
  bracket_tip_npc = 0.015,
  bracket_margin_npc = 0.02,
  text_size_pt = NULL,
  star_y_npc_offset = 0.01
)

Arguments

plot: A ggplot object with a discrete x-axis and continuous y-axis.
p_table: A data frame containing at minimum group1, group2, and significance columns.
y_npc: Numeric vector of label y positions in npc coordinates.
group1_col, group2_col, label_col: Column names in p_table.
x_levels: Optional character vector giving x-axis order.
facet_cols: Optional character vector naming the facet columns in p_table. If NULL, they are auto-detected.
draw_brackets: Logical; whether brackets should be drawn.
bracket_tip_npc: Length of bracket tips in npc units.
bracket_margin_npc: Gap between label and bracket top line in npc units.
text_size_pt: Text size in points. Defaults to theme text size minus 2.
star_y_npc_offset: Upward offset, in npc units, applied only to star-only labels like "*", "**", "***".

Value

A tibble with one row per annotation per matched panel.

Details

Star-only labels like "***" are kept as ASCII for PDF robustness and nudged upward slightly so they sit more like "ns".