Steps for Linux System Only

1. Install Conda. Instructions below:

   • Conda is used to build an isolated environment for specific programs to avoid software conflicts.

     https://docs.conda.io/projects/conda/en/latest/user-guide/install/linux.html

2. Install TomoNet. The download link can be found on GitHub.

   git clone <https://github.com/logicvay2010/TomoNet>
   
   cd TomoNet/bin
   chmod a+x *py
   

3. Edit your .bashrc

   ############### tomoStudio related setting ##################
   export PATH=YourPath2tomoNetParentFolder/TomoNet/bin:$PATH
   export PYTHONPATH=YourPath2tomoNetParentFolder/:$PYTHONPATH
   ############### End tomoStudio setting ##################
   

4. Create a new environment by typing/copying the following commands in the terminal one by one:

   #Please make sure you have your conda and GPU driver (nvidia-smi) installed correctly 
   conda create -n tomonet python==3.10 pytorch pytorch-cuda pyqt=5 -c pytorch -c nvidia
   
   ##### Important!!! check if pytorch gpu version is installed. 
   # run python -c "import torch; print(torch.cuda.is_available())"
   # 1. you should see it return "True", otherwise means the pytorch is not correctly installed
   # and you have to troubleshoot it before next step
   
   # 2. if you see error like undefined symbol: iJIT_NotifyEvent. need to reinstall the conda env
   # conda create -n tomonet python==3.10 pytorch pytorch-cuda pyqt=5 mkl==2024.0 -c pytorch -c nvidia -c conda-forge
   
   pip install numpy pytorch_lightning mrcfile scipy fire tqdm scikit-image matplotlib imodmodel psutil eulerangles starfile
   

5. Create a new environment by typing/copying the following commands in the terminal one by one (OLD version 1):

   ### First, please check that your pip version (24.x is having issue installing pytorch_lightning)
   ### if pip in version >= 24.x.x or some too old version, 
   ### please change your pip version to 23.1.2 by "pip install pip==23.1.2"
   
   conda create -n tomonet python==3.8 (new python(3.10) version is problematic)
   conda activate tomonet 
   # Please use the correct cuda version
   conda install pytorch==2.0.1 pytorch-cuda==11.8 -c pytorch -c nvidia
   
   ~~~~conda install cudatoolkit==11.8.0 cudnn==8.9.2.26 -y
   
   conda install pyqt
   #need pip version < 24.1
   pip install numpy==1.19 pytorch_lightning==1.6.4 mrcfile scipy fire tqdm scikit-image matplotlib imodmodel psutil eulerangles starfile
   
   

6. An alternative conda env installation, please check which one works for you. (OLD version 2)

   ### First, please check that your pip version (24.x is having issue installing pytorch_lightning)
   ### if pip in version >= 24.x.x or some too old version, 
   ### please change your pip version to 23.1.2 by "pip install pip==23.1.2"
   
   # Please use the correct cuda version, 11.8 or 12.1 or others
   conda create -n tomonet pytorch=2.0.1 pytorch-cuda=12.1 -c pytorch -c nvidia -y (new python(3.10) version is problematic)
   conda activate tomonet 
   
   conda install pyqt
   #need pip version < 24.1
   pip install numpy==1.19 pytorch_lightning==1.6.4 mrcfile scipy fire tqdm scikit-image matplotlib imodmodel psutil eulerangles starfile
   
   

7. check installation

   tomoNet.py &
   

Run the above line to launch TomoNet GUI

The following software needs to be installed for some of TomoNet’s modules.

• MotionCorr2 for Motion correction
• CTFFIND4 for CTF estimation
• Imod for basic operations on MRC files, visualization, and manual picking (REQUIRED)
• AreTomo for fiducial less Automatic tomography reconstruction
• PEET 1.15 for particle picking (REQUIRED for lattice particle picking)
• IsoNet for tomogram denoising and missing-wedge compensation
• ChimeraX for sub-tomogram place back

You do not have to install all of them if you do not need to use the specific function.