GenePattern News, July 2005, Number 2

In This Issue

  1. Announcements
  2. Featured Research Using GenePattern
  3. New Modules
  4. Updated Modules
  5. Training and Events
  6. Talk to Us

1. Announcements

GenePattern wins BioIT World Best Practices Award

The GenePattern team is proud to announce that GenePattern has won the Editor's Choice Award in the Bio-IT World 2005 Best Practices Contest. The Best Practices awards were judged by a peer review panel of experts, who reviewed a total of 33 submissions from organizations ranging from large pharmaceutical and biotechnology companies to academic institutions and agencies, detailing best practices in one of six categories.

The winning organizations were recognized for their noteworthy innovations and results in the research, development and manufacture of bio-IT-enabled therapeutics and diagnostics, as well as the underlying business strategies that support them. Read the Broad Institute and BioIT World press releases.

GenePattern 1.4.1 Released

In June the latest version of GenePattern, 1.4.1, was released. This version improves data handling and navigation, provides more information about the history of user analyses, and makes it easier to modify the server settings to your local requirements. See the release notes for the 1.4 and incremental 1.4.1 releases for more information.

2. Featured Research Using GenePattern

Lu, Getz, Miska, et al.

The in silico research used by Lu, Getz, Miska et al., in their recent paper, MicroRNA expression profiles classify human cancers. Nature. 2005 Jun 9;435(7043):834-8., is available as seven GenePattern pipelines. To reproduce their results in classifying cancer types using miRNAs, you can run these pipelines on your GenePattern server. Most pipelines currently run on Windows only. Other platforms are in development.

3. New Modules

Several new GenePattern modules are available on the GenePattern module repository. You can view comprehensive documentation on our module page and can install these modules by clicking on Install/Update in the Admin frame of your local GenePattern server.


Computes significance values for features using several metrics, including FDR(BH), Q Value, maxT P-Value, FWER, Feature-Specific P-Value, and Bonferroni. This module significantly expands and improves the functionality in ClassNeighbors.


Imputes missing values in a dataset by computing the average of the k nearest neighbors of each missing data point.

Lu, Getz, Miska et al. pipeline

These pipelines implement the analysis done in Lu, Getz, Miska et al., Nature 435, 834-838 (9 June 2005). Most run on Windows only. See the above note for more details.


Probabilistic Neural Network (PNN) calculates the probability that a new sample belongs to a given set of classes. This algorithm behaves like a Gaussian-weighted kNN classifier - closer training samples have a greater influence on the predicted class.

4. Updated Modules

The following modules have been released with significantly updated functionality:

Prediction modules save and load models

Prediction modules now provide the option of saving and re-loading prediction models. This functionality is available for KNN, SVM, and Weighted Voting, and will soon be available for CART.

5. Training and Events

GenePattern Training Workshops

The GenePattern workshops have been extremely popular. We will be scheduling the next round of workshops in the fall. Workshops are held at the Broad Institute at 320 Charles St. in Cambridge, MA, from 9:00 AM until 3:30 PM. Registration is free for attendees from academic or other nonprofit organizations, and $600 for attendees from for profit organizations. Space is limited.
These workshops teach participants the features of GenePattern, including:
  • an intuitive graphical user interface for users at all levels of computational sophistication
  • comprehensive repository of clustering, prediction, preprocessing, and visualization modules for analysis of microarray data
  • a pipeline environment that allows users to chain tasks together to create and share methodologies
  • a task integration environment that allows rapid, code-free integration of new tools
  • a programming environment that allows users to access GenePattern modules from the Java, MATLAB, and R programming languages
Watch our Web site for updates, or sign up to be notified immediately.

6. Talk to Us

Please let us know how you're using GenePattern!

User Survey

If you use GenePattern, we would like to know how your experience has been. Our user survey is a brief online form that lets you give us feedback about the software and other aspects of using GenePattern. Your responses are greatly appreciated - they will help us to understand how GenePattern is being used and how to make it as valuable a tool as possible.


If you've published a paper that makes use of GenePattern, we'd love to hear about it. Even if you are just using GenePattern in a novel way, let us know!
Email the GenePattern team

Early Adopters

If you'd like early access to new GenePattern releases to help us test new GenePattern features, please let us know. Join here

To remove yourself from this list, unsubscribe.