Skip to content

Fill in the Conversion Configuration File

The Salzburg MRI template creates a file called .mri_conversion_config.yml in your dataset project. This file tells the automatic conversion pipeline how to process your data. You need to fill it in before your first acquisition.

Editing the file in GitLab

  1. Navigate to your dataset project in your group on the ANC.
  2. In the file list, click .mri_conversion_config.yml. You may need to enable Show hidden files to see it, as the filename starts with a dot.
  3. Click the Edit button (pencil icon) in the top right of the file view and select Edit single file.
  4. Make your changes directly in the editor.
  5. When done, scroll down, add a short commit message (e.g. Fill in conversion config), and click Commit changes.

File contents

The template file looks like this:

project_slug: #exchange 'project_slug' with the actual project slug, e.g. 'anclang'
  volumesToDiscard: 6 #Number of initial volumes to discard [required, integer]
  hashParticipantID: true #Whether to hash participant IDs for anonymization [required, boolean (true, false)]
  email: #List of email addresses an email should be sent to after successful conversion [required, list of strings]
    - barbara.strasser-kirchweger@plus.ac.at
    - andreas.huttegger@plus.ac.at
  bidscoinMap: /path/to/file #for non-standard sequences [optional, string] file should be stored in the repository

Fields

project_slug (required)
Replace project_slug with your actual project slug — the URL identifier of your dataset project (e.g. the-brain). This must match the slug you set when creating the project.
volumesToDiscard (required, integer)
Number of initial volumes to discard at the start of each functional run. Typically 6 for Salzburg sequences — check with your acquisition protocol if unsure.
hashParticipantID (required, boolean)
Whether to hash participant IDs for anonymisation before conversion. Set to true to anonymise, false to keep IDs as-is.
email (required, list)
One or more email addresses that will be notified after a successful conversion. Add one address per line, each preceded by -.
bidscoinMap (optional, string)
Path to a custom BIDScoin mapping file, relative to the repository root. This is only needed if your sequences are not named according to the sequence naming conventions. If all your sequences follow the standard naming, you can leave this entry out entirely.

Example — fully filled in

the-brain:
  volumesToDiscard: 6
  hashParticipantID: false
  email:
    - barbara.strasser-kirchweger@plus.ac.at
    - andreas.huttegger@plus.ac.at
  bidscoinMap: ./bidscoin_map_coach23.yaml

In this example, participant IDs are not hashed, two people receive notification emails, and a custom BIDScoin map is used for non-standard sequences.