Installing Packages from Neuroconductor

John Muschelli

2017-10-09

Installing Neuroconductor Packages

Install and start the latest release version of R. Although the installer will try to download and install devtools, there may be some system requirements for devtools that you may need before going forward.

Then, you can install a package using the following command:

## try http:// if https:// URLs are supported
source("https://neuroconductor.org/neurocLite.R")
neuro_install("PACKAGE")

where PACKAGE is the name of the package you’d like to install, such as fslr. For example, if we want to install neurohcp and fslr we can run:

source("https://neuroconductor.org/neurocLite.R")
neuro_install(c("fslr", "neurohcp"))

neurocLite: an alias for neuroc_install

As Bioconductor uses the biocLite function to install packages, we have created a duplicate of neuro_install, called neurocLite, for ease of use for those users accustomed to Bioconductor. The same command could have been executed as follows:

source("https://neuroconductor.org/neurocLite.R")
neurocLite(c("fslr", "neurohcp"))

Installing the neurocInstall package

The neurocInstall package contains the neurocLite/neuro_install functions, as well as others relevant for Neuroconductor. You can install the package as follows:

source("https://neuroconductor.org/neurocLite.R")
neuro_install("neurocInstall")

After installation, you can use neurocInstall::neuroc_install() to install packages without source-ing the URL above.

Installing Neuroconductor Packages without upgrading dependencies

The neurocLite/neuro_install functions depend on devtools::install_github, which will upgrade dependencies by default, which is recommended. If you would like to install a package, but not upgrade the dependencies (missing dependencies will still be installed), you can set the upgrade_dependencies argument to FALSE:

neurocLite(c("fslr", "neurohcp"), upgrade_dependencies = FALSE)

Installing and starting R

  1. Download the most recent version of R from https://cran.r-project.org/. There are detailed instructions on the R website as well as the specific R installation for the platform you are using, typically Linux, OSX, and Windows.

  2. Start R; we recommend using R through RStudio. You can start R using RStudio (Windows, OSX, Linux), typing “R” at in a terminal (Linux or OSX), or using the R application either by double-clicking on the R application (Windows and OSX).

  3. For learning R, there are many resources such as Try-R at codeschool and DataCamp.

Packages not available on Neuroconductor

If a package is not in the Neuroconductor list of packages, then it is not located on the Neuroconductor Github. Therefore, when installing, you’ll get the following error:

Error in neuro_install(...) : 
  Package(s) PACKAGE_TRIED_TO_INSTALL are not in neuroconductor

Once a package is located on the list of packages, then it will be available to install.