Revision as of 18:03, 24 August 2007

Hackathon wikidb components

• middleware parts:
  • Chado to wiki:
    • modware to select gene attributes by gene name, print genes as wiki-string; Eric
    • wikiloader to add to create gene page, select gene page/table template, add gene wiki-string; Jim

• • Wiki to Chado:
    • XORT/chado xml scripts to load output of wiki/wikidb tables to chado; Josh

loadwiki.php

 usage: php loadwiki.php -p page_template -t table_template -f input_filename
 page_template == gene page template for wiki
 table_template == table edit template inside gene page
 input_filename == gene data in wiki-string format

input file (one line, for wiki table with '||' delimiters for wiki table columns)

 sadA    sadA||EGF repeat-containing 9 transmembrane molecule involved in substrate adhesion.||Jim, Don

or

 $gene_name."\t".$gene_name.'||'.$description.'||'.$synonym_string."\n";

We plan to extend the above to work with a fuller gene 'page' of output from chado. THis will use one common wiki Template:gene_page. This page template will have information linking the chado table output fields with the gene wiki table templates.

Planned outcome

Simple example to collect gene(s) information from Chado db, produce intermediate Wiki-text file (script 1). This is then loaded into Mediawiki database with gene page templates (script 2). Community folks edit the genes thru Table Edit mechanism as desired. Then updated gene info is dumped (from mysql wikidb), converted to chado xml, then loaded into Chado with transaction update checks, via XORT (script 3).

Genome wiki from chado notes

- From hackathon

• tasks:
  • locate sample chado data (some format) for some genes w/ attributes
  • convert to some format suited to wiki loading (as wiki xml?)
    • dump table via Chado SQL;

 see e.g. http://eugenes.org/gmod/genbank2chado/conf/v_genepage3.sql

• • • via xml/xslt transforms
    • via XORT perl parser
    • other

• • load to wiki

    >> this is larger;loading into wikipedia db via wikipedia.xml
         

• • dump wiki table edit (mysql db)

• • convert to chado xml (? xml transforms)

  ** flybase harvard has scripts for general bulk data to chado.xml  

• options:
  • use chado sql view/procedure to dump tables suited to wikibox_db ?
  • easier