The rise in large-scale cancer genomics data collection initiatives has paved the way for extensive research aimed at understanding the biology of human cancer. While the majority of this research is motivated by clinical applications aimed at advancing targeted therapy, cancer prevention initiatives are less emphasized. Many cancers are not attributable to known heritable genetic factors, making environmental exposure a main suspect in driving cancer risk. A major aspect of cancer prevention involves the identification of chemical carcinogens, substances linked to increased cancer susceptibility. Traditional methods for chemical carcinogens testing, including epidemiological studies and rodent bioassays, are expensive to conduct, not scalable to a large number of chemicals, and not capable of detecting specific mechanisms of actions of carcinogenicity. Thus, there exists a dire need for improvement in data generation and computational method development for chemical carcinogenicity testing. Here, we coin the term "carcinogenome" to denote the complete cancer genomic landscape encompassing both its repertoire of environmental chemical exposures, as well as its germ-line and somatic mutations and epi-genetic regulators. To study the carcinogenome, we analyze both the genomic behavior of real human tumors as well as profiles of the exposome, that is, data derived from chemical exposures in human, animal or cell line models. My thesis consists of two distinct projects that, through the generation and innovative analysis of multi-omics data, aim at advancing our understanding of the molecular mechanisms of cancer initiation and progression, and of the role environmental exposure plays in these processes. First, I detail our effort at data generation and method development for characterizing environmental contributions to carcinogenesis using transcriptional profiles of chemical perturbations. Second, I present the tool iEDGE (Integration of Epi-DNA and Gene Expression) and its applications to the integrative analysis of multi-level cancer genomics data from human primary tumors of multiple cancer types. These projects collectively further our understanding of the carcinogenome and will hopefully foster both cancer prevention, through the identification of environmental chemical carcinogens, and cancer therapy, through the discovery of novel cancer gene drivers and therapeutic targets.