Skip to contents

Use auto_glycan_size() as the size argument of geom_glycan(), geom_node_glycan(), scale_x_glycan(), scale_y_glycan(), guide_glycan(), or anno_glycan() to fit complete cartoons uniformly. Nodes, linkage text, lines, and spacing shrink together. Cartoons in the same panel, axis, legend, or heatmap annotation slice share a scale factor, preserving their relative residue sizes.

Usage

auto_glycan_size(max_size = NULL)

Arguments

max_size

Optional positive upper limit for the whole-cartoon scale multiplier. NULL uses 1 for panel cartoons and 0.4 for axis, legend, and heatmap labels. Automatic sizing only shrinks from this limit.

Value

An automatic sizing specification accepted by glycan plotting interfaces. It is not a ggplot2 aesthetic value.

Details

Panel cartoons fit inside the panel and between neighbouring anchors, and occupy at most 30% of its width or height. Axis and heatmap labels fit their row or column spacing; their reserved width or height is initially limited to 20% of the graphics device. Their size is refined when drawn in the actual annotation viewport. Legend cartoons fit a device-based budget for the complete collection of labels. These limits are layout heuristics; automatic sizing does not detect other plot layers or move anchors. Anchors on a panel boundary still need scale expansion, and coincident anchors still overlap. Extremely dense figures may need a larger output device to keep linkage text readable.

Supply a numeric size instead to retain a fixed whole-cartoon multiplier. A mapped numeric size aesthetic in a glycan layer also uses fixed sizing; use ggplot2::scale_size_identity() for literal multipliers.

Examples

glycans <- data.frame(
  structure = c("Gal(b1-3)GalNAc(a1-", "Gal(b1-3)[GlcNAc(b1-6)]GalNAc(a1-"),
  value = c(1, 2)
)
ggplot2::ggplot(glycans, ggplot2::aes(structure, value)) +
  ggplot2::geom_col() +
  geom_glycan(
    ggplot2::aes(structure = structure),
    size = auto_glycan_size(max_size = 0.7),
    orient = "up",
    vjust = 0
  ) +
  ggplot2::scale_y_continuous(expand = ggplot2::expansion(mult = c(0, 0.4)))