A web server for mining Comparative Genomic Hybridization (CGH) data

• Published in: Data Mining, Systems Analysis, and Optimization in Biomedicine (AIP Conference Proceedings Volume 953) Vol. 953; pp. 144 - 161
• Main Authors: Liu, Jun, Ranka, Sanjay, Kahveci, Tamer
• Format: Journal Article
• Language: English
• Published: 01.01.2007
• ISBN: 9780735404670; 0735404674
• ISSN: 0094-243X
• DOI: 10.1063/1.2817337

Advances in cytogenetics and molecular biology has established that chromosomal alterations are critical in the pathogenesis of human cancer. Recurrent chromosomal alterations provide cytological and molecular markers for the diagnosis and prognosis of disease. They also facilitate the identification of genes that are important in carcinogenesis, which in the future may help in the development of targeted therapy.A large amount of publicly available cancer genetic data is now available and it is growing. There is a need for public domain tools that allow users to analyze their data and visualize the results. This chapter describes a web based software tool that will allow researchers to analyze and visualize Comparative Genomic Hybridization (CGH) datasets. It employs novel data mining methodologies for clustering and classification of CGH datasets as well as algorithms for identifying important markers (small set of genomic intervals with aberrations) that are potentially cancer signatures. The developed software will help in understanding the relationships between genomic aberrations and cancer types.