Taxonomy 3 - A multivariate genetic analysis
get newsletter
email us
Introduction Methods Examples Publications
Software
Authors
Tax3 Java Application Sas Macros On-line Analysis
Cross-platform 'Tax3 Java application' for Unix, Mac and Windows

This application is freely available. It is developped for research purpose only. Several thousands of variables can be analyzed in a few minutes on a typical off-the-shelf computer. We would be happy to provide you with personalized advice on how to use this method and interpret the results.

Download version 1.2 (Mars 2008)    on-line help    

What's new in this version ?    - If upgrading, please uninstall previous version first -

Download version for latest JRE (v1.6)
    Windows self extracting installer
  Unix JAR installer
  Mac OS X Leopard JAR installer 
 
Download version for older JRE (v1.5)
    Windows self extracting installer
  Unix JAR installer
  Mac OS X Tiger and Leopard JAR installer

Windows & Unix: This application can be installed and run on any machine as long as the latest Java Runtime Environment (JRE) v1.6 is installed (it is installed by default on most Windows machines). If needed, you can download and update your JRE here : http://www.java.com/ .

Mac OS X: Tiger (X.4) and Leopard (X.5) come with the Java Runtime Environment (JRE) v1.5 pre-installed. It seems that only Leopard advanced users can upgrade to JRE v1.6 (see http://developer.apple.com/java/ ). Please download the version compatible with your system. Some minor functionalities are unavailble with the version compatible with JRE v1.5.

Installation: Download and execute the installer: it is either a Windows executable or an executable Java Archive (JAR). The installer will ask you where to install the program, examples and source code. Shortcuts will be created in Windows start menu. An un-installer is provided.

 

Capabilities       see screenshots below    top of page

The 'Tax3 Java Application' performs the taxonomy 3 multivariate analysis of relatively large case/control datasets including several variable types covering all genetic variables, genomic variables, and most sub-phenotypes:

 Variable type  Example
   Genetic, genomics variables          SNPs, HLA, Alu, repeats, expression levels, etc ...
   Binary, Categorical    gender, ethnicity, disease subtype, etc, ...
   Normally distributed    many biomarkers: BMI, most blood tests, disease severity, age, etc...
   Exponentially distributed     time to event: time to exacerbation, etc....
   Poisson distribution    number of events: number of relapses, etc...

A graphic user interface allows to easily upload, edit, analyse your data and save the results as text files and plots. The main data import format is a classical (ASCII) text file matrix with values for each observation and variable.

In addition, this application provides:

  • Investigation of the magnitude of signal providing easy-to-interpret 'p-values' using case/control permutations.
  • Multivariate inference: true positive and false positive rates (ROC curves), and individual inference of case/control status
  • Several algorithm & performance options for the principal component analysis: classical PCA & kernel PCA.
  • Many analysis and graph options

Example are included.

Screenshots    top of page

 

   top of page
Newsletter

    -> We plan to send (infrequent) emails regarding publications, talks, software updates, etc...

To subscribe, or manage your subscription, just enter your email address below:

email:

 

Send us your comments
Name   (optional)
Subject   (optional)
Email  
Comment  
    

You can also email us directly at:   taxonomy@delrieu.org