Would tree lengths change if a character was coded as ambiguous for each leaf Pol2009TreeSearch?
Arguments
- trees
List of trees of class
phylo, ormultiPhyloobject.- char
phyDatobject containing a single character.- concavity
Determines the degree to which extra steps beyond the first are penalized. Specify a numeric value to use implied weighting Goloboff1993TreeSearch;
concavityspecifies k in k / e + k. A value of 10 is recommended; TNT sets a default of 3, but this is too low in some circumstances Goloboff2018,Smith2019TreeSearch. Better still explore the sensitivity of results under a range of concavity values, e.g.k = 2 ^ (1:7). SpecifyInfto weight each additional step equally. Specify"profile"to employ profile parsimony Faith2001TreeSearch.
Value
LengthAdded() returns a named numeric vector listing the mean
absolute change to tree length resulting if the character were coded
ambiguous for each leaf in turn, under the specified concavity constant.
Details
High values for a leaf indicate that its coding contributes to instability ("wildcard" or "roguish" behaviour; see Roguefor further details). The coding is in tension with other data, which may indicate that the assumptions of homology that underlie the character's construction and scoring require careful scrutiny – or that the taxon in question has been subject to convergent evolution.
When inapplicable tokens are present in a character, the applicability of each coding is maintained: i.e. a leaf coded with an applicable token is never allowed to take an inapplicable value; and an inapplicable token remains inapplicable.
See also
Other tree scoring:
CharacterLength(),
IWScore(),
MinimumLength(),
MorphyTreeLength(),
TaxonInfluence()
Examples
trees <- inapplicable.trees[["Vinther2008"]]
dataset <- inapplicable.phyData[["Vinther2008"]]
char <- dataset[, 11]
added <- LengthAdded(trees, char)
PlotCharacter(
tree = trees[[1]],
dataset = char,
tip.color = 1 + added[trees[[1]]$tip.label] # Colour by added steps
) -> XX # Suppress return value; display plot only
