| Although it has long been appreciated that the human genome contains a size continuum of genetic variation ranging from single-nucleotide changes to large (> 3 Mb), microscopically visible karyotype alterations, the abundance and importance of mutation in chromosome structure between these two size extremes was essentially unknown five years ago. The gain or loss of DNA sequence at a given locus is often referred to as copy number variation (CNV). In this dissertation I describe statistical techniques to ascertain the location of CNVs in the genome, genotype them, and assess their functional impact. In chapter 2 I propose a simple method to identify deletions from SNP genotype calls, which I applied to data from the International HapMap Project to create the first high-resolution survey of copy number variation in the human genome. In chapter 3, I describe an adaptation of this method to identify cell line deletions that are the result of somatic, instead of germline, mutation processes. Chapter 4 consists of my attempts to develop CNV genotyping algorithms that cluster noisy array-based measures of DNA copy number into discrete genotype "calls". These make use of additional information, such as SNP genotypes and familial relationships, when available. Finally, in chapter 5 I assess the phenotypic impact of copy number variation in the Hutterites, by examining the genomewide association of CNVs with the numerous quantitative traits that have been measured in this population. |
