PROJECT SUMMARY Precision medicine requires an understanding of the origins and molecular control over complex traits and disease. The field is largely driven by human genetics, which adheres to a 1918 dogma that phenotype is determined solely by genetics and the environment. Yet, evidence from monozygotic twins and isogenic animal models indicate that up to 50% of phenotypic variation across diverse physiological traits and diseases cannot be explained by genetics or environment – there is something `more' that is unique to each individual, and that cannot be determined by analyzing population-level mean effects. These findings also indicate that even if we did have `complete' genetic and environmental knowledge, a substantial portion of disease heterogeneity would remain unaccounted for. The operating hypothesis for this project is that a substantial fraction of unexplained disease heterogeneity reflects inherently probabilistic properties of the biological system that lead to fixed, deterministic, real biological variation. There is compelling evidence for an evolved molecular circuitry that controls phenotypic variability as a quantitative trait. Thus, understanding variability as a quantitative trait is essential to understanding the etiology of phenotypic diversity (in general) and an individual's disease potential (in particular). Here, we will begin to finally answer the precision medicine questions of: what is the normal or expected disease potential for me? And, what are the origins and regulatory controls of non-genetic, non- environmental phenotypic and disease variability in humans? The first steps towards addressing these questions and identifying mechanisms through which probabilistic processes lead to disease heterogeneity is to create a catalogue of putative variance regulators and genes; a phenotypic, epigenetic, and cellular variance atlas charting the landscape of probabilistic variation in an isogenic model system (mice); and, to demonstrate that the regulatory architecture of variance control is conserved between mouse and humans. If it is true that a significant portion of unexplained disease heterogeneity is due to the molecular control of variability itself, then we will have uncovered an entirely new area of disease etiology that can be harnessed by the community to develop fundamentally new predictive, diagnostic, and therapeutic interventions, irrespective of the disease of interest.