The CID Cytoscape Plugin lets you import data from CID (CISBAN Interactomes Database) into Cytoscape. CID that contains integrated interactomes of different model organisms used in ageing research.

CID Plugin window loaded in Cytoscape

• The CID Cytoscape Plugin User's manual
  • Availability
  • Installation
    • Help! My installation wasn't successful!
  • Starting the plugin
  • Single species analysis
    • Loading a complete network
    • Retrieving the neighbourhood of a gene
    • Calculating all paths between two genes
  • Cross-species analysis
    • Reference and target species
    • Node neighbourhood and paths between nodes
    • Retrieving all interactions between a set of genes
  • Viewing the data and applying a layout
• Saving your data
• More information
• Limitations

Download

Before attempting to install the CID plugin, you should have been provided with the following file:

• cid_plugindate.zip

Where date is the date of the release, e.g. cid_plugin20110225.zip for 2 Feb 2011. Once you have this file, you are ready to proceed with the installation.

1. First of all, you need to have Cytoscape installed on your computer. The plugin is compatible with Cytoscape 2.7.0. Go to this page to download the latest version if yours needs to be updated.
2. Before you start, make sure Cytoscape is not running on your computer.
3. Extract cid_plugin20110225.zip. You should see a folder called "CID Plugin" containing two files:
   • postgresql-8.1-413.jdbc3.jar: a helper module that allows the plugin to talk to the database
   • CIDPlugin-1.0-SNAPSHOT.jar: the actual code for the plugin
4. Locate the Cytoscape installation folder on your computer. This folder is named after the version of Cytoscape. For instance, if you have version 2.6.2 then the folder will be called Cytoscape_v2.6.2. On Mac OS X, this folder is typically located under the Applications folder. On Windows, you might want to look in the Program Files folder or any other location you installed Cytoscape in. On Linux, Cytoscape tends to live in /usr/local/cytoscape.
5. Inside your Cytoscape installation folder, there is a folder called "plugins". Copy the two files (postgresql-8.1-413.jdbc3.jar and CIDPlugin-1.0-SNAPSHOT.jar) into the plugins folder of your Cytoscape installation folder. Make sure you copy both files directly and not the folder "cid_plugin20100225".
6. When you next start Cytoscape, there should now be a menu item named "CID" in the Plugins menu with two sub-menu items: 'Load CID Network' and 'Configuration' .

1. Stay calm!
2. Contact the CISBAN helpdesk for help.

Cytoscape menu bar highlighting the CID menu option

Easy! Just go to the Plugins Menu and select the CID > Load CID Network menu item. A dialog window should appear. In that window there are two tabs you can choose from. The first tab says "Single species analysis". Choose this one if you only want to retrieve data for a single organism. The second, "Cross-species analysis" can be used when you want to compare networks between different species.

There are three ways to retrieve data for a single organism:

• Load the complete integrated network for an organism
• Retrieve the neighbourhood, i.e. all the interaction partners, of a specific gene.
• Calculate all paths between two specific genes.

Before you can do any of these, you first have to choose a network from the "Choose a network" drop down list.  You'll notice  that there can be more than one network for a single organisms. When you choose a network, a brief, somewhat technical, description of the network appears in the Description box. The first line of this description tells you what type of network you're dealing with. If the line simply says "Integrated network according to the method described in:", followed by a reference then the network is created from direct experimental data on that organism. If it says "Integrated network created by combining a previously existing integrated network with inferred relations" then the network also contains relations inferred through orthology, i.e. based on data from another organism.

Depending on what you want to do, you want to prefer one type over the other. If you're not sure however, it's best to use a network that contains both types of data.

Complete D. melanogaster network loaded by CID

WARNING: Loading a complete network may require a lot of memory. It is not recommended to load a network with many interactions if you have 2GB or less of RAM. Check the table below to see how many interactions the network for each organism has. Note that these numbers may change in the future.

Currently, CID contains interactions, annotations and other data from 44 different sources. These data have been integrated and inferred to produce the following integrated networks:

Organism	Interactions from literature	Total number of interactions
Caenorhabditis elegans	24376	781494
Danio rerio	10	1666261
Dictyostelium discoideum	0	391354
Drosophila melanogaster	8624	1438520
Homo sapiens	527796	1568535
Mus musculus	46834	1685799
Paramecium tetraurelia strain d4-2	0	1499834
Podospora anserina	0	410642
Rattus norvegicus	52385	1236548
Saccharomyces cerevisiae	401563	475299
Schizosaccharomyces pombe	3027	432529

To load a complete network:

1. Select a network from the "Choose a network" drop down list.
2. Check the radio button labeled "Entire network"
3. Click the button that says "Get network".

The plugin will now retrieve the network from the database.

Loading the immediate neighbours of the S. cerevisae gene YKU70

1. Select the organism from the "Choose an organism" drop down list.
2. Check the radio button labeled "Node neighbourhood"
3. Fill in the name of the gene in the text box labeled "Node" for which you want to retrieve the neighbourhood. Make sure this is a gene from the organism you've selected in step 1.
4. Fill in the size of the neighbourhood that you want to retrieve in the field labeled "Size". A neighbourhood of size 1 means that you only want to retrieve the direct interaction partners of the specified gene, i.e. only one edge 'away'; from the central gene. Size 2 means you want to retrieve all genes up to 2 edges 'away', and so on. WARNING: Setting the neighbourhood size to a value of 3 or more may result in a very large network being retrieved which may cause your computer to run out of memory. You first might want to retrieve smaller neighbourhoods and try again later with a higher size setting.
5. Click "Get network" to retrieve the neighbourhood. If the gene you specified could not be found in the database or if the gene you specified does not have any interaction partners, you will get a notification. If not, the network will be displayed.

A

B

A  CID Dialog showing the settings to calculate the paths of 4 or less between TP53 and IGF1 in H. sapiens B  Resulting Cytoscape network from the dialog configuration in A, with a 'Spring Embedded' layout.

This option allows the calculation of all paths between two genes. This function is useful when you know that two genes are involved in the same pathway, but you do not know the exact interactions that regulate both. The plugin will calculate all possible paths up to a given length. For each path a score is calculated by averaging the scores of the individual edges in the path. The best-scoring paths are then returned and visualised in Cytoscape.

This function is an automated version of the pathway analysis performed in Joao Passos et al in Age (2009) 31:353-363.

To calculate the best scoring paths between two genes, follow these steps:

1. Select the organism from the "Choose an organism" drop down list.
2. Check the radio button labeled "Paths between nodes"
3. Fill in the start gene of the pathway in the field labeled "Start node" and the end gene in field labeled "End node"
4. Set the maximum path length in the field labeled "Maximum length". This is the maximum number of links allowed between the start gene and the end gene. Beware that calculating paths of length 6 or more may take a long time to compute, up to several hours. It is best to first calculate all paths of a smaller length and only try bigger values if no paths could be found or if the results are not satisfactory.
5. Set the number of paths to be returned after calculation in the "Number of paths" field. For instance, if you set this value to 10, only the 10 best scoring paths will be returned.
6. Click "Get network" to retrieve all paths. You will get a message if one of the specified gene names could not be found in the database, if one of the genes has no neighbours, or if no paths could be found. Otherwise, a network will be displayed upon completion.

After you have retrieved the data, you will see that all the nodes of your network are arranged in a grid. Typically you will want to change the layout.

• For entire networks and large node neighbourhoods, the best options are:
  • Layout → Cytoscape layouts → Edge-weighted Spring Embedded → weight
  • Layout → Cytoscape layouts → Force-Directed Layout → weight
• For networks with information from multiple species, it's best to use :
• Layout → Cytoscape layouts → Group Attributes Layout → organism

• For small node neighbourhoods, you also might want to use:
  • Layout → yFiles → Orthogonal

• For paths between two genes, you can get nice results by using:
  • Layout → Cytoscape layouts → Hierarchical Layout

The CID plugin sets a number of attributes that may prove useful when using Cytoscape's vizMapper or for understanding the data.

• Nodes
• DB_ids: used by the CID plugin to keep track of alternative splice variants
• canonicalName: the shortest alias for the gene with the most letters
• partOfInputSet: if this node was one of the ones that was originally queried for. The canonicalName of this node may be different than the one you queried for.
• organism: the organism from which this gene originates
• Edges
• interaction: 'pp' stands for protein-protein interaction; 'ortholog' indicates an orthology relationship
• isDirectional: whether the interaction is directional, e.g. a transcription factor and its binding site
• isEstablished: whether there is literature or public databases that support the existence of this interaction
• isInferred: whether CID predicted that this interaction exists
• pathIDs: when finding a path between two genes, this list indicates which pathways a gene is part of.
• pathWeight:the highest path weight of the paths this gene is part of. For example, gene A is involved in paths 1 and 2, with path 1 weighted at 1.5 and path 2 weighted at 2. The pathWeight attribute for gene A would be 2.
• pubmed_id: the publications that support the existence of this interaction, or the publications that support the existence of an orthologous interaction in other organisms.
• weight: the confidence in the existence of this interaction

You can save the network data you have retrieved and any layout and visual styles you may have applied using File → Save.

See the Cytoscape manual for more information on how to further analyse your network. The tutorial also includes a section on how to overlay network data with expression data and on how to create visual styles.

1. The software is still in alpha version. That's alpha as in before beta-version. Not as in alpha-male :-). This means that there still might be some bugs in it and not everything is quite finished. This also means that if you feel some features are missing, please email the CISBAN helpdesk.
2. At the moment it is not possible to view the gene annotation or interaction type.

Return to main CID page