Statistical Framework for Quantitative Analysis of Array CGH

Over the last several years there has been an explosion of microarray technology in the biosciences, medical sciences, biotechnology, and pharmaceutical industry. The technology has centered on providing a platform for determining the gene expression profiles of hundreds to tens of thousands of genes (or transcript levels of RNA species) in tissue, tumors, cells, or biological fluids in a single experiment. In recent years, this technology has been extended to include the use of microarrays to study genomic DNA for gains and losses of chromosomal regions. This has become possible through the attachment of large genomic fragments such as BACs (bacterial artificial chromosomes). In this paper, we present a methodology to model a CGH (comparative genomic hybridization) profile as a statistical process and solve for distribution parameters to determine genomic changes across the genome, including whole chromosome gains and losses, and focal point variations that are commonly seen in solid tumors and genetic disorders