Solving Structure with Sparse, Randomly-Oriented X-ray Data

摘要

Single-particle imaging experiments of biomolecules at x-ray free-electronlasers (XFELs) require processing of hundreds of thousands (or more) of imagesthat contain very few x-rays. Each low-flux image of the diffraction pattern isproduced by a single, randomly oriented particle, such as a protein. Wedemonstrate the feasibility of collecting data at these extremes, averagingonly 2.5 photons per frame, where it seems doubtful there could be informationabout the state of rotation, let alone the image contrast. This is accomplishedwith an expectation maximization algorithm that processes the low-flux data inaggregate, and without any prior knowledge of the object or its orientation.The versatility of the method promises, more generally, to redefine whatmeasurement scenarios can provide useful signal in the high-noise regime.