Scripts to make cryo-electron tomography a bit easier
Full options can be listed via:
tomotools --help
tomotools [subcommand] --help
- blend-montages: Blends SerialEM montages, write results to separate folder.
- Example:
tomotools blend-montages MMM*.mrc montages-blended
- Example:
- preprocess: Prepare tiltseries for reconstruction.
- Takes data directory from SerialEM or Tomo5 as input and writes the final stacks to the target directory.
- Frames are aligned using MotionCor2 and reordered if desired. Supports GainRef conversion from dm4 to mrc and the SerialEM-generated defects.txt.
- Example:
tomotools preprocess --mcbin 1 --gainref frames/GainRef.dm4 *.mrc ts-aligned
- reconstruct: Perform batch reconstruction using AreTomo or imod.
- Takes tiltseries and their associated mdoc files as input, automatically identified associated EVN/ODD stacks. Finds alignment using AreTomo, then applies it to EVN/ODD stacks. Alternatively, can move files and then open
etomo. Reconstruction is done using imod'stilt. - Example:
tomotools reconstruct --move --bin 4 --sirt 12 --do-evn-odd *.mrc
- Takes tiltseries and their associated mdoc files as input, automatically identified associated EVN/ODD stacks. Finds alignment using AreTomo, then applies it to EVN/ODD stacks. Alternatively, can move files and then open
- cryocare-extract: Extract cryoCARE training data from EVN/ODD reconstructions.
- cryocare-train: Wrapper for cryoCARE training.
- cryocare-predict: Wrapper for cryoCARE prediction.
- deconv: Python implementation of Dimitry Tegunov's tom_deconv.m script.
- imod2warp: Takes a list of imod directories or a file listing them and prepares everything for Warp or WarpTools.
- aretomo2warp: Takes a list of directories with AreTomo-aligned tiltseries or a file listing them and prepares everything for Warp or WarpTools.
- imod2tomotwin: Takes a list of imod directories and reconstructs for TomoTwin picking.
- aretomo2tomotwin: Takes a list of AreTomo directories and reconstructs for TomoTwin picking.
- fit-ctf: Run imod ctfplotter on a set of tiltseries and save results to their folder.
- merge-dboxes: Very beta, merges Dynamo DBoxes.
- reconstruct-3dctf: Perform reconstruction using imods
ctf3dfunction.
- semnavigator: Display SerialEM navigator files to find back you tomogram positions
- create-movie: Create a movie from a series of image files.
- update: Automatically pulls the most recent version from GitHub and runs
pip install --upgradeon it. - restore-frames: Restore SubFramePath to mdoc of tiltseries preprocessed with tomotools < 0.4.
tomotools depends on commands from MotionCor2 or MotionCor3, IMOD, and AreTomo 1.X or AreTomo2 for full functionality. IMOD should be in PATH.
MotionCor2/3 and AreTomo2/3 can either be in PATH as MotionCor2 / MotionCor3 or AreTomo / AreTomo2 respectively, or set using the envar MOTIONCOR_EXECUTABLE or ARETOMO_EXECUTABLE.
We suggest installing tomotools into its own conda / mamba environment. If you can only access your userspace, consider using micromamba.
If you want to use the scripts interfacing with cryoCARE denoising, please install tomotools[cryocare].
conda create -n tomotools python=3.8 cudatoolkit=11.0 cudnn=8.0 -c conda-forge
conda activate tomotools
pip install "tomotools[cryocare] @ git+https://github.com/tomotools/tomotools.git"
conda create -n tomotools python=3.8 -c conda-forge
conda activate tomotools
pip install "tomotools @ git+https://github.com/tomotools/tomotools.git"
With tomotools installed into a conda environment, you can then start tomotools with:
conda activate tomotools
tomotools --help
If you're using an sbgrid environment, make sure to set the following in your .sbgrid.conf file:
PYTHON_X=3.8.8 (anything > 3.8 works)
Additionally, try pip --version to make sure it's correctly working. This is required for tomotools update. Else, you can add the following line to your .bashrc:
alias pip='python -m pip'
Likely, you will also need to add your local Python path to .bashrc:
export PATH="/YOURHOMEFOLDER/.local/bin:$PATH"