Investigation of health effects of low doses of radiation as a field of study has been riddled with difficulties since its inception. In this document we will use 100 mGy as the cutoff upper limit for low-dose radiation, borrowing this definition from the U.S. Department of Energy, although other agencies and researchers sometimes include up to five-fold higher doses under the same title. Difficulties in this area of research are most often ascribed to the fact that effects of low doses of radiation are subtle and difficult to distinguish from the plethora of other low-grade stresses. Thus, for example, most epidemiological studies include hundreds of thousands of samples and generate risk estimates that are statistically meaningful only when they are considered on a scale of hundreds or thousands of people. A logical approach to remedy the situation for low-dose research was to conduct well-controlled animal studies with hundreds of animals; nevertheless, even after many such studies were completed, our understanding of the biological basis for risk from low-dose radiation exposure is still not conclusive. In this paper we argue that the problem lies in the fact that our approach to animal studies is not comprehensive but conceptually binary. While some researchers apply epidemiological models to animal data, others look into molecular and cellular biology only. Very few studies are conducted to bridge this gap and consider how a realistic model of DNA damage could be integrated into a realistic model of radiation carcinogenesis.