Protein domain lollipop

Variants along a protein, drawn over its domain architecture: a backbone spanning the sequence with domain rectangles on it, mutation stems whose head area is proportional to recurrence, and an optional post-translational modification track below. Hotspots inside a functional domain read very differently from truncating variants scattered across one, which is what this figure exists to show.

Usage

lollipop(
  variants,
  length,
  gene = NULL,
  uniprot = NULL,
  domains = NULL,
  ptms = NULL,
  classes = NULL,
  class_colors = NULL,
  domain_colors = NULL,
  label_top_n = 12,
  show_ptms = TRUE,
  show_domains = TRUE,
  show_legend = TRUE,
  min_head_radius = 3,
  max_head_radius = 11,
  y_label = "samples",
  backbone_color = "#E6DCC8",
  stem_color = "#93a1b8",
  theme = NULL,
  width = NULL,
  height = NULL,
  element_id = NULL
)

Arguments

variants

A data frame with columns position (amino-acid position, 1-based) and count (recurrence). Optional class (variant class) and label (e.g. "R175H") columns drive the colour and the text labels.

length

Protein length in residues.

gene, uniprot

Identifiers shown on the axis title.

domains

Optional data frame of domain rectangles with columns name, start and end.

ptms

Optional data frame of modification sites with columns position and type.

classes

Character vector fixing the legend order and colour assignment. Defaults to the classes present, most frequent first.

class_colors, domain_colors

Character vectors of hex colours. NULL uses the component's categorical palette.

label_top_n

Label the n most recurrent variants. Which stems get a label is resolved here and sent to the browser, so a redraw, an export and any static counterpart all label the same ones.

show_ptms, show_domains, show_legend

Toggle the surrounding tracks.

min_head_radius, max_head_radius

Stem head radius range in pixels. Head area is proportional to recurrence, so these bound the mapping rather than setting a size: raise max_head_radius when one hotspot dwarfs the rest and you want the difference to read at a glance.

y_label

Axis title for the recurrence axis.

backbone_color, stem_color

Hex colours for the protein backbone rectangle and the mutation stems.

theme

Optional named list of theme overrides.

width, height

Widget dimensions (any valid CSS size).

element_id

Optional explicit DOM id.

Value

An htmlwidget object.

Details

Stems and domains are canvas-drawn so a protein with thousands of variants stays responsive; labels, axis and legend are a vector overlay.

Examples

v <- data.frame(
  position = c(175, 248, 273),
  count = c(21, 15, 13),
  class = c("Missense", "Missense", "Missense"),
  label = c("R175H", "R248Q", "R273H")
)
d <- data.frame(name = "P53 DNA-binding", start = 100, end = 288)
lollipop(v, length = 393, gene = "TP53", uniprot = "P04637", domains = d)