RoguePlot {TreeTools}

Visualize position of rogue taxa

Description

Plots a consensus of trees with a rogue taxon omitted, with edges coloured according to the proportion of trees in which the taxon attaches to that edge, after Klopfstein and Spasojevic (2019).

Usage

RoguePlot(
  trees,
  tip,
  p = 1,
  plot = TRUE,
  Palette = colorRampPalette(c(par("fg"), "#009E73"), space = "Lab"),
  nullCol = rgb(colorRamp(unlist(par(c("fg", "bg"))), space = "Lab")(0.8)/255),
  edgeLength = NULL,
  thin = par("lwd"),
  fat = thin + 1L,
  outgroupTips,
  sort = FALSE,
  legend = "none",
  legend.inset = 0,
  ...
)

Arguments

trees

List or multiPhylo object containing phylogenetic trees of class phylo to be summarized.

tip

Numeric or character identifying rogue leaf, in format accepted by DropTip().

p

A numeric value between 0.5 and 1 giving the proportion for a clade to be represented in the consensus tree (see Consensus()).

plot

Logical specifying whether to plot the tree.

Palette

Function that takes a parameter n and generates a colour palette with n entries.

nullCol

Colour to paint regions of the tree on which the rogue is never found.

edgeLength

Numeric specifying edge lengths of consensus tree; NULL aligns tips by scaling edges proportional to clade size; 1 sets all edges to unit length.

thin, fat

Numeric specifying width to plot edges if the rogue tip never / sometimes does attach to them.

outgroupTips

Vector of type character, integer or logical, specifying the names or indices of the tips to include in the outgroup. If outgroupTips is a of type character, and a tree contains multiple tips with a matching label, the first will be used.

sort

Logical specifying whether to sort consensus tree using SortTree().

legend

Character vector specifying position of legend (e.g. "bottomleft"), or "none" to suppress legend. For fine-grained control of legend, use PlotTools::SpectrumLegend().

legend.inset

Numeric specifying fraction of plot width / height by which the legend's position should be inset.

...

Additional parameters to plot.phylo().

Details

Rogue taxa can be identified using the package Rogue (Smith 2022).

Value

RoguePlot() invisibly returns a list whose elements are:

• cons: The reduced consensus tree, in preorder;

• onEdge: a vector of integers specifying the number of trees in trees in which the rogue leaf is attached to each edge in turn of the consensus tree;

• atNode: a vector of integers specifying the number of trees in trees in which the rogue leaf is attached to an edge collapsed into each node of the consensus tree.

• legendLabels: A character vector suggesting labels for a plot legend; suitable for PlotTools::SpectrumLegend(legend = x$legendLabels).

Author(s)

Martin R. Smith ([e-mail])

References

Klopfstein S, Spasojevic T (2019). “Illustrating phylogenetic placement of fossils using RoguePlots: An example from ichneumonid parasitoid wasps (Hymenoptera, Ichneumonidae) and an extensive morphological matrix.” PLOS ONE, 14(4), e0212942. doi:10.1371/journal.pone.0212942.

Smith MR (2022). “Using information theory to detect rogue taxa and improve consensus trees.” Systematic Biology, 71(5), 986–1008. doi:10.1093/sysbio/syab099.

See Also

Other consensus tree functions: Consensus(), ConsensusWithout()

Examples

trees <- list(read.tree(text = "(a, (b, (c, (rogue, (d, (e, f))))));"),
              read.tree(text = "(a, (b, (c, (rogue, (d, (e, f))))));"),
              read.tree(text = "(a, (b, (c, (rogue, (d, (e, f))))));"),
              read.tree(text = "(a, (b, (c, (rogue, (d, (e, f))))));"),
              read.tree(text = "(rogue, (a, (b, (c, (d, (e, f))))));"),
              read.tree(text = "((rogue, a), (b, (c, (d, (e, f)))));"),
              read.tree(text = "(a, (b, ((c, d), (rogue, (e, f)))));"),
              read.tree(text = "(a, (b, ((c, (rogue, d)), (e, f))));"),
              read.tree(text = "(a, (b, (c, (d, (rogue, (e, f))))));"))
plotted <- RoguePlot(trees, "rogue", legend = "topleft", legend.inset = 0.02)
PlotTools::SpectrumLegend(
  "bottomleft",
  palette = colorRampPalette(c(par("fg"), "#009E73"), space = "Lab")(100),
  legend = plotted$legendLabels,
  cex = 0.4
)