Dynamic information fusion for genome annotation

Abstract

The need for demand-driven and scaleable integration of heterogeneous data sources is inherent to the process of genome annotation. In this paper we describe the Gene-EYe architecture for data integration and analysis, presenting a platform for dynamic fusion of genome data. The architecture allows uniform access to a large amount of heterogeneous data sources and provides domain specific analysis tools. As genome data is known to be erroneous the ability for data cleansing is a natural part of the annotation process described. Gene-EYe also supports the task of keeping derived annotation data up to date in case of changes in the original data sources.