A Data Warehouse Approach to Semantic Integration of Pseudomonas Data

Biological research and development are routinely producing terabytes of data that need to be organized, queried and reduced to useful scientific knowledge. Even though data integration can provide solutions to such biological problems, it is often problematic due to the sources’ heterogeneity and their semantic and structural diversity. Moreover, necessary updates of both structure and content of databases provide further challenges for an integration process. We present a new biological data warehouse for Pseudomonas species “PseudomonasDW” to integrate annotation and pathway data from highly different resources. The combination of knowledge from multiple disciplines and sources should advance the understanding of cellular processes and lead to the prediction of cellular behavior in its entirety. The key aspect of our approach is the combination of a materialized and a virtual data integration to exploit their advantages in a new hybrid approach. The data are extracted from the original data sources using SB-KOM (System Biology Khaos Ontology-based Mediator) and then stored locally in the data warehouse to ensure a fast performance and data consistency.