1. The first step to running pyCapsid on the cluster is installing pyCapsid in dedicated conda environment as described in its installation documentation.
2. Once you’ve done that, activate the environment using conda activate and note the path to the python interpreter of that environment using type python. For example:
3. Create a folder in which you will store the results of pyCapsid. Note the path of the folder using pwd.
4. Inside that folder create a file named config.toml or download an example config.toml from here. Adjust the contents of this config file to the desired set of parameters. A more detailed config file with more of the available options can be found here.
5. Next create a python file with your preferred name, i.e. run_pycap.py. The contents of that file should look like this:

    #!/usr/dataB/luquelab/members/ctbrown/miniconda3/envs/pycapsid/bin/python # This is the path to your python interpreter noted in step 2
    #PBS -l nodes=1:ppn=24 # This requests a single node with 24 processors per node. This corresponds to the 4 higher quality nodes on the CSRC cluster. Remove the ppn requirement to use any node.
    #PBS -l walltime=18:00:00 # This specifies the maximum time the job will run before being terminated
    #PBS -d "/usr/dataB/luquelab/members/ctbrown/pyCapsid/pyCapsid/run/" # This specifies the working directory, and should be the directory you created in step 3
    #PBS -j oe # This combines the standard output and standard error logs
       
    from pyCapsid import run_capsid_report
    run_capsid_report('config.toml') # make sure the filename provided here is the same as the config file you created.
   

6. Check that you set the parameters in config.toml correctly and submit the job using qsub run_pycap.py.
7. This will create a folder with the same name that was specified in the config file. You can copy this folder to your local machine using the scp command.