Web data sources constitute an important resource for Biological research. A simple tool that can retrieve information from different Web sites through a single interface and store the extracted data in a standardized format for efficient future use is critical to scientific discovery. In this paper we discuss an approach that combines agent and database technologies for biological data integration. To illustrate this, we employ two software tools: WinAgent, for building agents, and dbXML, for XML data management. WinAgent learns from its users by recording a browsing session on Web sites and successive data extraction from regions of interest on retrieved Web pages. The results are stored in a XML document and can be managed, queried and updated using a native XML database system such as dbXML. This approach is currently being evaluated at the Brain Tumor Cancer Unit of the Translational Genomics Research Institute (TGen), Phoenix, Arizona.