Two Datasets Are Better Than One: Method of Double Moments for 3-D Reconstruction in Cryo-EM

The introduction of the method of double moments (MoDM) marks a significant advancement in cryo-electron microscopy (cryo-EM), a technique essential for reconstructing three-dimensional molecular structures. By leveraging two datasets with distinct orientation distributions—one uniform and the other non-uniform—MoDM demonstrates that these second-order moments can uniquely determine the underlying molecular structure, albeit with some limitations such as global rotation and reflection. The development of a convex-relaxation-based algorithm further enhances the accuracy of this reconstruction process, relying solely on second-order statistics. This research underscores the critical role of dataset diversity, showing that collecting and modeling multiple datasets under varying experimental conditions can substantially improve the quality of reconstructions in computational imaging tasks. Such advancements not only enhance our understanding of molecular structures but also pave the way fo…