The role of declarative languages in mining biological databases

Biological and biomedical databases have become a primary application area for data mining. Such databases commonly involve multiple relational tables and a variety of data types, as in the biological databases that formed the basis for the KDD Cup 2001 and 2002 competitions. The diversity of such "multi-relational" data is likely to increase dramatically in the near future. For example, patient records at major medical institutions are being augmented to include a variety of genetic data, including data on single-nucleotide polymorphisms (SNPs) and mRNA levels from gene expression microarrays, in addition to clinical data. Data mining tools based on declarative languages are able to naturally integrate data of diverse types, from multiple tables, to arrive at novel discoveries.

Duke Scholars

Author David Page Biostatistics & Bioinformatics, Division of Biostatistics