Software name: 

Scipion is available to all HPC2N users.


Scipion is an image processing framework to obtain 3D models of macromolecular complexes using Electron Microscopy.


Scipion is an image processing framework for obtaining 3D models of macromolecular complexes using Electron Microscopy (3DEM). It integrates several software packages and presents a unified interface for both biologists and developers. Scipion allows you to execute workflows combining different software tools, while taking care of formats and conversions. Additionally, all steps are tracked and can be reproduced later on.


On HPC2N we have Scipion available as a module on Kebnekaise.

Usage at HPC2N 

To use the Scipion module, first add it to your environment. Use:

module spider Scipion

to see which versions are available and how to load the module and the needed prerequisites.

You can read more about loading modules on our Accessing software with Lmod page and our Using modules (Lmod) page.

Scipion is best used with its graphical frontend. To use the graphical frontend one should login to our Thinlinc based login node, see our support page for how to use thinlinc.

There are a few things that must be done before starting Scipion for the first time at HPC2N.

Scipion uses a per-user config file which is located in a directory that is not readable by batch system nodes. To fix this use the following sequence of commands.

rm -rf $HOME/.config/scipion
mkdir -p $HOME/Public/.config/scipion
mkdir -p $HOME/.config
ln -s ../Public/.config/scipion $HOME/.config

Python is using a cache for unpacking subclasses, this is located in $HOME/.cache/Python-Eggs, which is not writable from batch jobs.

So this also need to be redirected, but to /pfs/nobackup$HOME/.cache so batch jobs can write into it.

rm -rf $HOME/.cache/Python-Eggs
mkdir -p /pfs/nobackup$HOME/.cache/Python-Eggs
ln -s /pfs/nobackup$HOME/.cache/Python-Eggs $HOME/.cache

Now load the Scipion module and do

scipion config

This will create a scipion.conf file in $HOME/.config/scipion (which is now located under $HOME/Public).

Scipion also uses a separate directory for storing the projects in, the directory is defined in your scipion.conf file.

Change the SCIPION_USER_DATA definition in that file to use a directory in the parallel file system, /pfs/nobackup, we recommend to set it to /pfs/nobackup/home/x/xzzzz/ScipionUserData, where /home/x/xzzzz should match your normal home directory path.

Differences in the GUI from standard installations

At HPC2N we have modified how to run the different protocols to be able to use the batch system more efficiently. When opening a Scipion protocol definition, the "Host" field have multiple entries.

The "localhost" entry can be used to run shorter, non-cpu intensive tasks, on the login node, or submit tasks to the batch system and run on CPUs only, good for any protocol that doesn't utilize GPU resources. There is also a "GPU_batch_job" entry that can be used to submit GPU based tasks to the batch system.

When submitting tasks to the batch system one have to fill in the "SNIC Project" field. When using the "GPU_batch_job" selection one also have to choose which of the two GPU types to run on by selecting either the "GPU - K80" or the "GPU - V100" queue. Please read the support pages regarding how to use the GPU resources at HPC2N.

NOTE: Motioncor2 currently has a bug when running on the V100 GPUs and produces bad output. Use the K80s instead.

Additional info 

More information about Scipion and how to run it can be found on the Scipion home page.

Updated: 2019-05-17, 14:52