Fig. 1
From: A large-scale curated and filterable dataset for cryo-EM foundation model pre-training
Overview of cryo-EM data processing pipeline. The pipeline covers several steps from movie capturing, motion correction, CTF estimation, micrograph curation, particle picking and 3D reconstruction. (a) demonstrates rigid motion, bending motion and patch-wise motion estimation during of movies. (b) shows the 2D fitting data obtained during CTF estimation of micrographs, with the upper part of the image representing reciprocal space and the lower part showing simulated Thon rings. (c) evaluates the quality of images to determine their suitability for reconstruction. (d) accurately picks all particles from the micrograph. (f) performs 2D classification and averaging of the picked particles, removing unsuitable ones. (g) estimates particle poses and performs reconstruction without prior pose information. Finally, (h) refines the reconstructed volume data to achieve high-resolution results.
