R script for loading a corpus of community ecology data sets

Here's an R script for loading nine completely different community data sets, each associated with a matrix of environmental data. Most of the data comes from this wonderful website,

# Forest vegetation data from Vltava river valley
vltava.spe <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/vltava-spe.txt", 
    row.names = 1)
vltava.env <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/vltava-env.txt", 
    row.names = 1)

# Vegetation of acidophilous grasslands in Třebíč region
grasslands.spe <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/grasslands-spe.txt", 
    row.names = 1)
grasslands.env <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/grasslands-env.txt")

# Vegetation of Carpathian wetlands
bryo <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/bryophytes.txt", 
    row.names = 1)
vasc <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/vasc_plants.txt", 
    row.names = 1)
chem <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/chemistry.txt", 
    row.names = 1)

# Vegetation data from Bryce Canyon (Utah, USA)
library(labdsv)
data(bryceveg)  # matrix with species data (160 samples in rows and 169 species in columns)
data(brycesite)  # matix of environmental variables (160 samples in rows and 11 environmental variables in columns)

# Experimental data from wet meadow
ohrazeni.spe <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/ohrazeni-spe.txt", 
    row.names = 1)
ohrazeni.env <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/ohrazeni-env.txt")

# Tropical forest permanent plot in Barro Colorado Island (Panama)
# install.packages(vegan) install.packages(BiodiversityR)
library(vegan)
data(BCI)
library(BiodiversityR)
data(BCI.env)

# Community of coral reefs around Tikus island, Indonesia
tikus.spe <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/tikus-spe.txt", 
    row.names = 1)
tikus.env <- read.delim("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/tikus-env.txt")

# Doubs river fish communities
doubs.spe <- read.csv("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/DoubsSpe.csv", 
    row.names = 1)
doubs.env <- read.csv("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/DoubsEnv.csv", 
    row.names = 1)
doubs.spa <- read.csv("http://www.sci.muni.cz/botany/zeleny/wiki/anadat-r/data-download/DoubsSpa.csv", 
    row.names = 1)

# Oribatid mites
library(vegan)
data(mite)
data(mite.env)

Code for getting info about the data

We begin by making a vector with the names of the data sets, so we can refer to all the data simultaneously,

does.char.refer.to.data <- function(x) class(eval(parse(text = x))) == "data.frame"
data.objects <- ls()[sapply(ls(), does.char.refer.to.data)]

Here's a function that can be used to do the same thing to every dataset,

#' Character apply
#'
#' Pass several objects through a function where the objects
#' are refered to by character strings.
#'
#' @param objects Character vector with object names
#' @param FUN Function to apply
#' @param ... Additional arguments to pass to FUN
capply <- function(objects, FUN, simplify = TRUE, ...) {

    # differentiate between names (obj.names) and strings (objects) (currently
    # only necessary for naming the elements of the output)
    obj.names <- lapply(objects, as.name)

    # apply the function
    out <- lapply(obj.names, function(xx) FUN(eval(xx), ...))

    # put characters back in the names attribute
    names(out) <- objects

    if (simplify) 
        out <- simplify2array(out)
    return(out)
}

Here are some examples of getting info,

# capply(data.objects, dimnames, simplify = FALSE) # prints way too much!
capply(data.objects, dim)  # gives you the size of the data

##      BCI BCI.env brycesite bryceveg bryo chem doubs.env doubs.spa
## [1,]  50      50       160      160   70   70        30        30
## [2,] 225       6        11      169   30   15        11         2
##      doubs.spe grasslands.env grasslands.spe mite mite.env ohrazeni.env
## [1,]        30             48             48   70       70           96
## [2,]        27             16            171   35        5           33
##      ohrazeni.spe tikus.env tikus.spe vasc vltava.env vltava.spe
## [1,]           96        60        60   70         97         97
## [2,]           86         2        75  123         24        342