How to manage your data at Hoplab¶

This workflow (or technically speaking, Standard Operating Procedures) is the product of a consultation process with the Book A Data Manager team in 2025-2026. They aim to provide guidelines for collecting, organizing, and analyzing research data across the modalities used in our studies.

Data collection¶

1. Ethical approval¶

Check whether the correct ethical approval is in place (see this page for further information on the ethical procedure).

For behavioral and EEG studies, you need ethical approval from SMEC via the PRET platform;
For fMRI studies and studies involving patients, you need to register your study at the CTC and Radiology department of UZ Leuven, become and authorized user of the MRI scanner and get ethical approval from EC Research UZ/KU Leuven (more detailed information here).
If your research is related to another study that has already been ethically approved, you can also file for an amendment to the existing approval. Note that for most studies in the NeuroSPACE project, ethical approval has already been obtained (fMRI: S70813; EEG and behaviour: G-2024-8685).

Important

Regularly update the number of participants you have tested (with a signed ICF) on each ethical application in the ethics tracker.

Important

Make sure you are using the latest and correct (approved) recruitment materials, information letters, demographics questionnaires and informed consent forms (ICFs). Contact Klara in case you are not sure which version you should use.

2. Create project folder (SharePoint)¶

Create a study-specific folder on your personal SharePoint:

Give it a sensible name, e.g. 2026-StudyAcronym-YourInitials
Required subfolders:
- confidential/
- sourcedata/
- BIDS/
Encrypt the confidential/ folder (how to encrypt)
Add a README in the root (see guidelines here)

Tip

Encryption keys can be stored in Passwordstate, with the keys shared with the appropriate persons within the group.

3. Set up your experiment¶

Follow the relevant instructions:

4. Data acquisition¶

If applicable, get acquainted with specific testing procedures (EEG/fMRI) and start collecting your data.

For behavioural experiments, this can either be using one of the specified online platforms (Pavlovia, Meadows, Prolific, Microsoft Forms), directly from your PC or using paper questionnaires.

5. Store raw data¶

Save the experimental output to the sourcedata/ folder in your Sharepoint following the folder structure specified here.
- For paper questionnaires: Digitize the (pseudonymized) data using either an OCR solution or manual transcription into spreadsheet software. Save the data in an open format. Note that in BIDS, .tsv is the preferred format for spreadsheets. Excel does not allow saving in .tsv; instead, you can export as .txt and change the file extension manually. You can also use this workaround to convert .csv files to .tsv files.
- For data collected through an online platform: Export the data from the platform and save it in an open format (Pavlovia supports .csv and .json; Meadows supports .csv, .json, .mat and .log; Prolific supports .csv; Microsoft Forms only supports .xlsx). Make sure that the participant ID you use to store the research data is unique and thus different from the participant's SONA or Prolific ID. You can store the link between both in the confidential folder on your Sharepoint.
- In EEG experiments, raw EEG data are saved in the .bdf format (on the Windows PC) and associated behavioral data in .csv and .log formats (on the Linux PC). Move the data from both PCs using an SSD to your personal Sharepoint site. Ensure the output files are pseudonymized as far as possible without impacting research results before proceeding (e.g. through the BESA Anonymizer and using BIDS compliant subject IDs).
- In fMRI experiments, raw MRI data are automatically (temporarily) saved in the PACS system on the acquisition PC. To allow for optimal BIDS conversion of the data, export the data in DICOM format (more information here and here) using an SSD. Make sure to do the export pseudonomously by assigning a code name to the data (in BIDS format, e.g., sub-01), and to stick to a consistent DICOM layout across scans. Transfer the data from the SSD to your personal Sharepoint site.

Store physically signed ICFs properly in the designated secure location (i.e., a locked cupboard in your office). They must be stored for a retention period of at least 10 years (SMEC) or 25 years (EC).

At the end of your study, you can bring them to the lab's physical data archive (ask Klara in case you don't know where that is).

7. Confidential data¶

Confidential personal data that is requested in the context of participant payment, communication of study results or at-home testing (name, phone number, email address, home address and/or bank account details) should be stored in the confidential/ folder on your Sharepoint.

Important

These are not part of the research data and need to be stored separately and encrypted to avoid reidentification of the participants (how to encrypt).

These data should be deleted once they are no longer relevant (e.g., when the study is closed), unless participants explicitly agreed to store them longer for a specific purpose (discuss this with your PI before deleting them).

Data organisation¶

1. Convert to BIDS¶

Reorganize the raw data files in your BIDS/ folder in accordance with the BIDS metadata standard (behavioural data; EEG data; fMRI data).

Personal demographic information you collect (age, education level, handedness, etc.) is considered research data if you use them in your analysis, so they need to be stored pseudonymously as well. In BIDS, these can be described in a participant file in the root directory of the dataset.
Make sure to add all the modality agnostic files the BIDS format requires, such that your experiment and dataset is properly documented (see also the guidelines here).
For fMRI experiments, you can find a detailed in-lab example on how to do this for fMRI data here. If possible (i.e., if it doesn't impact your research results), deface your images in this step to make the data more pseudonymous.
Note that you can include additional subdirectories within your BIDS/ directory, such as a stimuli/ directory to store your stimulus files, or a docs/ directory with supporting documents (e.g., test protocol, reports, presentations). The latter is not formally part of BIDS, but can be added to a to .bidsignore file to exclude from BIDS validation. You can also decide to keep this outside of the BIDS folder but save it in the Sharepoint project folder instead.
Once you are ready, you can check whether your folder is BIDS compatible here.

2. Backup to ManGO¶

Push the data in your sourcedata/ and BIDS/ folders on Sharepoint to ManGO (central storage) on a regular basis during data collection (e.g., every week or after N participants). See this page for more information on how to do that.

Important

Only pseudonymized data can be transferred to ManGO. Do not transfer your confidential/ folder: this information should stay on Sharepoint only and should definitely not be in further back-ups or hard drives that you carry around. The confidential/ folder might be deleted once it is no longer necessary to keep your pseudonymization key (to be discussed with your PI based on your ethics application).

Tip

You don't have to delete anything from Sharepoint after the transfer (back-up), but if you need more space you can consider deleting intermediate steps that can be reproduced based on the raw data and essential scripts.

Data preprocessing & analysis¶

If the different steps below are followed in quick succession (or certain steps are done several times), repeated downloads/uploads make less sense and might be cumbersome. Therefore, they may be omitted, but researchers should keep in mind that ManGO is the main storage location for the research data and should thus be kept up to date.

Depending on the workflow, you may want to mount the ManGO platform as a virtual folder and read the data from that point rather than (re)downloading the data.

Throughout the process, you should make sure that any new or modified scripts are uploaded to the *GitHub environment *(see the page on coding practices).

1. Download data¶

Download the data from Sharepoint or ManGO to a local computer (e.g., your laptop, a more performant local computer or the HPC infrastructure).

2. Preprocess data¶

Preprocess the data (see the guidelines for setting up your fMRI analysis environment, preprocessing of fMRI data and preprocessing of EEG data).

3. Perform analysis¶

For fMRI data, perform first-level (subject) analysis and second-level (group) analysis (via the SPM GUI or a batch script). If applicable, define subject-level or group-level ROIs (see this page), and perform multi-variate analysis (see this page).
For EEG data, you can find the analysis guidelines here.
For purely behavioural data, process them through your programming script (R, Matlab, Python).

4. Store preprocessing outputs¶

Organize and store all relevant preprocessing scripts and outputs in the appropriate BIDS/ folders on your Sharepoint (preferably in open formats where feasible, with relevant metadata included).

All outputs from preprocessing can be stored in a derivatives/ folder in the main BIDS dataset root following BIDS conventions.
Create one subfolder per pipeline (e.g., fmriprep/, first-level/, group-level/, mvpa-rsa/, mvpa-searchlight/).
Each subfolder must include a dataset_description.json file describing how the data were generated (e.g., software, version, pipeline). Optionally, include a README file for additional context or explanations.
Any code used to generate the derivatives from the source data (e.g., the analysis scripts and pipelines), may be included in a code/ subdirectory. Extra documentation (and relevant images) may be included in a docs/ subdirectory. Logs from running the code or other commands may be stored in a logs/ subdirectory.

4. Upload & clean up¶

Push the resulting data to ManGO using your preferred client and clear any data from your local device or the HPC environment.