Differences between revisions 20 and 21
Revision 20 as of 2006-01-13 23:57:10
Size: 3035
Editor: GaryBader
Comment:
Revision 21 as of 2006-01-14 00:41:37
Size: 4552
Editor: GaryBader
Comment:
Deletions are marked like this. Additions are marked like this.
Line 28: Line 28:
  * [7. ID Mapping Services, Phase I] [Optional?] (Gary)   * 7. ID Mapping Services, Phase I (Gary)
   * To integrate e.g. gene/protein networks, we need to create a Cytoscape API to handle database identifier (ID) mapping. For example, if a gene network is defined by Affymetrix identifiers and a loaded PSI-MI file is defined by Entrez Gene identifiers, Cytoscape must be able to map Affymetrix IDs to Entrez Gene IDs. Note: cPath already maintains identifier mapping information and sends this information to Cytoscape in BioPAX or PSI-MI, however Cytoscape does not have an ability to automatically link the identifiers and currently represents the differently identified genes as two different nodes. A simple Cytoscape identifier mapping API that could handle linking these genes would solve this problem. This work will not create an extensive collection of identifier mappings because this task is difficult, outside of project scope and is being worked on by multiple different bioinformatics groups. Instead, this API will simply resolve identical biological objects using whatever existing mapping information is given to it. The API will support four use cases: 1. unification of network nodes, such as proteins, during dataset merging, 2. link out to related information, such as protein function annotation, 3. identifier translation from one type of identifier to another and 4. searching a network using a preferred gene name or alias.
   * ''TIMELINE'' Currently this task is optional. The MSKCC group has a design and can create an RFC, but are not yet sure if they can prioritize implementation by 2.3 release.

Use cases addressed:

  • First steps towards drawing pathways / metabolic networks.
  • Support for large networks
  • Key Usability Issues
    • Save State, Undo
    • Viz Mapper / Filters Specific Issues

New Features:

  • New rendering engine; lots of plugin dependencies
    • (Nerius, Allan as consultant) Also includes Layout algorithms for large networks
  • Meta-Nodes; Phase I
    • (Iliana, Allan, Melissa, Aditya, Alex Pico, Scooter)
  • Saving Project State (["Save current session as XML file"]) (Kei, Iliana); Depends on Meta-Nodes
    • TIMELINE Basic session reader and writer are done. Need to change them to accept Metanodes.

  • Undo Manager (Allan w/ help from friends)
  • Attribute browser
    • link attribute browser to network view (Kei)
    • Meta-node compatible browser window (Show/hide member nodes) (Kei)
  • New, improved Cytoscape Tutorials (Melissa)
    • There will be eight new tutorials: 1. Getting started, 2. Filters & editor, 3. Fetching external data, 4. Expression analysis, 5. Literature searching, 6. GO analysis, 7. Modules and complexes, 8. Domain networks and other topology topics. Other topics may be added if the demand is sufficient. Estimated timeline: 3-4 weeks.

  • Graph Merge, moved to core (already a plugin) (Ryan, UCSD)
  • [6. Clean-up Viz Mapper and Filter UIs (["VizMapUI"]); make friendlier; move to cytopanels] [Mike S, Allan, (possibly Alex?)]
    • TIMELINE

  • 7. ID Mapping Services, Phase I (Gary)
    • To integrate e.g. gene/protein networks, we need to create a Cytoscape API to handle database identifier (ID) mapping. For example, if a gene network is defined by Affymetrix identifiers and a loaded PSI-MI file is defined by Entrez Gene identifiers, Cytoscape must be able to map Affymetrix IDs to Entrez Gene IDs. Note: cPath already maintains identifier mapping information and sends this information to Cytoscape in BioPAX or PSI-MI, however Cytoscape does not have an ability to automatically link the identifiers and currently represents the differently identified genes as two different nodes. A simple Cytoscape identifier mapping API that could handle linking these genes would solve this problem. This work will not create an extensive collection of identifier mappings because this task is difficult, outside of project scope and is being worked on by multiple different bioinformatics groups. Instead, this API will simply resolve identical biological objects using whatever existing mapping information is given to it. The API will support four use cases: 1. unification of network nodes, such as proteins, during dataset merging, 2. link out to related information, such as protein function annotation, 3. identifier translation from one type of identifier to another and 4. searching a network using a preferred gene name or alias.
    • TIMELINE Currently this task is optional. The MSKCC group has a design and can create an RFC, but are not yet sure if they can prioritize implementation by 2.3 release.

Refactoring / Design Issues:

  • (["Node id issue"]) – creation of a default “label” CyAttribute (Ben)

  • New documentation options; docbook, wiki, etc. (Gary, Mike)
  • Take care of delete/hide/remove situation (Allan, in conjunction with Cytoscape Editor)
  • Flagging/selection of nodes/edges (Iliana)

Misc. Small Items

  • Ability to rename networks (Brad K at UCSD)
    • Brad is adding code to allow networks to be renamed. They are currently named based on the input file. This change will change that name throughout the system.
    • TIMELINE Nearly complete. All of the underlying code is in place, we just need to figure out how refresh the name in the network panel.

  • Script to update source license
    • Some of the Cytoscape source is missing copyright and license information. This needs to be fixed.
    • TIMELINE We have a script that needs to be updated with the proper information and tweaked so that it doesn't stomp on existing copyrights.

  • Generic Linkout Pop-up Menu (Doron B, MSKCC)
    • This is a plugin that will read URL patterns from a user-defined configuration file to populate the links in the context menu activated by right clicking on a node or edge. Configuration files for popular sites will be provided e.g. a yeast URL config file.
    • TIMELINE Estimated completion May 2006 (last estimated Jan.13.2006)

  • Re-organize save/load/import/export menu items (["New File Menu"]) (Kei)

Cytoscape_2.3 (last edited 2009-02-12 01:03:34 by localhost)

Funding for Cytoscape is provided by a federal grant from the U.S. National Institute of General Medical Sciences (NIGMS) of the Na tional Institutes of Health (NIH) under award number GM070743-01. Corporate funding is provided through a contract from Unilever PLC.

MoinMoin Appliance - Powered by TurnKey Linux