Human and mouse genome analysis using array comparative genomic hybridization

Almost all human cancers as well as developmental abnormalities are characterized by the presence of genetic alterations, most of which target a gene or a particular genomic locus resulting in altered gene expression and ultimately an altered phenotype. Different types of genetic alterations include: DNA copy number gains/amplifications, deletions, inversions, translocations, point mutations and epigenetic gene modifications. Over the years, different technological advances have made it possible to detect many of these in a high throughput fashion. One such technique, comparative genomic hybridization (CGH), allows detection of genetic aberrations that result in a change in DNA copy number. We introduced the first human genome-wide array CGH platform which could be used to detect both low-level DNA copy number aberrations including homozygous deletions as well as high level amplifications in cell lines and clinical samples (chapter 3). Given the accuracy and robustness by which one can quantitatively detect copy number aberrations, there are a plethora of applications for this technology. One can map recurrent regions of copy number aberration in a series of tumors and by doing so identify possible driver genes for tumor formation, progression or maintenance (chapters 6 and 8). One can also investigate the origin of metastases and the relatedness of local recurrences and second primaries (chapter 7). From a different point of view, one can start to investigate what the relationship is between underlying genetic defects and the types and numbers of DNA copy number aberrations present in the tumor (chapter 5). Using a mouse genome-wide array CGH platform, we show that different inbred and outbred laboratory mouse strains are characterized by the presence of multiple DNA copy number polymorphisms (chapter 4). The majority of these polymorphisms are recurrent among different individuals from one strain and all analyzed strains could be distinguished based on their polymorphism profile. We also introduce a novel approach for determining regions of heterozygosity in different backcross mice using array CGH. This approach is faster than conventional methods for determining regions of heterozygosity