Inductive Databases for Bio- and Chemo-Informatics

Inductive databases are a new form of databases in which one does not only store data but also patterns or regularities. A user can generate, manipulate and query for patterns of interest using an inductive query language. Data mining then becomes an interactive querying process. The inductive database framework is especially important for bioand chemoinformatics because of the existence of a large number of databases and the need to mine them in these research fields. Two examples of inductive databases are presented: MOLFEA and PROFEA. MOLFEA is the Molecular Feature Miner that mines for linear fragments of interest in the 2D-structure of chemical compounds. PROFEA mines for fragments of interest in the secondary structure of proteins. MOLFEA and PROFEA are examples of inductive databases that generate patterns in the form of strings. The underlying principles of this class of inductive databases are presented.