Copy Number Analysis of Cancer

By accurately describing cancer genomes, we may link genomic mutations to phenotypic effects and eventually treat cancer patients based on the molecular cause of their disease, rather than generalizing treatment based on cell morphology or tissue of origin. Alteration of DNA copy number is a driving mutational process in the formation and progression of cancer. Deletions and amplifications of specific chromosomal regions are important for cancer diagnosis and prognosis, and copy number analysis has become standard practice for many clinicians and researchers. In this thesis we describe the development of two computational methods, TAPS and Patchwork, for analysis of genome-wide absolute allele-specific copy number per cell in tumour samples. TAPS is used with SNP microarray data and Patchwork with whole genome sequencing data. Both are suitable for unknown average ploidy of the tumour cells, are robust to admixture of genetically normal cells, and may be used to detect genetic heterogeneity in the tumour cell population. We also present two studies where TAPS was used to find copy number alterations associated with risk of recurrence after surgery, in ovarian cancer and colon cancer. We discuss the potential of such prognostic markers and the use of allele-specific copy number analysis in research and diagnostics