A Metadata-Based Infrastructure for Integrating Analyzing and Visualizing Data in a High Throughput Influenza Laboratory
A Metadata-Based Infrastructure for Integrating, Analyzing, and Visualizing Data in a High Throughput Influenza Laboratory
Daniel Hecht, University of Applied Sciencies Wiesbaden
Prof. Dr. Reinhold Schaefer, University of Applied Sciencies Wiesbaden
Alexander Roth, University of California Los Angeles
The functionality for integrating multiple data sources, analyzing their data, and visualizing the resulting models is provided by many single tools on the market. The infrastructure we developed integrates all these functions into a single workflow using a highly flexible meta model. In general, this meta model provides core functionality for the generation of a web portal that allows highly compressed but also a highly detailed representation of various information related to the avian influenza virus and the High-Throughput Laboratory at the UCLA. In detail, the meta model describes the appearance of graphical user interfaces, the functionality of visualizing components, the structure of data which is processed, and enables the integration with a data warehouse solution.
Basis of the infrastructure is a data warehouse solution which is integrated using a custom data format consisting of generic data containers. This enables support for any desired data formats and data sources and thus, data analysis of all imported data using data mining tools and external data analysis tools which can be plugged-into the data warehouse. Data analysis results can be visualized based on a plug-in concept which integrates separately developed software components. They are developed focusing on functionality required for data presentation while benefiting from meta model’s core functionality automatically. Their configuration takes place using web-based software components which define (personalized) graphical user interface without requiring any technical knowledge. A role-based security model which is integrated into the meta model manages application configuration, data access, user control and administration functions.
The software prototype currently developed provides all functionality in a platform- and technology-independent fashion because of the usage of standardized markup languages and the implementation of a service oriented architecture.