Since it was set up, the MIG laboratory has worked on designing and developing databases. Databases constitute a major challenge for bioinformatics for structuring and exploiting the mass of data produced by genomic programs.
Our aim is to have at our disposal a coherent set of databases from the point of view of their design and their interfaces which will be the basis of the future information system of the unit.
The conceptual choices which we have made are :
The technical constraints which we have set concern on the one hand the use of free software enabling the distribution of the databases and their interfaces and on the other hand the use of software and standard modules enabling easy portage on different platforms. All the unit's databases are implemented on a PostgreSQL server (Relational Database Management System). The parsers and Web interfaces have been written in Perl using standard modules: (Database independent Interface) for connecting the database server, CGI (Common Gateway Interface) for the Web interfaces and BioPerl (Toolbox of Perl scripts for bioinformatics and genomics) for certain parsers.
We have already developed the following databases :
Other database projects are being developed in the laboratory. The most important is a relational database managing 3D structures of proteins extracted from the PDB bank. Apart from developing an information system which we mentioned previously, data relative to the 3D structures of proteins play a central role in analysing the 3D sequence-structure relation of the proteins, a subject of interest in the laboratory.