Quantifying Distributions of Cryo-EM Projections

In cryo-electron microscopy, a set of two-dimensional projections collected from different viewing directions may complicate image processing and subsequent model building if the distribution of these views is non-uniform. View distributions are traditionally represented as color-coded two-dimensional diagrams. However, such diagrams can introduce distortions, are cumbersome to manipulate, store, and compare across datasets; they do not provide a commonly accepted quantitative measure of uniformity. In this work, we propose a method to characterize these angular distributions quantitatively and to represent them as simple one-dimensional curves rather than two-dimensional colored diagrams. The suggested measures could be incorporated into databases such as the EMDB to provide a compact, standardized description of the angular distribution of particle views. Synopsis We propose a numerical measure to quantify the deviation of a cryo-EM 2D view distribution from the uniform one. This approach allows the traditional two-dimensional projection diagrams to be replaced, or completed with simple, easily generated curves that are straightforward to analyze.