Toxicogenomics has the potential to reveal the molecular pathways and cellular processes that mediate the adverse responses to a toxicant. However, the initial output of a toxicogenomic experiment often consists of large lists of genes whose expression is altered after toxicant exposure. To interpret gene expression changes in the context of underlying biological pathways and processes, new bioinformatics methods must be developed. We have used global gene expression profiling combined with an evaluation of Gene Ontology (GO) and pathway mapping tools as unbiased methods for identifying the molecular pathways and processes affected upon toxicant exposure. We chose to use the acute effects caused by the non-genotoxic carcinogen and peroxisome proliferator (PP) diethylhexylphthalate (DEHP) in the mouse liver as a model system. Consistent with what is known about the mode of action of DEHP, our GO analysis of transcript profiling data revealed a striking overrepresentation of genes associated with the peroxisomal cellular component, together with genes involved in carboxylic acid and lipid metabolism. Furthermore we reveal gene expression changes associated with additional biological functions, including complement activation, hemostasis, the endoplasmic reticulum overload response, and circadian rhythm. Together, these data reveal potential new pathways of PP action and shed new light on the mechanisms by which non-genotoxic carcinogens control hepatocyte hypertrophy and proliferation. We demonstrate that GO mapping can identify, in an unbiased manner, both known and novel DEHP-induced molecular changes in the mouse liver and is therefore a powerful approach for elucidating modes of toxicity based on toxicogenomic data.