limmaGUI          

Introduction

limmaGUI (linear models for microarrays Graphical User Interface) is for two-color microarrays only.
If you have single-channel Affymetrix data, you may wish to try affylmGUI.

limmaGUI Website Links

Main Webpage Contents

  1. Citation for limmaGUI
  2. Installing limmaGUI
    1. Download limmaGUI as an R package
    2. Windows Installation Notes
    3. Linux/Unix Installation Notes
    4. MacOS X Installation Notes
  3. Running limmaGUI
  4. Tcl/Tk Requirements
  5. R Packages Required

1. Citation for limmaGUI

The methodology behind limmaGUI/LIMMA differential expression analysis can be cited by:

  • Smyth, G. K. (2004). Linear models and empirical Bayes methods for assessing differential expression in microarray experiments. Statistical Applications in Genetics and Molecular Biology 3, No. 1, Article 3. (Full Text)
  • Smyth, G. K., Michaud, J., and Scott, H. (2005). The use of within-array replicate spots for assessing differential expression in microarray experiments. Bioinformatics 21(9), 2067-2075.

The LIMMA software package is described in:

  • Smyth, G. K. (2005). Limma: linear models for microarray data. In: Bioinformatics and Computational Biology Solutions using R and Bioconductor, R. Gentleman, V. Carey, S. Dudoit, R. Irizarry, W. Huber (eds.), Springer, New York, pages 397-420. (Published 8 August 2005, Publisher web site, PDF)

The limmaGUI software package is described in:

  • Wettenhall, J. M., and Smyth, G. K. (2004). limmaGUI: a graphical user interface for linear modeling of microarray data. In: Bioinformatics, 20:3705–3706, 2004., PDF)

2. Installing limmaGUI

Unless you are using the windows installation wizard, you should install the other required R packages before installing limmaGUI.
The other R packages required are listed at the bottom of this webpage.

2.1 Download limmaGUI as an R package

The latest version of limmaGUI is 1.8.1, for R 2.3.1 limmaGUI is distributed as open-source software under the LGPL license.

2.2 Windows Installation Notes (if not using the wizard)

An R 2.1.1 limmaGUI installation wizard (last update: July 4, 2005) is available. This will give you R, limmaGUI, affylmGUI and all of the required R packages and Tcl/Tk extensions. It will also give you some other Bioconductor packages recommended for use with limmaGUI and affylmGUI. If you use the wizard, you can ignore any warnings about ActiveTcl not being found in the Windows Registry.

If you have R already, you may not want to use the installation wizard. Instead, you can install limmaGUI as an R package by starting RGui and pasting in the following command:

   install.packages("limmaGUI",contriburl="http://bioinf.wehi.edu.au/limmaGUI")

Other R packages can be installed from the Packages pull-down menu. Most of the R packages required can be found on CRAN, the Comprehensive R Archive Network, or at Bioconductor

Download R 2.3.1 for Windows.

Download ActiveTcl 8.4.13 for Windows from ActiveState.

2.3 Linux/Unix:

Run R and paste in the following command. If it fails, check your R library paths using .libPaths() and if necessary use
.libPaths(MyRlibDir) to use a directory you have permission to write to.

  install.packages("limmaGUI",contriburl="http://bioinf.wehi.edu.au/limmaGUI")

2.4 MacOS X (with X11):

Run R and paste in the following command. If it fails, check your R library paths using .libPaths() and if necessary use
.libPaths(MyRlibDir) to use a directory you have permission to write to. 

  install.packages("limmaGUI",contriburl="http://bioinf.wehi.edu.au/limmaGUI",type="macbinary")

Eventually RAqua and TkAqua will be the best solution, but for now, it is best to run R and Tcl/Tk in X11. Install R from http://cran.r-project.org, X11 from here and Tcl/Tk from http://www.tcl.tk/, following the instructions in the R for MacOS X FAQ.

After installing Tcl/Tk for X11 on MacOS X you need to install the Tktable extension. It is available in binary form. (do not use the version from TclTkAquaBI, because that is an Aqua version, which is not compatible with X11). Here is a zipped-up binary version of Tktable2.9 built for MacOS X with X11. From the R prompt, you can type library(tcltk) then tclvalue("auto_path") to see your Tcl/Tk search path. The X11-compatible Tcl/Tk directories will generally be in /usr/local/lib, or wherever you installed them when you followed the instructions from the R for MacOS X FAQ. The Aqua-compatible (but X11-incompatible) Tcl/Tk directories will generally be in /Library/Tcl/.

3. Running limmaGUI

Start up R (or RGui), load the limmaGUI package with library(limmaGUI), and then type limmaGUI(). If in Windows, minimize your main RGui window.

If you are using Windows, you can run limmaGUI via a batch file, rather than starting up RGui every time. Here is a sample batch file, limmaGUI.bat and a short R script, runLimmaGUI.R. The path to runLimmaGUI.R in limmaGUI.bat needs to be modified depending on where you put runLimmaGUI.R. You can create a shortcut to limmaGUI.bat, choosing "run minimized" and you can even assign this icon to the shortcut if you like.

4. Tcl/Tk Requirements

limmaGUI uses the BWidget package and the TkTable package, neither of which come bundled with the default installation of R 2.x.y.

If using Windows (and not using the installation wizard), you need to install ActiveTcl 8.4.x.y from http://aspn.activestate.com/ASPN/Downloads/ActiveTcl/.

If using Red Hat Linux, install the Tcl and Tk 8.3 rpms from CD, e.g. into /usr/lib, then download BWidget from http://sourceforge.net/projects/tcllib/ and gunzip it and untar it into /usr/lib, and install Tktable from http://tktable.sourceforget.net, gunzip it and untar it, and run configure --prefix=/usr; make; make install, assuming that your Tcl and Tk rpms were installed in /usr/lib.

If using MacOS X, install X11 TclTk (MacOS X) (Preferably use customize when installing R and select the Tcl/Tk option).

Place the Tktable2.9 folder and the BWidget folder where Tcl/Tk will find it. This is determined by looking at the "auto_path" tcl variable. To do this, load tcltk (using library(tcltk) ) and type at the R prompt:

tclvalue("auto_path")

This will display something like the following:

[1] "/usr/local/lib/tcl8.4 /usr/local/lib ./lib /usr/local/share /usr/local/lib/tk8.4/Library/Frameworks/R.framework/Versions/2.3/Resources/library/tcltk/exec"

LimmaGUI adds extra locations to this search path after it has been loaded and run. Some of these locations may or may not exist on your Mac. The auto_path tcl variable will now look something like:

"/usr/local/lib/tcl8.4 /usr/local/lib ./lib /usr/local/share /usr/local/lib/tk8.4 /Library/Frameworks/R.framework/Versions/2.3/Resources/library/tcltk/exec /sw/lib/tcl8.4 /sw/lib /sw/lib/tk8.4 /Users/keith/TkExtensions /usr/local/Tcl/lib /usr/local/lib/Tcl /usr/lib /usr/lib/Tcl /usr/local/ActiveTcl/lib"

You should copy the BWidgets folder and the Tktable folder into one of these locations.

5. R Packages Required

The limma package is available from Bioconductor or from the WEHI Bioinformatics site (in Melbourne, Australia).

The sma, tkrplot, R2HTML, statmod and xtable packages are available from CRAN, the Comprehensive R Archive Network. If you are using Windows, you can install packages from CRAN using the Packages pull-down menu. Alternatively, go to http://www.r-project.org, click on CRAN, choose a local mirror, choose your operating system, e.g. Windows, and click on "contrib" for contributed packages.

Comments/Questions? Contact keith@wehi.edu.au.

Last modified: 06-06-2006