GeneClust Windows 2000 version

Table of Contents:

• Description
• Unpacking
• How To Run
• Analysis
• User Interface
• Input/Output

Description:

Gene Shaving is a method for clustering groups of similarly behaving genes whose changes in expression are most tightly linked to observed biological changes. The basic method is similar to observed principal components (singular value decomposition, maximum eigenvalue, etc.) with a sequential twist: a canonical "gene vector" is identified based on the eigenvectors, and the genes are ranked according to their agreement with this vector. The worst fitting are then "shaved off" and a new canonical vector is identified and fit.

The GeneClust distribution for Windows 2000 consists of a simple user interface and an S-plus back-end which calls various C functions implementing the statistical method.

Unpacking:

Create a folder named "GeneClust" on the computer's main drive. Copy GeneClust.zip and expand it in that folder using WinZip (a free evaluation version is available). The directory structure must be preserved for GeneClust to run properly. After unpacking, GeneClust.zip is not needed for execution of the application.

For convenience, a shortcut to the GeneClust folder may be placed on the desktop.

How To Run:

The GeneClust GUI can be started by double-clicking on the LaunchGC executable in the Windows Explorer or from the command line by typing the following:

C:\GeneClust> LaunchGC

It will launch SPlus2000 as well as the GeneClust application.
SPlus2000 must not be running at the time LaunchGC is executed!

The GeneClust application runs under both SPlus6 and SPlus2000. If both versions coexist on the same machine, the LaunchGC executable launches the one used last.

Analysis:

In order to run the analysis on different data sets, the data files (in tab-separated format) and their corresponding classification files (for supervised shaving) must be saved in the data subfolder. Make sure the file extension for the classification files is ".clf".

When performing supervised shaving, make sure the dimensions of the vector in the classification file and the number of samples match.

The output files are stored in the output subfolder by default. Directory browsing is not currently implemented. Advanced users may specify where output should be stored by modifying the gcoutput variable in the "GeneClust.ssc" file (set it equal to the full path of the desired folder) and saving the changes.

User Interface:

The GeneClust application can process raw data or generate it in demo mode. Data entry is not checked for validity; check input carefully. Tooltips containing the allowed ranges are provided for all numerical variables.

Input/Output:

• Files:
  • <supervision>.clf: Classification file used to perform supervised shaving.
  • <datafile>.tsv: ASCII file that contains tab-separated values.
  
  

Back to Home Page