GSEA v3.0 Release Notes

GSEA Home | Downloads | Molecular Signatures Database | Documentation | Contact

GSEA Desktop v3.0 Beta 2 Release Notes

Released October 14, 2016

GSEA Desktop v3.0 Beta 2 requires Java 7 or Java 8.  Java 8 is highly recommended.

Our tests show this Beta version produces equivalent results, but use the Production version if you have concerns.  At a minimum, verification with the Production version before publication is strongly recommended.

New Features

• There is a new option for saving the datasets backing the report heatmaps for downstream use & visualization. Use the ‘Create GCT Files’ parameter in the Advanced Fields. These files can be used with e.g. R, GenePattern, Morpheus, or GENE-E among other options.

Bug Fixes and Other Improvements

• Selection of multiple CHIP files for analysis is no longer allowed. This is a simplification that will permit further back-end improvements in the future.

• Collapse Dataset is no longer allowed with a GSEA Preranked analysis. This was a rarely-used feature that caused a lot of confusion.

• Fixed an HTML report launcher bug for users with non-English locale settings.

• Fixed an issue with running the Enrichment Map integration on Mac.

---

GSEA Desktop v3.0 Beta 1 Release Notes

Released August 15, 2016

GSEA Desktop v3.0 Beta 1 requires Java 7 or Java 8.  Java 8 is highly recommended.

Our tests show this Beta version produces equivalent results, but use the Production version if you have concerns.  At a minimum, verification with the Production version before publication is strongly recommended.

New Features

• In an effort to streamline the GSEA desktop application, the MSigDB Browser feature has been split out into a separate application. This will see its own Beta release at some point in the future - users requiring the feature can continue to use the production GSEA Desktop v2.2.x in the interim. Another option is the online MSigDB browser www.msigdb.org.

• Plots and other results can now be saved in the SVG format, which provides improved resolution suitable for publication, better post-analysis editing options, etc. Use the ‘Create SVG plot images’ parameter in the Advanced Fields.

• The Enrichment Map integration has been updated to work with current versions of Cytoscape (v3.3.0 and newer).

• New improvements for long-running file transfer operations include enhanced performance through I/O buffering and compression during transfer, and the addition of a progress bar.

• Analyses run with a ‘timestamp’ random seed, now record the timestamp value in the HTML Report ‘Other’ section. The same seed can be reused at any time to reproduce the analysis.

Bug Fixes and Other Improvements

• Fixed an issue where ‘NaN’ (division by zero) results were given a value of zero in the reports.

• Fixed issues handling C3 MIR Gene Sets and other gene sets with names containing comma characters.  Set the ‘Alternate Delimiter’ parameter and use a semicolon to separate gene set names.

• Improved control layout on the lower Tool Command panel.

• Improved layout of the result chooser panels on the Leading Edge Analysis and Enrichment Map Visualization screens.

• Updated the Browser Launcher component to fix issues with launching the web browser.

• Improved the error reporting when there are too few samples in the dataset.

• Fixed some basic aspects of integration with the Mac OS X menu bar.

• Removed Thread Controls from the UI as these rarely-used functions led to instability in the application.

• Improved memory resource use and clean-up during GSEA report generation.

• Fixed a failure in the Chip Chooser component when not connected to a network.

• Cleaned up clutter in the application menus and preferences.

• Added an override so the application can use a local GENE_SYMBOL.chip rather than FTP transfer.

• In preparation for a future open-source release of the GSEA application under an OSI-approved license, we made a number of changes toward cleaning up the code base, removing / replacing a number of third-party libraries, and removing unused resources from the GSEA jar file.