BioStar models of clinical and genomic data for biomedical data warehouse design. (28th April 2005)

• Record Type: Journal Article
• Authors: Wang, Liangjiang; Zhang, Aidong; Ramanathan, Murali
• Is Part Of: International journal of bioinformatics research and applications. Volume 1:Number 1(2005)
• Page Start: 63
• Page End: 80
• Publication Date: 2005-04-28
• Subjects: clinical data -- genomic data -- data integration -- multidimensional data modelling -- biomedical data -- data warehouse design -- bioinformatics -- data warehousing -- online analytical processing -- data mining -- computational biology; Bioinformatics -- Periodicals; 570.285
• Languages: English
• ISSNs: 1744-5485

Abstract

Biomedical research is now generating large amounts of data, ranging from clinical test results to microarray gene expression profiles. The scale and complexity of these datasets give rise to substantial challenges in data management and analysis. It is highly desirable that data warehousing and online analytical processing technologies can be applied to biomedical data integration and mining. The major difficulty probably lies in the task of capturing and modelling diverse biological objects and their complex relationships. This paper describes multidimensional data modelling for biomedical data warehouse design. Since the conventional models such as star schema appear to be insufficient for modelling clinical and genomic data, we develop a new model called BioStar schema. The new model can capture the rich semantics of biomedical data and provide greater extensibility for the fast evolution of biological research methodologies.