[Bioperl-l] (no subject)

[David Burdick]

Gabriel-

At the ISB, we've created a software tool called “Connect the Dots” that can make translations between databases that share identifiers. There is an example web interface at: http://t1dbase.org/cgi-bin/dotHome.cgi
though the software is primarily developed for use programatically where you parse only the databases you need. We currently have parsers for 26 databases that collect mostly identifiers (as opposed to major content), and new parsers are easy to write. 

To use Connect the Dots to translate identifiers, a small query language is used to join databases on identifier and extract the desired columns from each database into a relational table. There are a variety of options for output including flat files and XML. The software is built on the PostgreSQL database, and can be found at sorceforge: http://cvs.sourceforge.net/viewcvs.py/isbiology/connectDots_postgres/
We should have CPAN release soon, though for now you can download the software and read the documentation in the ConnectDots.pm file. It is written in Perl.

If you have questions, feel free to email me at dburdick at systemsbiology.org

Dave

> From: "Gabriel Berriz" <bucket_ospam at lycos.com>
> Subject: [Bioperl-l] Gene/protein ID synonyms
> 
> One recurrent headache for much of our bioinformatics development comes from
> the fact that for most model organisms any one gene or protein can be
> identified by several different IDs (e.g. for yeast genes one has common
> names, ORF names, SGD IDs, etc.).  We want to give users the option to refer
> to genes and proteins according to as many standards as possible, which
> means that our software has to be able to translate between all these
> standards and its own internal representation.  Moreover, this translation
> facility has to keep pace with updates to the source databases.
> 
> Are there Bioperl modules to deal with this problem?  If not, what would be
> a better mailing list or forum to enquire about available solutions to this
> problem?  I am aware of SRS, which some people find very useful for this
> kind of translation, but I'd like to learn more about what else is out
> there.
> 
> Many thanks,
>
> Gabriel