Conclusion

Bioinformatics data have special characteristics: they are stored in distributed databases, are highly heterogeneous, reflect expertise and form a complex web-based network. Jointly exploiting a data set, i.e. integrating these data, is therefore a particularly difficult task, since it involves dealing with data that may be contradictory and very often redundant. Nevertheless, the advancement of knowledge in biology depends on exploiting the wealth of this highly complementary information.

In this article, we provide an overview of existing databases and integration approaches. We have shown that certain easy-to-use solutions (portals) are well-suited to point-in-time information retrieval, since they are capable of unifying syntactically heterogeneous data. However, data warehousing solutions, which are heavier to implement, are the only ones capable of tackling...