MPicker

Welcome to use MPicker. It's a tool for membrane flatten and visualization.

See our website thuem.net for details.

• if no Conda

  • We recommend using Conda, because our offline environment package is very large and Conda is very commonly used in python. You can install conda from https://docs.conda.io/projects/miniconda/en/latest/
  • We generate the environment packages by conda-pack on CentOS7. It should work for other Linux system. You have to use conda if your system is not Linux.
  • Download environment.tar.gz and code.tar.gz
  • Uncompress them into 2 different empty folders. Here we call them /absolute/path_to/save_code and /absolute/path_to/save_env, replace them to real path you want to install mpicker at.

    mkdir /absolute/path_to/save_code
    tar -zxvf code.tar.gz -C /absolute/path_to/save_code
    mkdir /absolute/path_to/save_env
    tar -zxvf environment.tar.gz -C /absolute/path_to/save_env
    

  • Unpack the conda environment (just need once)

    source path_to/save_env/bin/activate
    conda-unpack
    source path_to/save_env/bin/deactivate
    

  • After unpack, you can activate the environment by

    source path_to/save_env/bin/activate
    

    and close the environment by

    source path_to/save_env/bin/deactivate
    

• Conda

  • Download code.tar.gz ( environment.tar.gz is not needed)
  • Uncompress it into an empty folder. Here we call it /absolute/path_to/save_code, replace it to real path you want to install mpicker at.

    mkdir /absolute/path_to/save_code
    tar -zxvf code.tar.gz -C /absolute/path_to/save_code
    

  • You can edit the file env.yml in folder path_to/save_code/mpicker_gui as you like
    • you can change "mpicker_full" to the environment name you like
    • you can change "cudatoolkit=10.1" to other version if your gpu driver doesn't support it (=11.0 or =9.2 for example)
  • Then you can create the conda environment from this file (may take 10min+)

    conda env create -f env.yml
    

  • Finally, you can activate the environment by

    conda activate mpicker_full
    

• No auto membrane segmentation by AI, but smaller installation size
• You can replace code.tar.gz by code_noseg.tar.gz , which without folder memseg_v3
• Just replace environment.tar.gz by environment_noseg.tar.gz (if no Conda)
• Just replace the file env.yml by env_simple.yml (Conda)

• Simple Start

  • Activate the environment at first, and you can see the result of

    which python
    

    should be something like

    path_to/save_env/bin/python # if no conda
    path_to/miniconda/envs/mpicker_full/bin/python # conda
    

  • Run this command to open Mpicker

    python path_to/save_code/mpicker_gui/Mpicker_gui.py
    

    You can add path_to/save_code/mpicker_gui into your PATH and just open Mpicker by

    Mpicker_gui.py
    

• Troubleshooting

  • For Linux, if your shell is not bash, for example csh, switching to bash may solve the problem about environment setup. You can check you shell by

    echo $0
    

  • If you fail to execute .py file directly you can try to run bin/mpicker.sh, which is a bash wrapper
  • You can add path_to/save_code/bin into PATH and activate the environment, then you can run Mpicker by

    mpicker.sh
    

  • You can replace PYTHON="$(which python)" in bin/mpicker.sh to PYTHON=/absolute/path_of/python , if your system just fail to find the right python we installed. By this, you can run Mpicker by mpicker.sh without activate environment at first

• Command Start

  • Mpicker support these parameters input from command line

    --raw Path to raw tomogram map

    --mask path to mask tomogram map

    --out Path to save all the result files

    --config Path to config file (config file for reloading all the process history)

  • Example: (Open a new GUI)

    Mpicker_gui.py --raw grid3_4_1_corrected.mrc \
    --mask segmentation_grid3_4_1_corrected.mrc \
    --out ../result
    

  • Example: (Reload a existing GUI)

    Mpicker_gui.py --config ../grid3_4_1_corrected.config
    

  • Then you will get the following GUI:

• For More Detailed instructions, Please see the Tutorial.md and Manual.md