UniDPI

This is the official UniDPI home page.

Version

December, 2005

Readme

UniDPI stands for Unified Database of Protein Interactions which gives access to documented interactions in several independent and heterogeneous databases disseminated on the web. As other interactions databases it contains experimental information about protein-protein interactions. However UniDPI has several features that distinguish it from those databases, these are:

1) UniDPI integrates information stored in the main protein interactions databases. The first version (1.0) contains interactions extracted from: DIP (http://dip.doe-mbi.ucla.edu/), BIND (http://bind.ca), MINT (http://mint.bio.uniroma2.it/mint/), GRID (http://www.cs.man.ac.uk/grid-db/) and IntAct (http://www.ebi.ac.uk/intact/).

2) For must of the proteins present in UniDPI we have added Uniprot information. Features such as: function, cellular location, enzymatic activity, Pfam domains, structure, related diseases, and other comments are available in UniDPI. This feature, joined with the full text index engine inherited from the Exist XML database; significantly enhance search capabilities of our database.

3) We have included in UniDPI a form for making correlations between any protein properties described as its corresponding UniProt identifier. Through the experimentally detected interactions network, the user can measure how often one property co-appear with another (e.g. different but related diseases).

4) Finally we have adapted the Cytoscape network analysis tool (http://www.cytoscape.org/) to be used inside UniDPI. We have built interaction networks for each of the must interesting UniProt features using as backbone the protein interactions networks. Taking advantage of the Cytoscape tools, the user can analyze the whole network, detecting important relations between some features like: Pfam and InterPro domains, transcription factors, diseases etc.

UniDPI is freely distributable under the GNU Public License (GPL), database and application's source code are available in our download page.

Protein Search

Use this form to search for protein interactions through any protein properties. UniDPI is a full text indexed database, so you can use any keywords to search against the interactions. You can take in advantage this feature by finding interactions related with specific diseases (e.g. use 'HIV' to search for interactions that are related to this disease) or making complex queries that relate different properties (e.g. searching by the query 'P53 cancer' shows experimental evidences that confirm the relation between the P53 transcription factor and cancer disease). In the results page you can filter the output adding any keywords to the previous query; it is also possible to filter the reported interactions by the number of experiments that confirm each interaction. This is a way of tuning the accuracy of the results.

Correlation Search

Through this search it's possible to correlate the simultaneous presence of related protein properties in the interactions. For one query property, expressed as it UniProt identifier, the system searches for all the properties that coexist with the queried property in the interactions and reports the percent of occurrences (e.g. use 'PF00262' to search which are the properties that coexist with the Calreticulim family). In the results page you can filter the output to display only specific properties (e.g. type InterPro to select only the related InterPro accession numbers).

Network analysis

UniDPI allows the analysis of the protein interactions networks from an original point of view. Focusing in one property present in the UniProt database it's possible to build the interaction network associated with this feature. We have built interactions networks for the following UniProt database crossreferences: GO, Pfam, Interpro, PDB, Reactome, HSSP, MIM and TRANSFAC. For the analysis ands visualization of these networks we have adapted Cytoscape. To lunch Cytoscape it's only required to have J2SE 1.4.1 or higher installed in your system.

Authors

Erik Torres <erik at bioinfo.cu>
Victor de la Torre <victor at bioinfo.cu>
Tirso Pons <tirso.pons at cigb.edu.cu>

Copyright

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

UniDPI-devel mailing list is the mechanism for end-users to utilize when reporting support issues to the project.

This is the right place for discussing about UniDPI development and usage. You are encouraged to share your opinion with the members of the list. Please note that our mailing list is a public forum. All mails sent to them will be archived publically, you may not post copyrighted texts without a license etc.

If you'd like to subscribe to UniDPI-devel mailing list, you can send an email (subject do not matter) to unidpi-devel-request@lists.sourceforge.net with the word subscribe in the body.

UniDPI© 2003
National Bioinformatics Center, Cuba