Skip to main content

Fetch metadata from EPU / SerialEM and launch on-the-fly pre-processing

Project description

A simple app to fetch acquisition metadata from a running EPU session or SerialEM. It parses the first found xml/mrc (EPU) or mdoc file (SerialEM) associated with a data collection session and launches Relion or Scipion pipeline. In case of SerialEM you need to enable saving mdoc file for each movie.


Dependencies are installed from pip automatically:

  • pyqt5 (GUI)

  • numpy (to parse MRC headers)

  • emtable (for some Relion schedules scripts)

  • watchdog (watch folder when running in daemon mode)


You can install either using pip or from sources.

  • from pypi (recommended): pip install MDCatch

  • from sources - you have two options:

  1. create python virtualenv:

python3 -m venv mdcatch
source mdcatch/bin/activate
git clone
cd MDCatch
pip install -e .
  1. create conda virtualenv (requires conda installed):

conda create -n mdcatch python=3.8
conda activate mdcatch
git clone
cd MDCatch
pip install -e .


  • Relion 3.1 or Scipion 3.0 is in in your PATH

  • Preprocessing templates: Schedules folder for Relion, template.json for Scipion

  • Edit to adjust it to your setup


To run with a GUI simply type mdcatch. If you want to run in daemon mode, run mdcatch –watch (or better setup a daily cron job)

Working principle

GUI mode (default)

The idea is to run the app on a processing server once EPU/SerialEM starts data collection and the first movie is acquired. The server has to have access to both EPU session folder and movies folder, or to SerialEM movie folder.

  1. find and parse the first xml/mdoc file, getting all acquisition metadata

  2. create a Relion/Scipion project folder username_microscope_date_time inside PROJECT_PATH (or inside Scipion default projects folder)

  3. create symlink for movies folder; copy gain reference, defects file, MTF into the project folder

  4. modify existing Relion Schedules/Scipion templates then launch Relion/Scipion on-the-fly processing

  5. setfacl -R -m u:uid:rwx is executed for the project folder, where uid is obtained from DEF_USER

Daemon mode

From version 0.9.7 onwards it’s possible to run the app in fully automatic mode. It will run in the background recursively watching for new directories (directory name should start with PREFIX, e.g. lmb_username_myEpuSession) inside METADATA_PATH. Once an xml (EPU) or a mdoc (SerialEM) file is created in such folder, the default pipeline will launch. All subsequent steps are equivalent to the GUI mode.


It’s probably useful to setup a daily cron job for mdcatch –watch to detect new EPU/SerialEM sessions.


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

MDCatch-0.9.8.tar.gz (54.9 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page