| fortify.isomap | R Documentation |
Prepares fortified versions of results from isomap objects.
## S3 method for class 'isomap'
fortify(model, data = NULL, axes = 1:6, what = c("sites", "network"), ...)
model |
an object of class |
data |
data.frame; additonal data to be added to the forified object. |
axes |
numeric; which axes to return. For |
what |
character; what aspect of the results should be fortified? |
... |
additional arguments for other methods. Currently ignored. |
Two different objects can be created from the results of an isomap object. The first is the standard scores representation of fortified ordinations in vegan, which results in a wide data frame where rows cotain observations and columne the coordinates of observations on the MDS axes. Because ISOMAP also produces a network or sorts, the coordinates of the edges of the network can also be returned in a tidy format using what = "network".
A data frame. For what = "sites", the data frame contains one variable per dimension of the multidimensional scaling embedding of the dissimilarities. Variables are named "DimX" with "X" being an integer. An adidtional variable is label, containing a label for each observation. For what = "network", the data frame contains four variables cotaining the coordinates in the chosen MDS axes for the start and end points of the network edges.
Gavin L. Simpson
library("ggplot2")
data(BCI)
dis <- vegdist(BCI)
ord <- isomap(dis, k = 3)
df <- fortify(ord, data = data.frame(Richness = specnumber(BCI)))
net <- fortify(ord, what = "network", axes = 1:2)
ggplot(df, aes(x = Dim1, y = Dim2)) +
geom_segment(data = net,
aes(x = xfrom, y = yfrom,
xend = xto, yend = yto),
colour = "grey85", size = 0.8) +
geom_point(aes(size = Richness)) +
coord_fixed()
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.