Quantitative Approaches to Understanding Cancer Genomes.

Abstract

Recent advances in technology have enabled the systematic, genome-wide analysis of cancer genomes, providing greater insight into the genetic basis of cancer development and a deeper understanding of the human genome. The focus of the current work is to identify genomic alterations potentially conferring risk for developing colorectal and breast cancers by performing genome-wide analysis with single nucleotide polymorphism (SNP) genotyping and next-generation sequencing (NGS) platforms. My first dissertation project involves deeply sequencing the genomes of individuals from a single family to identify novel mutations in hereditary mixed polyposis syndrome, a rare form of colorectal cancer with no known genetic basis. A novel candidate gene, ZNF426, was identified and decreased expression was confirmed in tumors from affected individuals. The second part of my dissertation evaluates methods for detection of somatic copy number alterations in colorectal cancer on chromosome 18 and the application of statistical methods for utilizing poor quality tumor data. Using genotyping and expression data from tumors, a variety of structural alterations were identified on chromosome 18. Additionally, I demonstrated the utility of applying new statistical methods to identity copy number alterations in array data with high background noise. The goal of my third project was to evaluate the contribution of consanguinity to breast cancer risk in Arab women without mutations in the BRCA1 and BRCA2 genes. The hypothesis in this study is that an increase in autosomal recessive genes responsible for genetic susceptibility to breast cancer is expected among families with consanguinity due to the increase in probability of sharing alleles identical-by-descent. Six unrelated individuals with breast cancer shared a 200kb overlapping region of homozygous SNPs on chromosome 9q332-33.3, which harbors an important candidate gene for cancer risk, LHX2. Whole-genome analysis allows for greater depth and higher throughput sequencing at lower costs, adding a new dimension to our understanding of cancer genetics. Future progress in these technologies and bioinformatics methods will improve the costs, sensitivity and accuracy of detecting mutations.