Technical Overview, dblist gpmnum.pl

---

Revision as of 19:27, 17 February 2009

This script searches the GPMDB for the twenty most reliably identified proteins in a specific result file.

The one required argument is gpmnum, a GPMDB result file identifier of the form "GPMxxxNNNNNNNN", where "xxx" is one of the three-digit identifiers of the search machines that form a part of the GPM, and the "NNNNNNNN" is a sequentially assigned number unique within the "xxx" machine. Two optional arguments are proex and ltype. If proex is not supplied, it defaults to -1, and is used solely in forming the links that lead to protein-level information for identifications in this file. If the ltype argument is not supplied, it defaults to 2; this value is both passed on through links to protein-level information pages, but also controls the display of the label elements; choices that can be made are "ensembl", "ipi", "symbol" (the default), "ncbi" and "swiss".

Display Elements

• a list of links to result file level information.
• a "show all" link to dblist_gpmnum_all.pl, to see all proteins from this result file.
• a dropdown menu to change the label display type.
• the per-protein lines of information.

The per-protein pieces of information are:

• the protein order number.
• the expect value of this protein identification.
• the accession number, according to the chosen label type.
• a coverage diagram for this protein identification. This is an Adobe SVG element depicting the sections of the protein which are covered by the peptides identified for this protein in the data file.
• an AJAX-generated identification comparison, comparing the quality of this identification to all other identifications of it.

This script uses the following tables: result, protein, peptide, proseq, paths.