← Revision 3 as of 2007-09-27 18:27:30
Size: 2022
Comment:
|
← Revision 4 as of 2007-09-27 18:28:05 →
Size: 2023
Comment:
|
Deletions are marked like this. | Additions are marked like this. |
Line 3: | Line 3: |
= FAQ: Frequently Asked Questions =; | = FAQ: Frequently Asked Questions =; |
= FAQ: Frequently Asked Questions =;
The plugins directory has all the relevant jar files and the shell from which I invoked Cytoscape indicates that all plugins have been successfully loaded. But I am unable to see loaded plugins from the Plugin drop-down menu. Am I missing anything?
The message �all plugins have been successfully loaded� means all of the plugins found in the plugins dir are actually loaded. The "plugins" menu is somewhat misleading as there is no requirement that a plugin actually present itself there. For example, the biopax plugin is accessed simply by importing a biopax formatted file. Other plugins are found throughout the application, like the cPath plugin that is accessed through the "File -> New -> Network -> Construct network using cPath" menu.
I'm using Cytoscape to visualize a large network (around 300K edges and 7000 nodes), the layout algorithms don't seem to operate correctly. Is the network too large?
By default, Cytoscape allocates 512MB for Java, which is definitely not enough memory for very large network. You can do a couple of things to increase the default heap size, depending on how you are running Cytoscape. (1) If you are executing it from the shell you want to edit cytoscape.sh and change '-Xmx512m' to something more reasonable for your machine, say '-Xmx1536m'. (2) If you are executing it from the finder (e.g.double-clicking the Cytoscape icon), you need to open a shell and go into the Cytoscape.app directory, then cd to Contents/Resources and edit the file i4jlauncher.config. Near the bottom of that file is a line that says 123=-Xmx512M -- change that one to 123=-Xmx1536M.
Now that that is done, you should have a little more memory to work with. The next question is which layout algorithms are you interested in using? Some algorithms are more sensitive to the number of edges than others. With that many edges, I suspect you will wind up with a pretty dense hairball by the time you visualize it.