Apart from the problem of having to consolidate information of different types of experiments, the storage of experimental data is made more complex when we consider that experimental data are often stored in different ways (pictures, tables, or text) and in different formats (flat files, databases, html files or even XML files). This probably explains why only few of the public available databases contain experimental data, and those that do generally contain only one type of experimental data, like expression profiles, enzyme properties or 2D gel electrophoresis (e.g. Brenda - enzymatic activities, enzyme properties; EcoCyc - expression profiles; KEGG - expression profiles; SWISSPROT -2D gel electrophoresis).
We are working on a database system that allows the comparison of experimental data from different sources and types. The data model has been created in such a way that it permits the storage of different types of experimental results (e.g. expression data, enzyme activity and time line data in general). The experimental data model has been built on the basis of an existing database modelled for the storage and management of data related to biochemical reactions. This extension allows us to present experimental results in the context of biochemical pathways. In order to keep track of the origin of the information stored, the meta-information (information about the method(s) used, the author(s), date, source, etc.) about the experimental results is also stored.
Complementing the database system we have developed a system for the visualisation of the experimental data. So far the system contains information related to 2D Gel electrophoresis, arrays and enzyme activity experiments. The system allows the user to combine information from experiments with other, related, information about biochemical reactions and its participants. For example, if the user imported data from a protein analysis experiment by 2D gel electrophoresis the gel spots are linked with information about the corresponding protein in the database. The user can then retrieve information about the protein (e.g. sequence, functional annotation, etc.), the corresponding gene or view the function of the corresponding protein within a pathway. One other feature is the chance to merge information from 2D gel electrophoresis with expression profiles of the same organism on the level of pathways. In this way the user can get an impression on how reliable the existence of a given reaction is within a particular organism.
