3. How to draw a pedigree

1.1 A simple pedigree graph

The trimmed simple_ped pedigree is drawn and displayed on the default graphics device. The addgen and addnum parameters need to be set to TRUE when tidying the pedigree using the tidyped function.

tidy_simple_ped <- tidyped(simple_ped)
visped(tidy_simple_ped, cex=0.3, symbolsize=10)
#> Label cex: 0.3. Symbol size: 10. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

Generation labels can be added on the left margin of the pedigree graph by setting genlab = TRUE. This is useful for deep pedigrees when you want to quickly identify the generation of each row.

visped(tidy_simple_ped, cex = 0.3, symbolsize = 10, genlab = TRUE)
#> Label cex: 0.3. Symbol size: 10. Generation label cex: 0.909090909. Adjust 'cex', 'symbolsize', and 'genlabcex' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

Instead of the default G1, G2, … labels, an unnamed character vector can provide one custom label for each displayed generation, from top to bottom. This is useful when the plotted pedigree layers have an application-specific interpretation. The labels are not inferred from a birth-year column, which may not correspond one-to-one with pedigree generations in overlapping cohorts.

generation_labels <- paste0("Gen_", sort(unique(tidy_simple_ped$Gen)))
visped(tidy_simple_ped, cex = 0.3, symbolsize = 10, genlab = generation_labels, genlabcex=0.6)
#> Label cex: 0.3. Symbol size: 10. Generation label cex: 0.6. Adjust 'cex', 'symbolsize', and 'genlabcex' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

In deep pedigrees the generation labels may appear small because their size is tied to the node scaling. The genlabcex parameter lets you set the generation-label size independently of the individual-label size (cex). For example, when cex is deliberately small to fit many individuals, you can still keep the generation labels readable:

# cex controls individual label size; genlabcex controls generation label size
visped(tidy_simple_ped, cex = 0.3, symbolsize = 15, genlab = TRUE, genlabcex = 0.8)
#> Label cex: 0.3. Symbol size: 15. Generation label cex: 0.8. Adjust 'cex', 'symbolsize', and 'genlabcex' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

Figures displayed in the RStudio Plots panel often have limited resolution. Individual IDs may overlap if the pedigree is large and the plot area is restricted. This can be resolved by saving the graph as a vector graphic in a PDF file. The visped function will suppress output to the default device if showgraph = FALSE.

suppressMessages(visped(tidy_simple_ped, showgraph = FALSE, file = tempfile(fileext = ".pdf")))

By setting the file parameter, you can generate a high-definition PDF version of the pedigree.

SVG output is also supported by using a file name ending in .svg. This is useful when the pedigree graph needs to be edited in vector-graphics software or embedded in web documents.

suppressMessages(visped(tidy_simple_ped, showgraph = FALSE, file = tempfile(fileext = ".svg")))

Custom node labels can be displayed either from a selected pedigree column or from a character vector with one value per input row using labelvar. Row-aligned vectors remain attached to the correct individuals if the pedigree is internally tidied or reordered. Both forms keep the underlying individual IDs available for tracing and highlighting, while changing the text shown on individual nodes. Compact full-sib family nodes still use the explicit FS×N family-size label when compact = TRUE.

tidy_simple_ped[, ShortLabel := paste0("N", seq_len(.N))]
visped(tidy_simple_ped, labelvar = "ShortLabel", cex = 0.3, symbolsize = 10)
#> Label cex: 0.3. Symbol size: 10. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

# A row-aligned vector is also accepted
visped(tidy_simple_ped, labelvar = rep("1x", nrow(tidy_simple_ped)),
       cex = 0.3, symbolsize = 10)
#> Label cex: 0.3. Symbol size: 10. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

visped(tidyped(small_ped), highlight = c("Y", "Z1"))

#> Label cex: 0.65. Symbol size: 1. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

# Highlight individual "Y" and all its ancestors and descendants
visped(tidyped(small_ped), highlight = "Y", trace = TRUE)

#> Label cex: 0.65. Symbol size: 1. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

visped(tidyped(small_ped), 
       highlight = list(ids = c("Y", "Z1"), 
                        frame.color = "#4caf50", 
                        color = "#81c784"))

#> Label cex: 0.65. Symbol size: 1. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.

1.1.2 Showing inbreeding coefficients

Inbreeding coefficients can be displayed on the pedigree graph using the showf parameter in the visped function. This requires that the pedigree has been processed with inbreeding coefficients calculated using the inbreed parameter in the tidyped function.

library(data.table)
test_ped <- data.table(
  Ind = c("A", "B", "C", "D", "E"),
  Sire = c(NA, NA, "A", "C", "C"),
  Dam = c(NA, NA, "B", "B", "D"),
  Sex = c("male", "female", "male", "female", "male")
)
tidy_test_ped_inbreed <- tidyped(test_ped, inbreed = TRUE)
visped(tidy_test_ped_inbreed, showf = TRUE)

#> Label cex: 0.65. Symbol size: 1. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.
#> Note: Inbreeding coefficients of 0 are not shown in the graph.

1.1.3 Drawing a plant pedigree with monoecious individuals

In plant breeding, many crop species are monoecious — a single plant produces both male and female gametes, enabling self-fertilization (selfing). When a plant serves as both sire and dam, tidyped() with selfing = TRUE assigns it Sex = "monoecious". In the graph, monoecious individuals appear as hexagons with teal edges and fill, and selfing edges (sire = dam) are drawn in teal.

The following example shows a multi-generation self-pollinating pedigree with inbreeding coefficients displayed:

library(data.table)

# P1 × P2 → F1;  F1 selfs → F2_1, F2_2;  F2_1 selfs → F3
plant_ped <- data.table(
  Ind  = c("P1", "P2", "F1", "F2_1", "F2_2", "F3"),
  Sire = c(NA,   NA,   "P1", "F1",   "F1",   "F2_1"),
  Dam  = c(NA,   NA,   "P2", "F1",   "F1",   "F2_1")
)

tp_plant <- tidyped(plant_ped, selfing = TRUE, inbreed = TRUE)
#> Selfing mode: 2 individual(s) appear as both Sire and Dam: F1, F2_1. These will be assigned Sex = 'monoecious'.

visped(tp_plant, compact = TRUE, showf = TRUE, cex=0.5, symbolsize=10)

#> Label cex: 0.5. Symbol size: 10. Adjust 'cex' and 'symbolsize' if labels are too large or small.
#> Tip: Use 'file' to save as a legible vector PDF or SVG.
#> Note: Inbreeding coefficients of 0 are not shown in the graph.

Self-fertilization produces rapid inbreeding: f = 0.5 after one generation of selfing (F2_1, F2_2), and f = 0.75 after two (F3). F1 and F2_1 are drawn as hexagons because they act as both sire and dam; selfing edges are shown in teal. For the general shape and color legend, see the introduction of Section 1.

See vignette("tidy-pedigree", package = "visPedigree"), section 3.9, for preparing plant pedigrees and calculating inbreeding coefficients.

1.2 A reduced pedigree graph

Warning messages will be shown when you try to draw the pedigree graph of the deep_ped dataset.

cand_J11_labels <- deep_ped[(substr(Ind, 1, 3) == "K11"), Ind]
visped(
  tidyped(deep_ped,
    cand = cand_J11_labels,
    tracegen = 3),
  cex=0.08,
  symbolsize=5.5
  )

  Too many individuals (>=3362) in one generation!!! Two choices:
1. Removing full-sib individuals using the parameter compact = TRUE; or, 
2. Visualizing all nodes without labels using the parameter outline = TRUE.
Rerun visped() function!

The function indicates that there are too many individuals in a single generation to draw a standard pedigree graph. It is recommended to use the compact or outline parameters to simplify the pedigree.

First, let’s try the compact parameter and output it to a PDF file. The plot on the default device may suffer from significant overlapping due to the high density of individuals.

cand_J11_labels <- deep_ped[(substr(Ind,1,3) == "K11"),Ind]
visped(
  tidyped(
    deep_ped,
    cand = cand_J11_labels,
    trace = "up",
    tracegen = 3
  ),
  cex=0.08, 
  symbolsize=5.5,
  compact = TRUE,
  showgraph = TRUE,
  file = tempfile(fileext = ".pdf")
)
#> Pedigree saved to: /private/var/folders/rz/zk22mcsx26lc9jy74shpdfwr0000gn/T/RtmpVfrc8A/file134ba625f5feb.pdf
#> Label cex: 0.08. Symbol size: 5.5. Adjust 'cex' and 'symbolsize' if labels are too large or small.

You can open the generated PDF file to view the high-definition pedigree vector graph. Most rectangles at the bottom are compact family summaries, and labels such as FS×10 report the number of represented full-sib individuals. The summary is sex-neutral because a compacted family may contain individuals of different or unknown sex. Individual labels may be shorter than their shapes and might not align perfectly. Labels can be resized using cex; increasing it makes labels larger, while decreasing it makes them smaller. The value typically ranges from 0 to 1, but can be larger, and adjustments of 0.1 are usually sufficient. The visped function reports the cex value used to draw the graph.

visped(
  tidyped(
    deep_ped,
    cand = cand_J11_labels,
    trace = "up",
    tracegen = 3
  ),
  compact = TRUE,
  cex = 0.83,
  showgraph = FALSE,
  file = tempfile(fileext = ".pdf")
)
#> Pedigree saved to: /private/var/folders/rz/zk22mcsx26lc9jy74shpdfwr0000gn/T/RtmpVfrc8A/file134ba1e6392d6.pdf
#> Label cex: 0.83. Symbol size: 1. Adjust 'cex' and 'symbolsize' if labels are too large or small.

You can open the generated PDF file to view the high-definition pedigree vectorgraph. The labels align better with the shapes compared to the previous version. You can continue to adjust cex until the labels are sized appropriately.

1.3 An outlined pedigree graph

Setting outline = TRUE produces an outlined pedigree graph. Individual labels will not be shown in the graph. This is highly effective for large pedigrees with many individuals.

The following code generates an outlined pedigree graph in a PDF file.

suppressMessages(visped(
  tidyped(
    deep_ped,
    cand = cand_J11_labels, 
    tracegen = 3),
  compact = TRUE,
  outline = TRUE,
  showgraph = TRUE,
  file = tempfile(fileext = ".pdf")
))

1.4 How to use this package in a selective breeding program

suppressWarnings(
  K110550H_ped <- tidyped(deep_ped, cand = "K110550H")
)
if(interactive()) {
  visped(K110550H_ped, compact = TRUE)
}

cand_2007_G8_labels <-
  big_family_size_ped[(Year == 2007) & (substr(Ind, 1, 2) == "G8"), Ind]
suppressWarnings(
  cand_2007_G8_tidy_ped_ancestor_2 <-
    tidyped(
      big_family_size_ped,
      cand = cand_2007_G8_labels,
      trace = "up",
      tracegen = 2
    )
)
sire_label <-
  unique(cand_2007_G8_tidy_ped_ancestor_2[Ind %in% cand_2007_G8_labels,
                                          Sire])
dam_label <-
  unique(cand_2007_G8_tidy_ped_ancestor_2[Ind %in% cand_2007_G8_labels,
                                          Dam])
sire_dam_label <- unique(c(sire_label, dam_label))
sire_dam_label <- sire_dam_label[!is.na(sire_dam_label)]
sire_dam_ped <-
  cand_2007_G8_tidy_ped_ancestor_2[Ind %in% sire_dam_label]
#> Warning: Subsetting removed parent records. Result is a plain data.table, not a tidyped.
#> Use tidyped(tp, cand = ids, trace = "up") to extract a valid sub-pedigree.
sire_dam_ped <-
  sire_dam_ped[, FamilyID := paste(Sire, Dam, sep = "")]
family_size <- sire_dam_ped[, .N, by = c("FamilyID")]
fullsib_family_label <- unique(sire_dam_ped$FamilyID)
suppressMessages(
  visped(
    cand_2007_G8_tidy_ped_ancestor_2,
    compact = TRUE,
    outline = TRUE,
    showgraph = TRUE
  )
)