Frame alignment and dose weighting were performed with MotionCor2 (using Fourier space cropping by a factor of 2, resulting in a pixel size of 1.35 Å) (41), and contrast transfer function was determined with CTFIND4 (42). Initially, 8563 (apoElp123), 10888 (Elp123-tRNA lobe), and 9757 (Elp123-2tRNAs) particles were manually selected using EMAN2 boxer swarm tool (43). 2D classes obtained from these subsets in RELION2 (44) were used as templates in the autopicking procedure with the full datasets. Classification, refinement, post-processing, and local resolution estimation were performed with RELION2.

For the Elp123 structure, 1 million autopicked particles were extracted and binned four times to reduce computational costs (fig. S1G). These particles were subjected to reference-free classification, from which 490,000 particles (49%) were selected for subsequent processing. 3D classification was performed with the selected unbinned particles using as a starting model the available Elp123 map at a resolution of 27 Å (EMD-4151) low pass–filtered to 60 Å. Three classes (containing 32% of the particles) were selected on the basis of their recognizable Elp123-like shape and subjected to a second round of 3D classification, using either a mask that covered one lobe of the Elp123 subcomplex (for the Elp123 lobe map) or a mask covering the whole Elp123 subcomplex [for the Elp123 lobe DD map]. For the Elp123 lobe, the classification with the lobe mask yielded two classes that showed better resolution and gathered 54% of the particles, which were pooled and refined using a lobe mask. We obtained the Elp123 lobe map with a resolution of 3.4 Å and 84,135 particles, which after post-processing yielded the final 3.3 Å resolution map. For the Elp123 lobe DD, the classification using a full mask yielded two classes with both lobes nicely defined, so we pooled the 81,709 particles and then refined them using a mask covering the whole molecule that yielded the Elp123 lobe DD map at a resolution of 3.97 Å. The post-processing of this map yielded a final 3.7 Å resolution map that included well-resolved Elp1-DD. Alternatively, for the Elp123 full map, after the first round of 3D classification, we selected the best resolution class that showed an apparent twofold symmetry axis, although no symmetry was previously imposed. This class (with 45,236 particles) was refined with C2 symmetry, applying a mask that covered the whole molecule. This procedure yielded the Elp123 full map at a resolution of 7.5 Å, which was post-processed, yielding the final map at a resolution of 4.5 Å.

For the Elp123-tRNA lobe reconstruction (fig. S4F), 1.5 million particles were autopicked, extracted with bin 4, and reference-free–classified. A total of 751,000 particles (50%) were selected and subjected to 3D classification using the same starting model as for apoElp123, the available 27 Å map of Elp123 (EMD-4151) low pass–filtered to 60 Å as a reference. Two classes (that gathered 194,000 particles) were selected on the basis of their Elp123-like shape and the presence of an extra density on one or two lobes of the molecule.

The unbinned particles were used for a second round of 3D classification, yielding a good class of 65,000 particles with a conspicuous extra density on one lobe. A subsequent focused 3D classification step was performed after partial signal subtraction of the lobe (45). Only the signal coming from the extra density was considered, and masked 3D classification without alignment was performed yielding one class with a tRNA-shaped density in the lobe. This class with 23,020 particles was refined using a lobe mask, yielding an Elp123-tRNA lobe map at a resolution of 4.5 Å. The final map after post-processing was obtained at a resolution of 4.4 Å.

For the Elp123-2tRNA complex (fig. S6C), 798,000 autopicked particles were extracted, binned four times, and subjected to reference-free classification, from which 654,000 particles (82%) were selected. 3D classification was performed with the subset of binned particles using the Elp123 map (EMD-4151) low pass–filtered to 60 Å. Two classes were selected on the basis of their Elp123-like shape and the presence of an extra density on one or two lobes of the molecule. The unbinned 189,000 particles from the two classes were subjected to a second round of 3D classification. One class with 52,000 particles was selected on the basis of the presence of a strong extra density on each lobe. This subset of particles was subjected to partial signal subtraction of the two lobes. Then, masked 3D classification without alignment was performed, yielding two classes with extra densities in both lobes. The 12,921 particles from these classes were pooled and refined independently with and without C2 symmetry, using a mask covering the whole molecule and yielding two equivalent reconstructions at resolutions of 7.8 Å (without symmetry) and 6.9 Å (with C2 symmetry) (fig. S6A). The Elp123-2tRNA reconstructions were post-processed, yielding the final 7.6 Å resolution map (without symmetry) and the 6.7 Å resolution map (with C2 symmetry).

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.