Cryogenic-sample electron microscopy (CryoEM) has become a leading technique for determining the structure of biological macromolecules. However, many biological molecules are structurally heterogeneous, occupying a broad range of possible structures. In this talk, I present a novel approach to analyzing Cryo-EM that attempts to estimate the structural probability distribution of biomolecules. Our work casts the analysis of Cryo-EM data as a Bayesian Inverse problem and employ structural modeling tools to construct a physical meaningful prior into our posterior. Proof-of-concept suggest that our approach could recover the probability distribution underlying a biomolecules conformational degrees of freedom, even in the presence of disorder. Finally, I discuss ongoing work on accelerating the evaluation of the likelihood using fast algorithms and dynamically tuning the ensemble of protein structures.