See scripts/buildPathVisio.sh for protocol for processing of CS databases into Derby databases.

NOTES:
 * For Sc databases: an if statement in buildPathVisio.sh gets http://downloads.yeastgenome.org/curation/chromosomal_feature/SGD_features.tab and loads the first and last columns
into MySQL to overwrite SGD primary IDs.  Ensembl only provides the symbol redundantly as both primary_id and display_id. The ETL 
device has collect both and included symbol in the attribute table. So, by overwriting the primary_id in the 'gene' and 'attr'
tables, the databases will be fixed.  The proper SGD IDs are important to support the majority of yeast pathways already in WikiPathways.

* For copying derby and sql files to bridgedb.org, you can use the following command to copy just the last 5 (adapt as necessary) created databases:
 scp `ls -lht *bridge | head -5 | awk '{print $9}'` bridgedb@bridgedb.org:data/gene_database/pre-release/.
 scp `ls -lht *sql.dump.gz | head -5 | awk '{print $9}'` bridgedb@bridgedb.org:data/gene_database/sql/.

* For copying derby files to wikipathways.org, you can use the following command to copy just the last 5 (adapt as necessary) created databases:
 scp `ls -lht *bridge | head -5 | awk '{print $9}'` wikipathways@www.wikipathways.org:database/. 

* For creating symbolic links to the derby databases, use LinkDatabases.pl. It detects all *.bridge files and asks if you want to link them all or run interactively. If you choose to link a particular database it will first remove any preexisting symlink and the folder it was linking to, then unarchive the database and create a new symlink. It uses the HashSpeciesList Perl module (http://svn.bigcat.unimaas.nl/bridgedb/trunk/dbbuilder/src/org/bridgedb/extract/HashSpeciesList.pm), which locates a species list and loads it into a hash.
