octopucs {octopucs}

Support for Hierarchical Clusters

Description

The global function of the "octopucs" package is to estimate support for hierarchical clusters throughout the dendrogram's topology, considering their repeatability, integrity, and contamination based on bootstrapping procedures.

Usage

octopucs(data, n=NULL, prop=NULL, resample = TRUE, distance =NULL, 
method=NULL, orientation=NULL, disp = NULL, lab.pos=NULL, 
sep.lab=NULL, xlim=NULL, ylim=NULL, verbose=NULL)

Arguments

data

A matrix whose rows (sites/conditions) will be grouped based on the columns (species/elements).

n

The number of bootstrap repetitions.

prop

Proportion of information used in the bootstrapping procedures. When matrices contain more than 18 columns, it refers to (i) the proportion of columns used in every run to produce a new grouping array when "resample=TRUE" and (ii) the number of cells in the body of the matrix (1–prop; e.g., 1–0.8=0.2) that will be replaced with column averages when "resample=FALSE". When matrices contain less than 18 columns, the procedure with "resample=FALSE" will be forced. In such a scenario, "prop" corresponds to the number of cells in the body of the matrix (1–prop; e.g., 1–0.8=0.2) that will be replaced in the bootstrapping procedures.

resample

When TRUE, "octopucs" will select random subsets of columns in every bootstrap run to produce a new grouping array. When FALSE, "octopucs" will replace cells in the body of the matrix with column averages instead of entire columns (see description for "prop" for further details).

distance

Distance method to be used (default="bray"). Options include those of the function "vegdist" ("vegan"): "manhattan", "euclidean", "canberra", "bray", "kulczynski", "gower", "morisita", "horn", "mountford", "jaccard", "raup", "binomial", "chao", "altGower", "cao" and "mahalanobis", as well as those referred in Koleff et al. (2003), available either by name or number: 1 or "w", 2 or "-1", 3 or "c", 4 or "wb", 5 or "r", 6 or "I", 7 or "e", 8 or "t", 9 or "me", 10 or "j", 11 or "sor", 12 or "m", 13 or "-2", 14 or "co", 15 or "cc", 16 or "g", 17 or "-3", 18 or "1", 19 or "", 20 or "hk", 21 or "rlb", 22 or "sim", 23 or "gl", 24 or "z". For open distance metrics (not restricted to zero-one ranges), the axis is re-scaled to cover a zero-one range. If so, the axis will be labeled a "Similarity (rescaled)"; if no re-scaling is necessary, the label will appear as "Similarity".

method

Agglomerative procedure for hierarchical clustering, including: "average" (=UPGMA), "ward.D", "ward.D2", "single", "complete", "mcquitty" (=WPGMA), "median" (=WPGMC) or "centroid" (=UPGMC).

orientation

Orientation of the plot (default="r": right; metric in the x-axis is similarity). "l" = left-wise orientation (metric in the x-axis is dissimilarity). "d" = down-wise orientation (metric in the y-axis is dissimilarity).

disp

Result display in the dendrogram (default="sig"; only supported nodes will populated with given results). "all" = all nodes will be populated with support metrics, with significantly supported nodes in blue/red and non-significantly supported nodes in black.

lab.pos

This argument modifies the position of the labels relative to the tips of the dendrogram (sites/conditions).

sep.lab

This argument modifies the position of supporting values from the nodes and lines in the dendrogram.

xlim

Limits of the x-axis.

ylim

Limits of the y-axis.

verbose

TRUE = display all relevant setup values used in the analysis. FALSE = display only the dendrogram with marked significantly grouped clusters.

Value

"octopucs" generates a list object that contains:

$gcluster

hclust output of the generalized clustered dendrogram.

$support

Table with the support value for every group and subgroup across the topology of the generalized clustered dendrogram.

$grps

List with the memberships of every group and subgroup in the topology.

Author(s)

Roger Guevara ([e-mail]) & Ian MacGregor-Fors ([e-mail])

References

Koleff, P., Gaston, K.J., Lennon, J. J. 2003. Measuring beta diversity for presence-absence data. Journal of Animal Ecology 71:367. https://doi.org/10.1046/j.1365-2656.2003.00710.x

Examples

data(dune)
octopucs(dune, n=100, prop=0.8,
resample=TRUE,distance="bray",
method="average", disp = "sig",
verbose=TRUE)