OpenDvM

A comprehensive Python package for EEG, eye-tracking, and behavioral data analysis, built on top of MNE-Python. open_dvm covers the full pipeline from raw-data preprocessing through ERP analysis, time-frequency decomposition, multivariate decoding, and spatial encoding models -- especially suited to integrating EEG data with behavioral and/or eye-tracking experiments (e.g. via OpenSesame).

EEG analysis includes:

• Semi-automatic preprocessing pipeline (filtering, ICA, artifact rejection)
• ERP analysis
• Time-frequency decomposition
• Multivariate decoding (BDM)
• Forward encoding / channel tuning function models (CTF)
• Eye-tracking integration and quality control

Full documentation, including runnable tutorials with real output, is at open-dvm.readthedocs.io.

Installation

pip install open-dvm

For development (editable install with test/lint dependencies):

git clone https://github.com/dvanmoorselaar/open_dvm.git
cd open_dvm
pip install -e ".[dev]"

Quick Start

from open_dvm.analysis import ERP
from open_dvm.support import FolderStructure

# Load preprocessed data
df, epochs = FolderStructure().load_processed_epochs(sj=1, fname='ses_01_main', preproc_name='main')

# Compute condition-specific ERPs
erp = ERP(sj=1, epochs=epochs, df=df, baseline=(-0.2, 0))
erp.condition_erps(pos_labels={'target_loc': [2, 6]})

Tutorials

The tutorials/ folder contains 10 Jupyter notebooks that walk through the toolbox end to end, from preprocessing to advanced decoding. For rendered versions with real output already included, see the hosted tutorials.

Folder Structure

Analyses are run from project-specific scripts, pointed at a project folder containing:

1. eeg/raw (raw EEG files, .bdf or .edf)
2. behavioral/raw (raw behavior files, .csv)
3. eye/raw (raw eye-tracking files, if available: .asc or .csv)

For detailed file naming conventions and folder organization standards, see NAMING_CONVENTIONS.md.

Known Limitations

See KNOWN_LIMITATIONS.md for a list of known gaps, deferred features, and open methodological questions.

Contributing

Contributions are welcome -- see CONTRIBUTING.md for dev setup, running tests, and the PR process.

Citation

If you use open_dvm, please cite it via CITATION.cff (or GitHub's "Cite this repository" button). The accompanying methods paper is still in preparation; the citation entry will be updated with full details once it's available.

This toolbox is built on MNE-Python -- please also cite:

Gramfort, A., Luessi, M., Larson, E., Engemann, D. A., Strohmeier, D., Brodbeck, C., ... & Hämäläinen, M. S. (2014). MNE software for processing MEG and EEG data. Neuroimage, 86, 446-460.

Questions or Issues?

Please open a GitHub issue, or contact Dirk van Moorselaar at dirkvanmoorselaar@gmail.com.