conducting the relative cladogenesis test for all slices through a tree
rc(phy, plot=TRUE, ...)Table of results with four columns: Number of ancestors, Maximum descendents, p-value, Bonferroni-corrected p-value
a phylogenetic tree of class 'phylo'
whether to plot tree with significant branches highlighted
arguments passed for plotting (see plot.phylo)
LJ Harmon
A list of nodes is returned, along with the number of lineages alive just before that node, the maximum number of descendents that any of those lineages has at the present day, a p-value for this observation under the null hypothesis of a birth-death process (that is, given the null, what is the probability that one of these lineages had at least that many descendents), and the p-value after Bonferroni correction (given that a total of n-1 comparisons are made).
If a plot is made, asterisks will mark significantly diverse clades. These asterisks appear just to the right of the MRCA of the diverse clade.
The Bonferroni correction used here is exceedingly conservative for a tree of any reasonable size (and not necessarily recommended, especially given the exploratory nature of this test and the non-independence of the comparisons). Plotting defaults to indicating which nodes are significant without a Bonferroni correction and a P-value of 0.05 as a cutoff (see Examples for modifying this behavior).
One will often see significant results "trickle down" nodes in the tree - that is, if one clade is expecially diverse, then one or more of its parent clades will also be diverse. The most parsimonious place to attribute this effect is to the most shallow significant branch - that is, the branch closest to the tips (see Moore et al. 2004).
Purvis A, S Nee, and PH Harvey. 1995. Proc. R. Soc. London Ser. B 260:329-333.
Moore BR, KMA Chan, and MJ Donoghue. 2004. Detecting diversification rate variation in supertrees. In O.R.P. Bininda-Emonds (ed.), Phylogenetic Supertrees: Combining Information to Reveal the Tree of Life, pp. 487-533. Kluwer Academic, Netherlands:Dordrecht.
geo <- get(data(geospiza))
## WITHOUT BONFERRONI CORRECTION
rc(geo$phy)
## WITH BONFERRONI CORRECTION and ALPHA=0.15
rc(geo$phy, bonf=TRUE, p.cutoff=0.15)
Run the code above in your browser using DataLab