リム傑威
Lim Jackwee
Advances in cryo-EM
This is a walkthrough in using the latest advances in cryo-EM including graphene oxide grid preparation, etc... and GPU-Motioncor2.
Graphene oxide (GO) grid preparation
The video is available through figshare: http://dx.doi.org/10.6084/m9.figshare.3178669. Thomas, Andreas and Anthony have added additional washing steps to improve the quality of GO in http://dx.doi.org/10.1038/ncomms11293. In addition, we found the need to double the glow discharge time if necessary to obtain better reproducibility of the GO.
The preparation was inititated by Rado Pantelic et al. Please cite http://dx.doi.org/10.1016/j.jsb.2009.12.020.
The following steps are modified from http://dx.doi.org/10.6084/m9.figshare.3178669.
1. Graphene oxide (Sigma Aldrich #763705) diluted 10X to 0.2mg/ml (click).
2. Spin for 30 seconds at 400 g to remove aggregates.
3. Glow discharge grids with carbon side up (Quanitfoil Au 300 R1.2/1.3 1minute at 0.2mBar and 40mA; Quanitfoil Cu 300 R1.2/1.3 60 seconds at 0.2mBar and 30mA). The glow discharge time may be doubled if necessary.
4. Prepare a flat and clean working area (e.g. with parafilm).
5. Prepare 3x 20µl drops of ddH20 for each grid you want to prepare.
6. Take up grids with anti-capillary tweezers.
7. Place 3µl of the diluted GO solution (1) on the carbon side of the grid.
8. Incubate for 1 minute.
9. Remove solution with filter paper.
10. Take up first drop of ddH20 with the carbon side and remove it with the filter paper.
11. Same for the second drop.
12. Take up third drop of ddH20 with the back side of the grid.
13. Dry for 5 minutes or 1h before use.
Workflow
1. Newstack utility in IMOD
2. Whole frame alignment with Motioncor2.sh
3. CTF estimation with SCIPION: gctf or ctffind4 (note Min and Max minimas)
4. Particle picking with SCIPION: EMAN/RELION/ XMIPP or Gautomatch
5. 2D classification with SCIPION: RELION/ XMIPP
6. 3D Reconstruction with SCIPION: RELION auto-refine
7. Recenter.py and repeat particle re-extraction
8. Back to step 5 with high resolution frames of total dose controlled at 20e/Å^2
Softwares needed
Graphic card (To run GPU-programs including the coming Graphic Processing Unit (GPU)-RELION)
Cuda-library compatible to GPU (v7.5 or better is needed for -Patch for local alignment in Motioncor2) (click)
Gautomatch (http://www.mrc-lmb.cam.ac.uk/kzhang/, for particle picking)
Gctf (http://www.mrc-lmb.cam.ac.uk/kzhang/, for Contrast Transfer Function correction)
EMAN (http://blake.bcm.edu/emanwiki/EMAN2)
RELION (http://www2.mrc-lmb.cam.ac.uk/relion/index.php/Main_Page)
IMOD (http://bio3d.colorado.edu/imod/)
Motioncor2 (http://msg.ucsf.edu/em/software/index.html, preferred to unblur)
PyMol (http://pymol.org/, for structure visualization)
UCSF Chimera (https://www.cgl.ucsf.edu/chimera/, for structure visualization)
SBGrid (https://sbgrid.org/, structural biology programs platform)
SCIPION / CCP-EM (Integrated cryo-EM softwares platforms)
After collecting the multipage-tif stack from cryoEM Titan...
1. To convert a multipage-tif stack containing all movie-frames to mrc format by IMOD
tif2mrc FrameStack.tif FrameStack.mrc
2. To unpack 4-bits stack.mrc from data acquisition by IMOD
clip unpack FrameStack.mrc UnpackedFrameStack.mrc
3. To process a packed.mrc (multiply by defects.txt from SerialEM) by IMOD to super-resolution mode image (click)
clip unpack -D defects.txt FrameStack.mrc UnpackedFrameStack.mrc
4. To create substack from the mrc stack by IMOD
newstack -fileinlist -fromone file FrameSubStack.st
Name of file with list of input files and sections to read (from Relion star file).
The format of the file is:
number of input files
name of first file to read
list of sections to read from first file
name of second file to read
list of sections to read from second file
...
5. Batch processing of (1) and (3) collected by SerialEM and gain reference image, SuperRef_rot90.mrc. The output is _gN.mrc
#!/bin/sh -x
#batch IMOD
cd /emdata4/data/frames_in_tif/
for FILE in myprotein_???.tif; do
mrc_file=`echo $FILE | sed 's/tif/mrc/'`
#mrc_stack_file=`echo $FILE | sed 's/tif/mrcs/'`
gainNorm_file=`echo $FILE | sed 's/\.tif/_gN.mrc/'`
tif2mrc $FILE $mrc_file
clip unpack -D ../test_unzip/defects.txt $mrc_file ../test_unzip/SuperRef_rot90.mrc $gainNorm_file
done
To create an inital volume reference 1ryp_8A.mrc at 8Å from PDB 1ryp.pdb by EMAN
e2pdb2mrc.py --apix 1.35 --res 8 --box 320,320,320 1ryp.pdb1 1ryp_8A.mrc
To query image information about 1ryp_8A.mrc by EMAN
e2iminfo.py -H 1ryp_8A.mrc
To rotate 1ryp_8A.mrc to to a preferred orientation for initial 3D volume reconstruction by EMAN
e2proc3d.py --rot=0,90,0 1ryp_8A.mrc 1ryp_8A_90.mrc
Checklist for cryo-EM data collection