Genetic polymorphisms in drug metabolizing enzymes and drug–drug interactions are major sources of inadequate drug exposure and ensuing adverse effects or insufficient responses. The current challenge in assessing drug–drug gene interactions (DDGIs) for the development of precise dose adjustment recommendation systems is to take into account both simultaneously. Here, we analyze the static models of DDGI from in vivo data and focus on the concept of phenoconversion to model inhibition and genetic polymorphisms jointly. These models are applicable to datasets where pharmacokinetic information is missing and are being used in clinical support systems and consensus dose adjustment guidelines. We show that all such models can be handled by the same formal framework, and that models that differ at first sight are all versions of the same linear phenoconversion model. This model includes the linear pharmacogenetic and inhibition models as special cases. We highlight present challenges in this endeavor and the open issues for future research in developing DDGI models for recommendation systems.