BiblioList in depth

In this section, we will detail a little bit more how BiblioList© really works and what's under!
First of all, BiblioList© is a web application written in pure Perl using Catalyst framework. It follows MVC (Model View Controller) architectural pattern.

As mentioned above, the first import must be made manually by the user (red arrow). To do it, a Perl script is provided. It requests PubMed litterature for an organism of interest and fetch it, parse the results (in XML format) and load BiblioList© database with it. In order to do it, the BiblioList© database schema (See database section) must be created first and sufficient right for the user must be supplied. Once this script has been ran, another script must be ran to create links between genes and publications. The most the organism has publications the most the script will take time, so be aware of it.
After these manual steps (only running scripts), your litterature is ready to be curated by users through BiblioList© web interface. NOTE: The user who curate the litterature must have sufficient role(s) to manage bibliography.

As soon as the user wants to access BiblioList© to manage litterature for his/her organism of interest, (s)he wiil need to login. If no account exists yet, the user can create it. However, its rights will be restricted to only view data from the database. To be able to manage litterature, (s)he will need to contact the administrator of the site to get enough rights. After that, the user will have to choose a gene of interest. This search is performed by a search box where the user can dig the database for a gene name or a gene description. If results are found, then the curator will be faced to 4 categories of link between a gene and a publication:
Gene name is found in abstract, best link (light green)
Gene name is found in abstract, but need to be curated (dark green)
Words from gene description are found in the abstract with a cretain distance
and have been curated by user (orange)
Words from gene description are found in the abstract with a cretain distance
and haven't been curated by user (red)
At this step, the curator just need to check or uncheck link between a gene and the publication. Also a direct link exists to completely remove and don't display the publication. This is how easy it is to curate the bibliography for an annotated organism using BiblioList©! Now, the remote database, containing annotations, is simply updated and/or populated with the new curated litterature by just clicking a button that automatically insert the data to the remote database schema. That's all!