Multi-State Modeling Of Biomolecules By Satisfaction Of Spatial Restraints From Single-Particle Electron Microscopy Images

Single proteins and macromolecular complexes can exist in multiple conformational and compositional states. Accurate modeling of these ensembles of states is key to understanding and modulating the molecular function. Single-particle electron microscopy (EM) can produce a wealth of structural information. Here, we describe an integrative method for computing a multi-state model of a biomolecular system, based primarily on input particle images from EM. The approach also leverages prior knowledge about the stereochemistry of molecules, image noise, and potentially other sources of information.