AceBrowser Enhancements - AGR displays
The AGR database uses a number of AceBrowser scripts that are based on standard scripts that come as part of the AceBrowser package, but which have been substantially modified.

Whilst these modifications frequently refer to specific data in AGR (which might not be present in other acedb databases), they may prove useful to other database curators who are wishing to customise their database.

This page details three of our most customised scripts that we use for displaying sequences (DNA), proteins, and author information. Note, use of these scripts will require modifications to the SiteDefs.pm file (you need to declare these additional display scripts and specify which databases will use them, and configure the mapper routine so that they are used properly).

Sequences
The standard way of displaying Sequence information in most acedb databases is the tree display. With AGR, we prefer to use our much customised 'AGR-sequence' script. Whilst this shows less information than what is present in the tree display, the layout is hopefully clearer, and more helpful (and of course it links back to the tree display with the (hopefully) useful 'More information...' hyperlink).

It is also possible to add to the information present in the tree display. E.g., whilst AGR contains the original EMBL database annotation for each sequence (minus the DNA sequence itself), this script allows links to be made to the full underlying record at the EBI and also to the equivalent GenBank record. These additional hyperlinks are not present in the tree display and indicate how external database links can be added if URLs can be formed with information present in the tree display (e.g. a sequence accession). This would be more useful if you do not have space in your database to store the original database annotation.

This script is also quite neat for viewing predicted coding sequences within larger sequences, e.g. look at this typical AGR coding sequence. The remarks at the top of the page would normally just show Text items within the tree display. The script has been changed though so that it also shows 'Evidence' and 'Gene function' information.

Proteins
Whereas the AGR-sequence script is a fairly straightforward hack of the the 'sequence' script, that comes with AceBrowser, the AGR-protein script is completely novel (though obviously based on the AGR-sequence script). Whilst AcePerl can connect to a database and retrieve a list of all features (along with coordinates) that are on a sequence, it cannot do the same for peptides. This makes it harder to grab the necessary info to display a table of peptide homologies in the same way that the AGR-sequence script does.

I solved this problem with a not very elegant solution which involves looking up all the homology info in the tree display and constructing a new table from this. This script also displays peptide sequences automatically if they are below a certain length.

cropnet-biblio
Again, this script is a modification of the basic 'biblio' script that comes with AceBrowser. It hopefully organises the results in a more intuitive fashion, and the external link to PubMed should be useful. Also useful is the fact that this script works out which sequences and patents are associated with each author and allows you to directly access them. Contrast an AGR author object when viewed with the old biblio script and the new cropnet-biblio script.