|
The PLASdb web site is implemented using the programming language PERL and the database management software ORACLE 8i. The relational schema consists of 17 tables and is available here.
In the database, each gene product has a property that indicates its phylogenetic fingerprint.
Specific information about each gene product from Arabidopsis was obtained from The Institute for Genomic Research (TIGR). These data are available in the files ftp://ftp.tigr.org/pub/data/manatee/mysql_data/euk_manatee and ftp://ftp.tigr.org/pub/data/a_thaliana/ath1/PSEUDOCHROMOSOMES. These data include the following properties: annotation, molecular weight, isoelectric point, clone identifiers, gene ontology, and PFAM statistics. The automatic functional assignments for the Arabidopsis genes represented in PLASdb were obtained from the MIPS Arabidopsis thaliana database (http://mips.gsf.de/proj/thal/db/index.html). Additional properties were obtained using the programs TMHMM for prediction of transmembrane helices and TargetP for prediction of subcellular localization as described in the Supplemental Material. In addition, gene family information has been provided using the BLASTCLUST program from NCBI as described in the Supplemental Material. All the data was automatically entered into the database using custom-made PERL programs.
The user interface consists of HTML code generated by PERL programs using the PERL/DBI interface to the Oracle database. The design goal was to allow the user to employ filters to restrict the gene list to categories of interest and to allow these shortened lists to be sorted by various gene product properties. The filtering was achieved using SQL statements and the sorts were done using the PERL sort function against row sets returned from the database.
| 
