Computational protocol: Structural basis for energy transduction by respiratory alternative complex III

Similar protocols

Protocol publication

[…] ACIII sample at 0.025 mg ml−1 was negatively stained with 1% (w/v) uranyl acetate, pH ~4. Electron micrographs were recorded on a CCD camera (Gatan Ultrascan 4000) with a Tecnai Spirit at 120 kV under low-dose conditions, at a magnification of ×51,000 corresponding to a 2.34 Å pixel size at the specimen. Approximately 2000 particle images were picked manually in EMAN boxer and used to generate templates. In total, 25,848 particle images were autopicked and used for two-dimensional reference-free classification in RELION 1.4. The 2D class averages with recognizable features were selected and used to generate a low-resolution initial model with EMAN2. [...] A set of 2479 movies was collected. Whole-image drift correction and dose weighting of each movie were performed using MotionCor2. Particles were picked manually using EMAN Boxer or automatically by template matching in Gautomatch (by Kai Zhang, MRC-LMB Cambridge, UK), and the micrograph-based CTF was determined using CTFFIND4 on drift-corrected, non-dose-weighted images. Automatically picked particles were subjected to a first round of reference-free two-dimensional classification with ISAC within Sphire, to exclude false positives. The initial clean data set contained 131,995 particle images (288 pixels × 288 pixels). Dose-weighted particles were subjected to 2D classification in RELION 1.4. Visual selection of particle classes with interpretable features resulted in a data set of 103,756 particle images for 3D classification. The initial ACIII map generated with EMAN2 from negative-stained specimens was low-pass filtered to 60 Å and used as an initial reference for the 3D classification in RELION 1.4. The best 3D classes were selected for 3D refinement in RELION 2.0. Individual frames were B-factor weighted and movements of individual particles were reversed by movie frame correction in RELION 2.0. The resulting data set of polished particles was used for a new 3D refinement, producing a final map at 3.87 Å resolution. Particles from 3D classes with a weak density at the membrane level were re-extracted with a larger box (300 pixels × 300 pixels) and extensively classified in 3D. The refinement of the final pool of particles with the extra density resulted in a 20 Å map. UCSF Chimera was used for visualizing cryo-EM maps and docking of atomic models to the 20 Å map (ACIII from this work and T. thermophilus caa3 oxygen reductase—PDB 2YEV). Model building was performed with Coot and real space refinement was performed in Phenix followed by manual rebuilding. MolProbity and EMRinger were used for model validation. Local resolution was estimated using ResMap. Figures were drawn with UCSF Chimera. […]

Pipeline specifications

Software tools EMAN, RELION, MotionCor2, Gautomatch, CTFFIND, SPHIRE, UCSF Chimera, Coot, PHENIX, MolProbity, EMRinger, ResMap
Applications cryo-EM, Protein structure analysis
Organisms Homo sapiens, Rhodothermus marinus