Skip to contents

Custom styling and export

Source: vignettes/v08_custom_styling_and_export.Rmd
v08_custom_styling_and_export.Rmd

Custom styling and export

vennDiagramLab separates analysis from rendering. Once you have a RegionResult, you can render it with custom names, custom colors, post-process the SVG, embed it in a ggplot2 chain via geom_venn(), or export to PNG / PDF.

library(vennDiagramLab)
result <- analyze(load_sample("dataset_real_cancer_drivers_4"))

Custom names

Pass a per-letter mapping (A-I) to override the dataset’s set names:

svg <- render_venn_svg(
    result,
    set_names = c(A = "Vogelstein\n(2013)",
                   B = "COSMIC CGC",
                   C = "OncoKB",
                   D = "IntOGen"),
    title = "Pan-source cancer driver agreement"
)
substr(svg, 1, 60)
#> [1] "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<!-- Created by Zolta"

Custom colors

Pass a per-letter hex map. Each letter’s color is applied to the matching shape AND the legend bullet (and, where present, the Euler extra shape).

svg <- render_venn_svg(
    result,
    colors = c(A = "#E69F00", B = "#56B4E9", C = "#009E73", D = "#CC79A7")
)
nchar(svg)
#> [1] 6464

Hide the count labels 

svg_clean <- render_venn_svg(result, show_counts = FALSE)
nchar(svg_clean)
#> [1] 6425

(show_names = FALSE does the analogous thing for set names.)

Post-render SVG manipulation with xml2

The returned SVG is a plain string; parse it with xml2 to make targeted edits (e.g. set the page background or add a watermark):

svg <- render_venn_svg(result)
doc <- xml2::read_xml(svg)
xml2::xml_attr(doc, "viewBox")
#> [1] "0 0 700 700"

Embed in a ggplot2 chain

geom_venn() returns a list of layers that draws the venn on a unit-square coordinate system, ready to compose with titles, themes, and other annotations.

library(ggplot2)
ggplot() +
    geom_venn(data = result) +
    theme_void() +
    labs(title = "4 cancer-driver sources",
          subtitle = "Vogelstein, COSMIC CGC, OncoKB, IntOGen") +
    theme(plot.title = element_text(size = 14, face = "bold"),
          plot.subtitle = element_text(size = 10, colour = "grey40"))

Multi-format export

render_venn_svg() returns a string. Convert it to PNG or PDF via the rsvg package (already a hard import of vennDiagramLab):

svg <- render_venn_svg(result)
png_path <- tempfile(fileext = ".png")
rsvg::rsvg_png(charToRaw(svg), png_path, width = 1200)
file.size(png_path)
#> [1] 206202

pdf_path <- tempfile(fileext = ".pdf")
rsvg::rsvg_pdf(charToRaw(svg), pdf_path)
file.size(pdf_path)
#> [1] 31926

For a multi-page composite report (venn + upset + statistics + network + about), use to_pdf_report() — see vignette("v07_pdf_reports").

What’s next