Hu Hai, Brzeski Henry, Hutchins Joe, Ramaraj Mohan, Qu Long, Xiong Richard, Kalathil Surendran, Kato Rand, Tenkillaya Santhosh, Carney Jerry, Redd Rosann, Arkalgudvenkata Sheshkumar, Shahzad Kashif, Scott Richard, Cheng Hui, Meadow Stephen, McMichael John, Sheu Shwu-Lin, Rosendale David, Kvecher Leonid, Ahern Stephen, Yang Song, Zhang Yonghong, Jordan Rick, Somiari Stella B, Hooke Jeffrey, Shriver Craig D, Somiari Richard I, Liebman Michael N
Windber Research Institute, 600 Somerset Avenue, Windber, PA 15963, USA.
Pharmacogenomics. 2004 Oct;5(7):933-41. doi: 10.1517/14622416.5.7.933.
The Windber Research Institute is an integrated high-throughput research center employing clinical, genomic and proteomic platforms to produce terabyte levels of data. We use biomedical informatics technologies to integrate all of these operations. This report includes information on a multi-year, multi-phase hybrid data warehouse project currently under development in the Institute. The purpose of the warehouse is to host the terabyte-level of internal experimentally generated data as well as data from public sources. We have previously reported on the phase I development, which integrated limited internal data sources and selected public databases. Currently, we are completing phase II development, which integrates our internal automated data sources and develops visualization tools to query across these data types. This paper summarizes our clinical and experimental operations, the data warehouse development, and the challenges we have faced. In phase III we plan to federate additional manual internal and public data sources and then to develop and adapt more data analysis and mining tools. We expect that the final implementation of the data warehouse will greatly facilitate biomedical informatics research.
