Drug Databases

Although database systems that cover metabolic and gene regulation networks are already available and mostly freely accessible for academic research, drug databases are not, and recent data has to be acquired from proprietary data sources, e.g., literature databases or lists of agents and medical products. In addition, public databases of medical products do not contain the real names of the agents but the trademarks (more than 9000 products are registered in Germany alone). To analyze drug actions in the context of metabolic and gene regulation networks, modelling and implementing of custom databases is necessary. This section will give an overview about the modelling of static relationships in the certain domains.

Corresponding to our inhouse database DrugDB, the first case study is a drug database which provides data about biochemical mechanisms of drug action and their direct targets, as well as the resulting effects. Because of the highly complex relationships in DrugDB, only a subset its structure is summarized in Figure 11.4. In fact, the database is finely granulated and contains a total of 53 classes. In a static conceptual schema a classes defines a special type of objects having well defined properties. Edges between the classes of objects indicate that the two classes are related to each other. Naming the relationship will define its semantics. Different types of relationships can be defined, e. g. association, aggregation or inheritance. In the following examples only aggregations and associations are included.

Most drugs have a designated target, which is often highly specific to the drug. Administration of the drug results in activation or deactivation of the target entity. The mechanisms of action of each drug are stored in the database and linked to the resulting changes in metabolic systems, which are often the actual goal of administering the drug. The goal of this database is to classify and reveal the activities of different drugs and to link to the biochemical networks that they influence.

The access to drug databases integrated with metabolic data gives us information about the activation of alternative pathways around metabolic blocks. The possibility

Fig. 11.4 Conceptual scheme of a drug database.

of searching for drugs that can modulate these pathways at different biochemical levels should offer new possibilities for the treatment of metabolic diseases.

0 0

Post a comment