Tomography tutorial
Contents
Tomography tutorial¶
This tutorial shows how to process tilt-series from the HIV-1 Gag (EMPIAR-10164) dataset.
We first download and decompress a tbz file containing a subset of 5 tilt-series (down-sampled 2x compared to the original data), and an initial model:
# cd to a location in the shared file system and run:
wget https://nextpyp.app/files/data/nextpyp_tomo_tutorial.tbz
tar xvfz nextpyp_tomo_tutorial.tbz
Step 1: Create a new project¶
Next, we create an empty folder where all files for the tutorial will be saved:
mkdir EMPIAR-10164
cd EMPIAR-10164
Step 2: Pre-processing¶
Data pre-processing consists of movie frame alignment, tilt-series alignment, tomogram reconstruction, CTF estimation and virion detection:
# launch pre-processing
pyp -data_mode tomo \
-scope_voltage 300 \
-scope_pixel 1.35 \
-data_path="/path_to_frames/*.tif" \
-ctf_max_res 5.0 \
-scope_tilt_axis 85.3 \
-tomo_rec_binning 8 \
-tomo_rec_thickness 2048 \
-tomo_vir_method "auto" \
-tomo_vir_rad 500.0 \
-slurm_tasks 23 \
-movie_pattern "TILTSERIES_SCANORD_ANGLE.tif"
Step 3 (optional): Virion segmentation¶
In this step we use IMOD over a remote X11 connection to interactively select virion segmentation thresholds. Execute the command below and select the column in the image where the yellow curve more closely matches the membrane. Go over all virions in the tilt-series and save the model when you are done:
( export tomoedit=tomoedit &&
singularity exec -B /your_binds --no-home -B $HOME/.ssh /path_to_container/pyp.sif
/opt/pyp/bin/run/pyp -vir -skip )
Tip
To skip a virion, simply select the rightmost column (these virions will be removed from the downstream processing).
Step 4: Particle detection¶
Detect spikes using the membrane values selected above:
pyp -data_mode tomo \
-tomo_vir_detect_method "mesh" \
-tomo_vir_detect_dist 8 \
-tomo_vir_detect_band 800 \
-tomo_spk_rad 50.0
Step 5: Reference-based refinement¶
If a 3D reference is available, the csp command can be used to align the particle projections using constrained refinement:
# launch coarse refinement
csp -refine_parfile_tomo=`pwd`/frealign/EMPIAR-10164_original_volumes.txt \
-refine_model="EMPIAR-10164_init_ref.mrc" \
-particle_mw 300.0 \
-particle_rad 150.0 \
-particle_sym "C6" \
-extract_box 192 \
-extract_bin 2 \
-extract_fmt frealign \
-refine_skip \
-refine_fboost \
-refine_maxiter 2 \
-refine_rhref "8.0" \
-csp_UseImagesForRefinementMax 10 \
-csp_refine_particles \
-csp_NumberOfRandomIterations 50000 \
-csp_ToleranceParticlesShifts 50.0 \
-csp_ToleranceParticlesPhi 10.0 \
-csp_ToleranceParticlesTheta 10.0 \
-reconstruct_mintilt -50 \
-reconstruct_maxtilt 50
Tip
To only search for in-plane rotations (i.e., rotation angle Psi), set the tolerance of the other two rotations
csp_ToleranceParticlesPhiandcsp_ToleranceParticlesThetato zero.cspcan also use initial alignments from other software packages such as Relion or EMAN. For example, see Tomo import/export to import alignments from Relion.
Step 6: Fully constrained refinement¶
New, we do additional local refinement:
# launch coarse refinement
csp -refine_maxiter 5 \
-refine_rhref "8:10:8:6" \
-csp_OptimizerStepLength 100.0 \
-csp_NumberOfRandomIterations 0 \
-csp_ToleranceParticlesShifts 20 \
-csp_ToleranceParticlesPhi 20.0 \
-csp_ToleranceParticlesPsi 20.0 \
-csp_ToleranceParticlesTheta 20.0 \
-csp_refine_micrographs \
-dose_weighting_enable \
-dose_weighting_fraction 4
All results from 3D refinement are saved in the folder frealign/maps, including png files for visual inspection corresponding to each refinement iteration.
Tip
Tolerance parameters determine the range used for searching. If you think particle alignments or tilt-series alignments are not accurate, you can increase the corresponding tolerances.
Step 7: Filter particles¶
The next step is to remove particles with low correlation scores:
mv frealign/maps frealign/fully_constrained && mkdir frealign/maps
pcl -clean_parfile=`pwd`/frealign/fully_constrained/EMPIAR-10164_r01_05.par.bz2 \
-clean_threshold 2.5 \
-clean_dist 10.0 \
-clean_mintilt -15.0 \
-clean_maxtilt 15.0 \
-clean_min_num_projections 1 \
-clean_check_reconstruction
Step 8 (optional): Permanently remove bad particles¶
It is often a good idea to permanently remove any bad particles identified in the previous step:
pcl -clean_discard
Step 9: Region-based refinement before masking¶
The following command performs region-based constrained alignment:
mv frealign/maps frealign/filter_particles && mkdir frealign/maps
csp -refine_parfile=`pwd`/frealign/filter_particles/EMPIAR-10164_r01_02_clean.par.bz2 \
-refine_model=`pwd`/frealign/filter_particles/EMPIAR-10164_r01_02.mrc" \
-particle_rad 100.0 \
-extract_box 384 \
-extract_bin 1 \
-refine_iter 2 \
-refine_maxiter 3 \
-refine_rhref "6:5" \
-csp_UseImagesForRefinementMax 4 \
-csp_refine_particles \
-csp_refine_micrographs \
-csp_ToleranceParticlesShifts 20.0 \
-csp_Grid "8,8,2"
Step 10: Create shape mask¶
The next step is to create a shape mask:
mv frealign/maps frealign/region_refine && mkdir frealign/maps
pmk -mask_model=`pwd`/frealign/region_refine/EMPIAR-10164_r01_03.mrc \
-mask_threshold 0.42 \
-mask_normalized \
-mask_edge_width 8
Step 11: Region-based refinement after masking¶
Next, we do further refinement using the mask calculated in the previous step:
mv frealign/maps frealign/mask && mv frealign/region_refine frealign/maps
csp -refine_maxiter 6 \
-refine_rhref "6:5:5:4:3.5" \
-refine_maskth=`pwd`/frealign/mask/mask.mrc"
Step 12: Particle-based CTF refinement¶
In this step we refine the CTF parameters on a per-particle basis:
csp -refine_maxiter 7 \
-refine_rhref "3.1" \
-no-csp_refine_micrographs \
-no-csp_refine_particles \
-csp_refine_ctf \
-csp_UseImagesForRefinementMax 10
Step 13: Movie frame refinement¶
Next, we refine the raw movie frames against the most recent 3D reconstruction:
mv frealign/maps frealign/ctf_refine && mkdir frealign/maps
csp -refine_parfile=`pwd`/frealign/ctf_refine/EMPIAR-10164_r01_07.par.bz2 \
-refine_model=`pwd`/frealign/ctf_refine/EMPIAR-10164_r01_07.mrc \
-particle_rad 80.0 \
-extract_fmt frealign_local \
-refine_iter 2 \
-refine_maxiter 2 \
-refine_rhref "3.2" \
-refine_spatial_sigma 200.0 \
-refine_transreg \
-no-csp_refine_ctf \
-csp_frame_refinement \
-csp_UseImagesForRefinementMax 4
Step 14: Refinement after movie frame refinement¶
Using the refined frame averages for each tilt, we perform additional constrained refinement:
csp -refine_maxiter 3 \
-refine_rhref "3.3" \
-csp_refine_micrographs \
-csp_refine_particles \
-no-csp_frame_refinement \
-csp_ToleranceMicrographShifts 10.0 \
-csp_ToleranceMicrographTiltAngles 1.0 \
-csp_ToleranceMicrographTiltAxisAngles 1.0 \
-csp_ToleranceParticlesPsi 1.0 \
-csp_ToleranceParticlesPhi 1.0 \
-csp_ToleranceParticlesTheta 1.0 \
-csp_ToleranceParticlesShifts 10.0 \
-csp_RefineProjectionCutoff 2
Step 15: Map sharpening¶
The final step is to sharpen the map and produce FSC plots:
psp -sharpen_input_map=`pwd`/frealign/maps/EMPIAR-10164_r01_half1.mrc \
-sharpen_automask_threshold 0.35 \
-sharpen_adhoc_bfac -50