Three Dimensional Reconstruction of Single Particle Specimens using Reference Projections Without Defocus Groups

Introduction

This document contains a protocol for creating a 3D reconstruction utilizing 'gold standard resolution workflow' from electron micrographs without sorting the images into defocus groups. This method is an alternative to the legacy SPIDER method of creating a 3D reconstruction using defocus grouping. In both methods, once a set of particle images has been obtained, an initial 3D reconstruction is calculated using coarse projection angles. This is followed by refinement, which iteratively adjusts the angles for finer resolution.

In the traditional method, ensembles of particles with similar defocus values are grouped together. A separate 3D reconstruction is computed for each "defocus group." CTF-correction is performed when these reconstructions are merged into a final reconstruction.

In this alternative method, the CTF-correction is applied at the level of windowed particle images. This method presents certain advantages. First, it circumvents one of the approximations when using defocus groups, namely that all particles in a defocus group follow the same CTF profile. At high resolution, where the CTF oscillates more rapidly, this assumption may not hold. Second, parallelization can be more efficient, since groups can be of identical size, independent of the number of particles at each defocus. Third, particles from what would be sparsely populated defocus groups need not be thrown out. In principle, though not yet implemented, astigmatic images can be corrected for with this method, as can tilted-specimen images. Lastly, interoperability with other software packages, which also correct for the CTF at the level of particle images, will be more straightforward.

However the strategy of using defocus groups may have some advantages. First is that it can readily account for the non-uniform distribution of signal-to-noise in projection data (Penczek, 2012). Second, we find that reconstructions using particle-level CTF-correction sometimes show artifacts when using iterative backprojection methods, such as 'BP RP' or 'BP CG', whereas the use of defocus groups does not present such limitations.


Contents


Links to further information


Running SPIDER procedure files

In the sections below, various file types are denoted by different fonts:

The data below are sometimes shown plotted with gnuplot; see the Gnuplot manual for details.

After each step check the output files to make sure the results are sensible. If you are not sure what is "sensible," ask an expert.


Quick-start guide

This is a brief listing of steps in this protocol. For simplicity, options have been limited. For further information, see below. A procedure for testing a complete reconstruction using the Nature Protocols paper data set is provided.

Data extension is assumed to be: dat

In your working directory (e.g.user):

  1. cp $SPIDER_DIR/docs/techs/recon1/spiproject.tar.gz .
  2. tar -xvf spiproject.tar.gz
  3. cd myproject

In your project directory (myproject):

  1. cp   `which spider`   ./spider
  2. Data loading choices:
    1. mv   /actual-location-of-your/micrographs/raw*    Micrographs/
      spider spi/dat @make-params
      spider spi/dat @resizevol
    2. tar -xvf $SPIDER_DIR/docs/techs/recon1/natproc_data_mics.tar.gz

In Micrographs/:

  1. spider spi/dat @make-mic-list
  2. spider spi/dat @shrink-mic
  3. montagefromdoc   ../sel_micrograph.dat   sm_mic_* &

In Power_Spectra/:

  1. spider spi/dat @ctffind
  2. montagefromdoc   ../sel_micrograph.dat   power/pw_avg*
  3. ctfmatch.py   power/ctf* &
  4. spider spi/dat @make-ctf-cor

In Particles/:

  1. spider spi/dat @make-noise-img
  2. Particle picking choices
    1. spider spi/dat @pick-at
    2. spider spi/dat @pick-lfc
  3. montagefromdoc   win/sel_part_0001.dat win/winser_0001.dat &
  4. [Optional] Initial verification
    1. spider spi/dat @denoise-imgs
    2. Use Montage operation in WEB.
  5. spider spi/dat @number-parts

In Reconstruction/:

  1. spider spi/dat @restack-n-ctf
  2. spider spi/dat @make-ref-views
  3. Alignment choices
    1. ./spider spi/dat @align
    2. ./spider spi/dat @pub-align

In Averages/:

  1. spider spi/dat @select-byview
  2. spider spi/dat @avg-filter-byview
  3. spider spi/dat @plot-cc-vs-view
  4. spider spi/dat @show-ref-views
  5. [Optional] Classification-based verification
    1. spider spi/dat @verify-class-byview
    2. verifybyview.py views/prj001
    3. spider spi/dat @verify-combine-classes
    4. spider spi/dat @plot-cc-vs-view
    5. spider spi/dat @show-ref-views
    6. spider spi/dat @plot-ref-views
    7. spider spi/dat @parts-vs-defocus
  6. spider spi/dat @best-imgs

In Reconstruction/:

  1. spider spi/dat @recon-regroup
  2. Reconstruction choices
    1. ./spider spi/dat @recon
    2. ./spider spi/dat @pub-recon
  3. spider spi/dat @plot-fsc-curve

In Refinement/:

  1. Refinement choices
    1. ./spider spi/dat @refine
    2. ./spider spi/dat @pub-refine
  2. spider spi/dat @plot-fsc-curves

Creating a new reconstruction project

At the start of a reconstruction project, a project directory has to be set up with the proper subdirectories and procedure files.

Run these procedures in the project/ directory.



Preparing micrographs

Place digitized micrographs in the Micrographs/ directory.

Run these procedures in the Micrographs/ directory.

  1. If you have successive frames from a K2 Direct Electron Capture Camera in MRC format align the frames using 'framealign.spi' which implements methodology from: X Li, P Mooney, S Zheng, C R Booth, M B Braunfeld, S Gubbens, D A Agard & Y Cheng.
    This is a global alignment over the whole image area and does not consider any movement of individual particles within the ice.

    It reads filenames and ###.frames.mrc (Where ### denotes stem name of input set). It creates:
    ¤ ###: SPIDER micrograph stacks.
    ¤ filtstk_###: Filtered micrograph stacks.
    ¤ shift_doc_###: Raw alignment shift doc file.
    ¤ mlr_shift_doc_###: Refined alignment shift doc file.
    ¤ shstk_###: Aligned frame image stacks.
    ¤ ali_###: Summed aligned image.

  2. If you are using SPIDER, Hiscan TIFF, PerkinElmer, ZI, or Nikon Coolscan micrograph data:
    mv /actual-location-of-micrographs/*###*   Micrographs/

    The procedures below can accept SPIDER, Hiscan TIFF, PerkinElmer, ZI, and Nikon Coolscan data. These files may have been compressed with gzip, if so be sure to indicate compression in the params file.

  3. If using the test data provided in the SPIDER distribution from the Nature Protocols paper load the data set and unpack it in your project directory:
    tar -xvf $SPIDER_DIR/docs/techs/recon1/natproc_data_mics.tar.gz
    This creates:
    ¤ sel_micrograph: Micrograph selection doc file listing 4 micrographs used later in the protocol.
    ¤ Micrographs/raw****: Four micrograph images.
    ¤ orig_reference_volume.dat: Reference volume.
    ¤ reference_volume.dat: Reference volume.
    ¤ params: Reconstruction parameter file.
    If you are using this data you can skip the next two steps.


Selecting micrographs

Generate a list of micrographs and screen the micrographs.

Run these procedures in the Micrographs/ directory.



Contrast Transfer Function Parameter Estimation

Estimate defocus of each micrograph by calculating its power spectrum.

For more information, see:
  1. Contrast transfer function correction tutorial.
  2. ctfdemo, a program that demonstrates effects of various parameters on the CTF.
  3. Penczek et al., 1997 for a fuller discussion.

Run these procedures in the Power_Spectra/ directory.


Particle Windowing and Initial Verification

A particle picking procedure file analyzes each micrograph, cutting out small windows of likely particle candidates. This is followed by a manual selection process that identifies the good particle images and rejects the bad ones. Particles automatically output by the procedure file are said to be windowed; the subset that are manually chosen are said to be verified.

Run these procedures in the Particles/ directory.


Alignment

Reference images are generated from the reference volume. Data particles are compared to each reference to find the best match, and the corresponding transformations (shifts and rotations) are written to a doc file. Finally, the transformations are applied to the data images, aligning them to the references.

Run these procedures in the Reconstruction/ directory.



Compute Averages and Verify Particles

For all projections, all aligned particles of a given reference view are averaged together. Further particle selection is made by selecting a correlation cutoff threshold to reject some particles. The distribution of particles among projections can be displayed.

Run these procedures in the Averages/ directory.

Note: *** denotes image group, ### denotes view, ??? denotes class


3D Reconstruction

Use the selected aligned particles to create an initial 3D volume. To estimate the nominal resolution of the resulting structure, the particle images are split into two equal sets, and the two resulting reconstruction volumes are compared.

Run these procedures in the Reconstruction/ directory.