CLAG R package

CLAG (for CLusters AGgregation) is an unsupervised non hierarchical clustering algorithm originally described in this article: CLAG: an unsupervised non hierarchical clustering algorithm handling biological data, Linda Dib and Alessandra Carbone, BMC Bioinformatics 2012, 13:194.

This website provides a CLAG package for the R programming language that allows for easier use of this method than the original set of Perl scripts. It provides a function (CLAG.clust) which takes as input a matrix and returns the clustering, like standard R function do (hclust, kmeans, etc.)

How to install it

Linux and Mac OS X

In R (at least version 2.14), run:

tmp <- tempfile(fileext=".tar.gz")
download.file("http://clag.r-forge.r-project.org/CLAG.tar.gz", tmp)
install.packages(tmp, repos=NULL, type="source")

Once installed, you can type library(CLAG) to load it. It should say "CLAG ready".

Windows

As CLAG is written in the Perl programming language, you first need to install Perl if you haven't done it already. Perl can be downloaded for free here: http://strawberryperl.com/. Simply download the installer and install with all the default options.

Then, click here to download the CLAG package. After that, in R, run: 

install.packages(file.choose(), repos=NULL, type="source")

and select the file which you just downloaded.

Once installed, you can type library(CLAG) to load it. It should say "CLAG ready".

How to use it

We provide a tutorial (PDF).

You can get detailed help with the regular help function in R like this:

library(CLAG)
help(CLAG.clust)

Links

Authors

The CLAG algorithm was designed by Linda Dib and Alessandra Carbone, while the R package was created by Raphaël Champeimont.

This work was done at the Laboratory "Génomique des Microorganismes", UMR7238 CNRS-Université Pierre et Marie Curie (Paris 6).

CLAG is open source and can be redistributed under the terms of the BSD license or the GPL license.

Last Update May 2013

UPMC LGM CNRS