Automatic Particle Picking from Cryo-Electron Microscopy Images by Using Partial Least Squares

Single-particle analysis is a three-dimensional (3D) structure determining technique using electron microscopy. In this technique, a large number of projections is required to create 3D reconstruction. Manual selection of particles, by visual observation of the micrographs on a computer screen, is recognized as a bottleneck in automated single-particle reconstruction. In this paper, we established a brand-new method which is based on partial least squares theory. After initial selection for candidate particles, a PLS recognition model is set up to identify the needed particles from the candidates.