read.multitable: Multiple table data input

In stevencarlislewalker/multitable: Simultaneous Manipulation of Multiple Arrays of Data

Description

These functions are used to read several text files and create a data list from them.

Usage

read.multitable(files, dimids, fill = rep(NA, length(files)), ...)

read.multicsv(files, dimids, fill = rep(NA, length(files)), ...)

read.multidelim(files, dimids, fill = rep(NA, length(files)), ...)

read.fourthcorner(community, environment, traits, dimids=c("sites", "species"), 
community.name = "abundance", ...)

multifile.choose(n)

read.matrix(...)

Arguments

files	A character vector with the names of the files containing the tables (possibly created with multifile.choose).
community	A character string of the name of the file containing the community data of a fourth-corner problem.
environment	A character string of the name of the file containing the environment data of a fourth-corner problem.
traits	A character string of the name of the file containing the trait data of a fourth-corner problem.
dimids	A vector with the names of the columns associated with replication dimensions. For read.fourthcorner this is simply the names of the replication dimensions.
fill	See dlcast.
community.name	Character string of the name of community matrix.
n	Number of files to choose.
...	Additional arguments to pass to read.table. The most important such argument is sep, which specifies how entries are separated in the data files. Note that headers MUST be present in order to distinguish columns for replication dimensions from columns for variables, and therefore header is set to TRUE and cannot be changed. Exception: header may be missing in read.fourthcorner.

Details

The read.multitable function is a multiple-table version of read.table. It is largely a wrapper for dlcast, so the tables that are read need to produce ‘long’ format data frames. The implementation of read.multitable is very simple: (1) repeatedly call read.table to load in the files with names in files and then (2) call dlcast to combine these tables into a data list. Therefore, the dimids and fill arguments are simply passed to dlcast.

The read.multicsv and read.multidelim simply wrap read.multitable with the appropriate value for the sep argument (compare with read.csv for example).

The read.fourthcorner function reads in three files (with names passed to community, environment, and traits), and creates a data list out of the resulting files. The community table is coerced to a matrix so that the rows and columns are treated as two different dimensions of replication, whereas the other two tables are left as data frames. This function is useful because many community ecologists will store their data in this way.

The multifile.choose function allows the interactive selection of the names of n files. Compare with file.choose.

The read.matrix function is identical to read.table but returns a matrix instead of a data frame. The definition of this function is simply a call to read.table wrapped in a call to as.matrix.

Value

A data.list object.

Examples

abundance.file <- tempfile()
environment.file <- tempfile()
trait.file <- tempfile()

abundance <- data.frame(
	sites=c(
		"midlatitude","subtropical","tropical","equatorial",
		"arctic","midlatitude","tropical","equatorial",
		"subtropical"
	),
	species=c(rep("capybara",4),rep("moss",4),"vampire"),
	abundance=c(4,10,8,7,5,6,9,3,1)
)

environment <- data.frame(
	sites=c(
		"arctic","subarctic","midlatitude","subtropical",
		"tropical","equatorial"
	),
	temperature=c(-30,0,10,20,50,30),
	precipitation=c(20,40,20,100,150,200)
)

trait <- data.frame(
	species=c("capybara","moss","vampire"),
	body.size=c(140,5,190),
	metabolic.rate=c(20,5,0)
)

write.table(abundance,abundance.file,sep=",")
write.table(environment,environment.file,sep=",")
write.table(trait,trait.file,sep=",")

files <- c(abundance.file,environment.file,trait.file)
dimids <- c("sites","species")
read.multicsv(files,dimids,fill=c(0,NA,NA))

## Modifications necessary to use the read.fourthcorner function
abundance <- data.frame(
	capybara = c(4,10,8,7,0,0),
	moss = c(6,0,9,3,5,0),
	vampire = c(0,1,0,0,0,0),
	row.names = c(
		"arctic","subarctic","midlatitude","subtropical",
		"tropical","equatorial"
	)
)
write.table(abundance,abundance.file,sep=",")
read.fourthcorner(abundance.file,environment.file,trait.file,sep=",")

stevencarlislewalker/multitable documentation built on May 30, 2019, 4:44 p.m.