BioNetBuilder Plugin

What is BioNetBuilder?

BioNetBuilder is a Cytoscape plugIn that offers a user friendly interface to create biological networks integrated from several databases. Users can create networks for ~1000 different species, including most common model organisms and human. The interface offers different options to specify a set of initial genes/gene-products for which to find molecular interactions (including loading a simple text file, finding genes with specified Gene Ontology annotations, and finding genes whose common name match a given string). Depending on the user's computer capabilities, biological networks for whole organisms can also be created and displayed. Currently supported databases include: DIP, BIND, Prolinks, KEGG, HPRD, The BioGrid, and GO, among others. Users can set database parameters through the user interface to control which information from each database can be included in their networks (for example, interaction type, p-value, threshold, etc). BioNetBuilder also attaches to each gene cross reference IDs and URL links to public databases, including RefSeq, PIR, Prolinks, and The Yeast Resource Center.

What are the requirements to run BioNetBuilder?

Most computers will be ready to go!
You need to have Java 2 Standard Edition 1.4.x (later versions are probably fine) installed and an internet connection.
If you use the Java-Web Start link below, you will also need Java Web-Start. Most computers have Java Web-Start installed by default since it is included in the Java Runtime Environment.

How can I get BioNetBuilder?

Currently, the easiest way of using BioNetBuilder is clicking on this Launch BioNetBuilder link which will start Cytoscape on your computer with the plugIn loaded.

If you are a more advanced user, you can download the plugIn jar and place it within your Cytoscape plugins directory. If you do this, please download a recent version of Cytoscape 2.3. Important note: the plugIn jar cannot be loaded into earlier versions than Cytoscape 2.3.

If you download the plugIn jar and place it within Cytoscape's plugins directory to run it, you will need to place this properties file in your Cytoscape top directory (where cytoscape.jar is located).

The source code is licensed under the LGPL.

For anonymous svn access to development code:
   svn checkout https://err.bio.nyu.edu/svn/bionetbuilder/trunk/

NOTE: BioNetBuilder relies on Cytoscape libraries to build properly. Please also download that code and possibly adjust the CYTOSCAPE PATHS properties in the BioNetBuilder build.xml file.

How do I use BioNetBuilder?

On the Cytoscape Desktop, click on Plugins->BioNetBuilder Wizard... to start the wizard.

The plugIn was designed to be as user friendly as possible. We have developed a tutorial to aid in learning how to use it.

If you wish to use the HPRD database, please send me an email stating that you are not a commercial entity, and we will send you the necessary username and password. Also, if you want, please let us know how you heard about BioNetBuilder and what you're interested in using it for.

NOTE: As of 8-01-06, the username and password has changed for the HPRD database due to our server transition to NYU. If you are receiving "access denied" errors, this may be the reason. Please send an email to kdrew [at] nyu.edu for the updated username and password. Sorry for the inconvience.

What's next for BioNetBuilder?

BioNetBuilder was designed to be extensible and therefore adding new databases is very easy. We are looking into how to include several new databases to the set of supported databases. These databases may include: MINT (Zanzoni, et al., 2002), MIPS(Pagel, et al., 2005), MetaCyc(Caspi, et al., 2006), Predictome (Mellor, et al., 2002), RegulonDB (Salgado, et al., 2006) and others. If you have a perfered database that you would like to see integrated, you can tell us about it and we will put it on the list or you can develop the database handler by following our instructions and send us the java code.

Other Tools

Cytoscape has an excellent set of plugins which allows users to visualize and analyze many different networks. CyGoose is one such plugin which acts as an interface between Cytoscape and the Gaggle, a tool to create a data exploration and analysis environment for Systems Biology. Another plugin which allows users to query protein-protein interactions is cPath which is specifically geared towards cancer pathways.

Known problems

BioNetBuilder is in version 1.0 but still has some remaining problems. Here are the ones that we are aware of:

  1. When selecting an organism for a network build and then hitting the back button, the program sometimes does not reset and the state of the organism is saved even if another organism is selected. This can be correct by completing the wizard and building the network. If it is still a problem you can restart Cytoscape.

  2. When adding interactions to an existing network through the last panel of the wizard (ie. Network Name panel), sometimes the BioNetBuilder progress bar continues to run even after the interactions have been added to the network. You can safely close the progress bar window. If while dealing with large networks and you cannot verify all requested interactions were added, you can create a new temporary network and merge the two networks with Cytoscape's GraphMerge plugin.

  3. When loading in a file with gene ids other than GI numbers, occasionally translation does not occur and BioNetBuilder creates a network with all of the nodes but no interactions. This can be solved by loading a file with the corresponding GI numbers (ie. manually translating gene ids to GI numbers in the file).

  4. When starting the BioNetBuilder plugin, a message pops up stating "There were errors while starting the BioNetBuilder plugin. This probably means the server at [server address] could not be found or is not running". Please check to make sure the [server address] is correctly set to http://err.bio.nyu.edu/bionetserver . If it is correctly set and the message persists, please send an email to kdrew [at] nyu.edu . The server is occasionally overloaded by downloads of large networks and the server needs to be recycled.

Credit: Kevin Drew, Jay H. Konieczka, Iliana Avila-Campillo, Kevin Belasco, Alex Pine, John Lin, David J. Reiss, Richard Bonneau, Parker B. Antin
Brought to you by the Bonneau Lab at NYU and the Antin Lab at the University of Arizona

Please sumbit any comments or questions to the BioNetBuilder disscussion group.