This is the new server for GMOD.org. Please let us know if you notice anything weird while it's getting broken in.
LuceGene for Daphnia genome
Daphnia's genome now has lots of gene annotations, wanting both search and reporting for these. I've dusted off LuceGene which is designed for this, and installed over the Labor Day weekend at http://wfleabase.org/lucegene/. It was relatively easy to do: most of my time was in designing an XSLT stylesheet for gene page display. Other organism databases should be able to follow these steps, with more detail forthcoming, and have a similar gene search and report service without special effort.
- Convert Daphnia's gene annotations (some 28,000) from GFF to GeneSummary XML. This took a simple perl script and the GeneSummary XML package, at http://eugenes.org/gmod/gene-report-examples/. See bin/gff2ugpxml.pl. This is Daphnia specific, but should be easy to revise for other species GFF. You also want the lib/ and conf/ sections here to generate UGP-XML. See also the genepages project in SVN.
- install from the GMOD Lucegene distribution the web archive (lucegene.war), having already a Tomcat servlet server for this.
- copy and edit the conf/ugpxml.* configuration files, tweaking for Daphnia. The ugpxml.properties handles Lucegene indexing configuration, and that mostly worked but needed some field updates for Daphnia. The ugpxml.xslt for GeneSummary page reports needed the most work, to tune it to the specific gene annotations for this data set.
- also add Daphnia fasta gene and EST sequences to Lucegene search configurations, and the wfleabase.org web documents (updating current conf/ files for those).
- run the admin/lucegene-index.sh script to index the gene page XML, fasta sequences and web documents (or use admin/make-indices for all).
- update various html documents with Daphnia search examples.
- Restart Tomcat, already configured for the wfleabase.org webapps folder with this lucegene web archive. View results at http://wfleabase.org/lucegene/
Note that this is all what-you-see-is-what-you-get, in that the source gene page XML is exactly what is being searched, with no hidden or extra search tables, and is exactly what is sent to the client web browser. The browser does the boring job of converting it to HTML display (if it wants). Or one can use these gene pages in XML for data processing without tedious HTML page scraping.
Here is a Daphnia gene page example, http://wfleabase.org/lucegene/lookup?id=NCBI_GNO_292134 (view the page source to see structured gene page XML). This corresponds to these screen shots daphnia gene page and gene page xml.
Dongilbert 16:23, 4 September 2007 (EDT)