Main content area

CanGEM: mining gene copy number changes in cancer

Scheinin, Ilari, Myllykangas, Samuel, Borze, Ioana, Böhling, Tom, Knuutila, Sakari, Saharinen, Juha
Nucleic acids research 2008 v.36 no.suppl_1 pp. D830
Internet, carcinogenesis, comparative genomic hybridization, data collection, databases, gene dosage, genes, genome mining, humans, microarray technology, neoplasms, nucleic acids, screening
The use of genome-wide and high-throughput screening methods on large sample sizes is a well-grounded approach when studying a process as complex and heterogeneous as tumorigenesis. Gene copy number changes are one of the main mechanisms causing cancerous alterations in gene expression and can be detected using array comparative genomic hybridization (aCGH). Microarrays are well suited for the integrative systems biology approach, but none of the existing microarray databases is focusing on copy number changes. We present here CanGEM (Cancer GEnome Mine), which is a public, web-based database for storing quantitative microarray data and relevant metadata about the measurements and samples. CanGEM supports the MIAME standard and in addition, stores clinical information using standardized controlled vocabularies whenever possible. Microarray probes are re-annotated with their physical coordinates in the human genome and aCGH data is analyzed to yield gene-specific copy numbers. Users can build custom datasets by querying for specific clinical sample characteristics or copy number changes of individual genes. Aberration frequencies can be calculated for these datasets, and the data can be visualized on the human genome map with gene annotations. Furthermore, the original data files are available for more detailed analysis. The CanGEM database can be accessed at