geWorkbench 2.2.1 (2011-07-29) ============================== --- New features and changes in release 2.2.1--- ANOVA - #2618 - Control buttons repositioned for easier use. - #2694 - The simple range check as a test of whether data was log-normalized was removed. ARACNe - #2494 - Bonferroni correction option added. - The "Choose edges with hightest MI" option was renamed to "Merge multiple probesets". - #2648 - The adjacency matrix is no longer made symmetric. Each line now starts with a hub gene. - Similarily, the adjacency matrix, when saved to file, is not made symmetric. - #2679 - The "merge multiple probesets" option was not recorded in the dataset history. - Merging of multiple probesets is now just within a hub gene, not global. - #2684 - When a subset of markers was activated, the grid service version used an incorrect offset into the marker list. BLAST - #2665 - Previous search results were not cleared if a following search returned no hits. Color Mosaic - #2401 - Export button and right-click menu - both now support export of significance results (p-values) from both t-test and ANOVA. CNKB - Export of network adjacency matrix from Project Folder omitted gene/probeset names. Cytoscape - t-test overlay on network did not work in release 2.2.0 due to a late change. Fixed. - #2650 - Display all networks at the gene level. When multiple probesets represent a gene in the network, the network will be summarized at the gene level. As a result, the count of the number of nodes and edges displayed in Cytoscape may differ from the counts present in, e.g. a complete ARACNe adjacency matrix (ARACNe probeset merging option not used). - #2589, #2638 - If a network might be too large to view in Cytoscape, the user is offered an alternate visualization (text). The threshold for warning can be changed by the user. This replaces warnings that were previously built in to several components e.g. ARACNe. Now, an adjacency matrix is always created, regardless of size. - #2597, #2598, #2599 - Inconsistencies in how selected nodes in Cytoscape were reflected as markers in the Markers component were found and fixed. File Open Dialog - #2698 - The "merge" checkbox is no longer disabled when a file type is selected which cannot be merged. There were synchronization problems. Fold-Change Analysis - New component accidentally omitted from release 2.2.0. Network Import - Networks can now be imported either from Adjacency matrix files (#2432) or from SIF-format (#2434) files. The networks can be represented by gene symbols, Entrez IDs, or probeset names, or by some other identifier (#2432). Networks are represented in memory with the same identifiers that were imported. That is, if a network is read in using gene symbols, then it is represented at this level within geWorkbench. Probesets for each gene can be obtained as needed for particular analyses. - #2608, #2645 - On import, .SIF and .ADJ file extensions are automatically recognized. - On export, .ADJ is automatically added to file. Gene Ontology Analysis - #2512 - The analysis now supports using an alternate annotation file rather than an Affymetrix annotation file. Organism annotation files from the geneontology.org website can now be used. GenSpace - The J2EE framework used in geWorkbench 2.2.0 was replaced with one based on SOAP. - #2666 - The "Remove Friend" feature did not properly update the GUI. Marker Annotations - #2658 - a missing CGAP URL was added to the export file. MarkUs - #2622 - Added support for job description and email notification fields. Master Regulator Analysis - #2610 - MRA can now be run from gene-level adjacency matrices created by CNKB. - #2301 - When saving results (target lists for MRs), a ".csv" file chooser is used and will append this extension automatically. MINDy - #2659 - If MINDy was run remotely on the grid service, sorting on the modulators column of the results table did not work. - #2657 - The "All Markers" checkbox was removed, as its function has been taken over by a separate filter control. - #2314 - An inconsistency in how modulators and targets were selected/deselected with respect to the "Displayed Targets Filter" was fixed. Pattern Discovery - #2682 - Some parameters were not properly captured in the dataset history. - #2672 - Pattern nodes were not always properly restored when the workspace was saved and restored. - #2664 - In Exhaustive discovery mode, the "min support" parameter is now input only as an integer representing the floor for support. The option to enter the minimum exhaustive support as a percent of initial support was removed. - #2663 - If discovered patterns were saved to file and reloaded, they were not always displayed. - #2593 - Problems that occurred when a run was canceled or the server was unreachable were fixed. - The Pattern Discovery (Splash) server URL was changed to splash.c2b2.columbia.edu. PCA - #2670 - PCA did not run if geWorkbench was installed to a path containing a space. Project Folders - #2626 - Mousing-over a network node will display the number of nodes and edges in hover text. - #2646 - When saving a network, the ".adj" file filter will append ".adj" to the file created. Sequence Retriever (#2630, #2631, #2632) - For DNA, now queries the "refGene" table rather than UCSC "known genes", so can now retrieve genomic sequence for all UCSC supported species. - The transcription start point was off by one in query and display. - Identifiers for retrieved sequences were being truncated if long. - The "local" option for sequence retrieval was removed. - The sequence location "spinner" controls can now be set to zero, and step in increments of 100. geWorkbench Installer - When run in Windows, the installer version of geWorkbench now displays its correct icon in the Windows task bar. ---Online Help updates--- Help files were updated for the following components: - ANOVA - ARACNe - BLAST - CNKB - Cytoscape - Filtering - JMOL - MarkUs - Master Regulator Analysis - MINDy - Normalization - Pattern Discovery - Promoter Analysis - Pudge - t-test ---Versions of external files/components included in release 2.2.1--- - Cytoscape 2.7.0 (no change). - gene_ontology.1_2.obo downloaded 2011-06-29 from geneontology.org. - Ontologizer.jar version 2.0, file released 2010-03-10 (no change). - Jaspar_CORE (http://jaspar.genereg.net/) SQL files last updated on server 2009-10. (no change). - JMOL 12.0.45 (updated).