BioGRID Plugin 2.0
BiogridPlugin2 is a plugin for the Cytoscape network visualization tool (http://www.cytoscape.org), designed to allow easy import of interactions from BioGRID data files. It is an update of the BiogridPlugin 0.2 release to allow import of data from our new tab2 file format. Additional features of BiogridPlugin2 include filtering of data during import based on gene lists and various interaction attributes, expanding of nodes and saving of networks in tab2 format. The plugin can be downloaded at http://thebiogrid.org/download.php.
BioGRID interaction data can be downloaded in several file formats (PSI XML 1, PSI XML 2.5, MITAB, and BioGRID specific formats tab and tab2). All data curated by BioGRID is present in the tab2 files (see http://wiki.thebiogrid.org/doku.php/downloads) . Please note that importing data from larger files will be slower, and it is recommended to use the BioGRID organism specific files, if possible. The plugin must be set to retrieve data from a single BioGRID tab or tab2 file by choosing Plugins>BioGRID Operations >View/Change current file. The currently set file will also be automatically set when a file is loaded using Plugins>BioGRID Operations>Load network. If the Cytoscape session is saved (File>Save), the currently set BioGRID file will be reinstated when the session is reloaded.
Filters to include or exclude interaction data during importcan be created by choosing Plugins>BioGRID Operations>Filters. Any number of named filters can be created, and the currently set filter will be applied to data being imported from tab and tab2 files. Interactions can be filtered according to several attributes include interactor names and IDs, interaction evidence codes, pubmed ids and paper throughput. In the case of importing data by expanding a node in the current network (see ‘Importing Data’ below), the ‘Filter by gene’ section of the filter will be ignored. Any filters saved will be reloaded when a saved session is reopened.
Data can be imported by using Plugins>BioGRID Operations>Create network, which imports data from the current BioGRID tab or tab2 file according to the currently set filter. If a filter-limited network has been loaded, further interactors can be pulled in by right-clicking on a node in a network and choosing Biogrid Operations>Expand or Expand All Selected, which imports all interactions for those nodes from the current BioGRID file , according to the attributes of the currently set filter, disregarding filtering by gene. New nodes are laid out circularly around the selected node avoiding overlap with other nodes; in the case of a new node which connects with more than one expanded node, the new node will be placed around one of the expanded nodes. Newly added nodes from expansion will be the only selected nodes after expansion (highlighted yellow) so that they can be conveniently repositioned together and/or have a layout applied. To immediately undo an expansion, simply choose Edit>Delete Selected Nodes and Edges while the nodes are still selected.
Biogrid tab and tab2 files contain a single row for each unique combination of interactors A and B (A → B and B → A are also on separate rows), experimental system (and Modification for tab2 files) and publication. When data is imported using BiogridPlugin2, a single Cytoscape interaction is created for each unique combination of interactors, regardless of any other attributes. During import Cytoscape attributes are created with names beginning ‘biogrid.’ , for data in corresponding columns in the BioGRID file, e.g. ‘biogrid.EXPERIMENTAL_SYSTEM’, making a unique list of values where multiple rows with the same unique interactor pair have different values for the column. Other ‘metadata’ attributes are generated during import (see table). When data is imported from a tab2 file, each column also has an EXPANDED equivalent, which has a mapping between the BIOGRID_INTERACTION_ID and the value for that interaction (e.g. 249420=physical, 250620=genetic, 264501=physical). This system of loading attributes preserves the data structure contained in the tab2 file.
|biogrid.INTERACTION_DIRECTION||The A-B directionality of the interaction in the BioGRID tab file compared to the directionality of the interaction in Cytoscape (i.e. source node – target node). ‘forward’ (same directionality), ‘reverse’ (opposite directionalities) or ‘bidirectional’ (data loaded from file contains the interaction in both directions|
|biogrid.INTERACTION_DIRECTION||As biogrid.INTERACTION_DIRECTION, but there is a list of values, one for each row imported from file, with the format biogrid_interaction_id=value.|
|biogrid.INTERACTION_TYPE||This column is present in tab2 files but absent in tab files. When data is imported from tab files, this attribute is generated by mapping the EVIDENCE_CODE to ‘physical’ or ‘genetic’. For unique interactions having >1 imported row and both physical and genetic values, this attribute value is set to ‘physicalANDgenetic|
|biogrid.INTERACTION_TYPE_EXPANDED||As biogrid.INTERACTION_TYPE, but there is a list of values, one for each row imported from file, with the format biogrid_interaction_id=value.|
|Biogrid.NUMBER_OF_EXPERIMENTAL_SYSTEMS||Number of values in biogrid.EXPERIMENTAL_SYSTEMS|
|Biogrid.NUMBER_OF_PMIDS||Number of values in biogrid.PUBMED_IDS|
|Biogrid.TOTAL_RAW||Total number of rows imported from file with the unique pair of interactors.|
On loading data from a BioGRID file, a new visual style (BiogridViz) is loaded into Cytoscape and applied to the network. Nodes have biogrid.OFFICIAL_SYMBOL as labels. Edge colors are mapped to biogrid.INTERACTION_TYPE; green for ‘genetic’, orange for ‘physical’ and blue for ‘physicalANDgenetic’. Edge width is mapped to biogrid.NUMBER_OF_EXPERIMENTAL_SYSTEMS , with the default width of 1 for edges with 1 experimental system, and 3 for edges with greater than one experimental system.
A BioGRID tab2 file can be created from a network loaded from a BioGRID tab2 file. This is particularly useful for creating custom-filtered interaction datasets. Choose Plugins>BioGRID Operations>Save network to file.
The Cytoscape visual mapper applies a visual property to a node based on the mapped attribute. If that attribute is a list, only the first value in the list is used to decide the visual property of the node. The BiogridPlugin allows the current visual style’s node color mapping to be applied to nodes, using all colors mapping to the node's list values. Choose Plugins>BioGRID Operations>Pie Colors>Add/Refresh colors to add coloring (or refresh after changing the Visual Style mapping or node attribute changes), and Plugins>BioGRID Operations>Pie Colors>Remove to remove pie coloring.
All options previously in the ‘Plugins>BioGRID Operations’ menu have been moved to a tab with title ‘BioGRID’ in the left hand control panel.
The ‘Create network’ menu option is now a ‘Import into new network’ button and the new ‘Add to network’ button will load additional interaction data into the currently-selected network from the current BioGRID data source, according to the currently-selected filter .
Data can now be imported from the new BioGRID REST service, which is set as the current data source by default, but can be changed under ‘Select Data Source’.
*Due to an upgrade of our main REST service the URL for use use in the Cytoscape plugin must be changed to: http://sce-bio-c02577.bio.ed.ac.uk:8080/BiogridRestService/resources/interactions*
The BioGRID data fields Systematic Name and Official Symbol can now be mapped to Cytoscape node IDs during import, as well as the Entrez Gene ID field. Please note, the latter is the only ID guaranteed to be a unique gene identifier in BioGRID. This setting can be changed under ‘Select Data Source’.
When Cytoscape is closed, current settings, including all filters are saved and will be imported on start up. This differs from the previous version (2.0) where settings were saved within session files only.
The ‘Pie Colors’ options are no longer available.