SpRay


SpRay – Visual Analytics Tool

SpRay supports the visualization of the high dimensional data; such as microarray data using parallel coordinates and information visualization methods.

Trends and clusters can be explored through the application of specific transparency modulations and colormaps. However, often the raw data does not provide enough structure to allow a comprehensive analysis. Therefore, in SpRay visual exploration with statistical analysis methods are combined. This combination allows to discover relations that were difficult to reveal with visual methods alone, since it allows the identification of irrelevant data, which can henceforth be removed from visual representation.

Another valuable advantage of this combination is the possibility of visualizing the effect of the various analysis methods, it has been shown with the half-marathon dataset. Reliability or instability of the individual methods can be examined and considered for a specific application and allows this way a better understanding of them.


How-To

Coloring of BarPlots created under ICA and PCA Methods

When you run SpRay and start working with the Dimension Reductions methods such as ICA (Independent Component Analysis) or PCA (Principal Component Analysis) under the Feature Extractor Module, you will see various different features which will help you moving on with your work.

When you can select and run the Covariance or Correlation PCA methods on your dataset than you can also create BarPlots using the R integration in SpRay.

All you need to do is to select the Reduction Method (and/or the Reduction Rule) afterwards you should select an output and the ‘Run R’ button will become available for execution.

When you click ‘Run R’ button, in the release folder of SpRay, you will see automatically generated pdf files, titled ‘PCA overview’ and ‘PCA overview with clusters’.

What you need to know is the coloring of these BarPlots are fixed and could only be changed by editing the script files located under the ‘contrib/colors’ folder. You will find this folder also in the ‘release’ folder.

Changing the defined colors and groups in ‘colors_ICA.R’ or ‘colors_PCA.R’ files is a easy and straight forward process. Edit the scripts according to your preferences and number of groups you prefer to see in the generated BarPlots.

If you wish to edit the Legend in the ‘PCA Overview with Clusters’ file than you need to edit the ‘legend_PCA.R’ file and make changes according to your needs.

Generating the Neighborhood Evolution Matrix as ColorDotPlots using Locally Linear Embedding feature

When you use non linear feature extraction methods of SpRay, you may also have the opportunity to apply LLE (Locally Linear Embedding) on the data.

After selecting the LLE options like the Metric and Neighborhood, SpRay makes the ‘Run R’ button available for the LLE. Next step is to click the button and the ‘lNEM-xN.pdf’ containing the l-Neighborhood Evolution Matrix will be generated under the release directory.

LLE applied on the columns is shown below, using the Scatterplot Matrix visualizer of SpRay.

If you may have any question regarding this procedure you can contact polatkan[@]informatik[.]uni-tuebingen[.]de for further assistance.


Download

Before starting SpRay, make sure you’ve downloaded and installed the preliminary libraries.


Linux

In order to run the SpRay releases in order to use the full functionality of the SpRay.

  • Install R on Linux
  • Download Rcpp and RInside, then run the command:
    $ sudo R CMD INSTALL filename.tar
  • Using the Synaptic Package Manager or the command
    $ sudo apt-get install librarynamehere

    install the Glew, QWT and GSL libraries.

  • Download RColorBrewer, then run the command:
    $ sudo R CMD INSTALL filename.tar

Mac

In order to run the SpRay releases with full functionality, you need to install the required packages and dependencies.

SpRay release for Mac depends on

1) A x86_64 bit MacOS kernel & libraries
2) Glew, QWT, GSL libraries
3) R 64bit
4) Installed QT Framework (4.7)

We would like to provide you a walkthrough of steps you have to follow, below:

  • Install Xcode
  • Install R on mac
  • Download Rcpp and RInside, then run the command:
    $ sudo R CMD INSTALL filename.tar
  • Install Macports
  • Install qmake via Macports by running the command:
    $ sudo port install qt4-mac

    (remember it may take about a hour long and 6 GB of space)

  • Install Glew via Macports by running the command:
    $ sudo port install glew +universal
  • Install QWT via Macports by running the command:
    $ sudo port install qwt-60
  • Install GSL via Macports by running the command:
    $ sudo port install gsl
  • Install Doxygen
    $ sudo port install doxygen
  • Download RColorBrewer, then run the command:
    $ sudo R CMD INSTALL filename.tgz

Download Releases

Releases for version 1.1 (stable):

 

Current and previous versions:

Releases for version 1.1  – available on February 1st, 2012 (current)