The treatment of many cancers centers on risk-adapted therapy. Patient stratification methodologies are, therefore, essential to the design of effective treatment strategies. Our research focus is to develop transcriptional regulatory networks of cancer that can predict a patient’s response to various therapies and their risk of progression. This approach associates mechanistic underpinnings to the various disease states of a given cancer type. Thus, it enables rational approaches to directing high-risk transcriptional states to become low-risk. In this way, we hope to not only identify the best available treatment for a given patient, but to discover new therapies for effectively modulating gene regulation.