sovabids

Visit the documentation

• sovabids is a python package for automating EEG to BIDS conversion (experimental MEG support).
• New to sovabids? Start with the Quickstart guide.

• sovabids can be used through (click to see the examples):

  1. its python API
  2. its CLI entry points
  3. its terminal user interface (TUI)
  4. its JSON-RPC entry points (needs a server running the backend)
  5. its minimal web-app GUI

Note

The advantage of the JSON-RPC way is that it can be used from other programming languages.

Limitation:

Do notice that at the moment the files have to be on the same computer that runs the server.

Warning

MEG support is experimental. sovabids correctly routes MEG data to the meg BIDS datatype, but does not expose MEG-specific metadata requirements such as empty-room recordings, manufacturer calibration files, or digitization coordinate systems. For complex MEG datasets (Elekta/Neuromag, CTF, KIT) those steps must be handled manually after conversion using MNE-BIDS directly — see the MNE-BIDS MEG conversion guide. The output format is automatically set to FIF for MEG data; you do not need to set non-bids.output_format manually (but you may still override it if needed).

Supported Formats

sovabids reads EEG files via MNE-Python's read_raw, which supports a wide range of formats including BrainVision, EDF/BDF, EEGLAB, FIF, CNT, KIT/SQD, CTF, and more. See the full list in the MNE documentation.

Output is always written as a valid BIDS dataset. The table below lists formats that can also be exported natively (i.e., the BIDS data files stay in that format rather than being converted to BrainVision):

Format	Extension	Extra needed	Notes
BrainVision	.vhdr	(core)	Default output format; pybv included in base install
EDF	.edf	sovabids[formats]	Requires edfio; date must be in 1985–2084
EEGLAB	.set	sovabids[formats]	Requires eeglabio; montage with fiducials needed
FIF	.fif	(core)	MNE native format; use for MEG data (experimental — see warning above)

Install export support for EDF and EEGLAB:

pip install "sovabids[formats]"

For all other readable formats, sovabids converts the data to BrainVision on output (the default mne-bids behaviour).

Tip

sovabids supports incremental conversion: files whose BIDS output already exists are skipped automatically, so you can safely re-run after adding new participants without re-converting the whole dataset. To force a full re-conversion, delete the output folder and start over. The Python API returns {'succeeded': [...], 'skipped': [...], 'failed': [...]} so you can distinguish newly converted files, skipped files, and failures. Skips are also logged at WARNING level so they appear in CLI output without -v.

Architecture

The main elements of sovabids are:

• A source path with the original dataset.
• A bids path that will be the output path of the conversion.
• A rules file that configures how the conversion is done from the general perspective.
• A mapping file that encodes how the conversion is performed to each individual file of the dataset.

Internally sovabids uses MNE-Python and MNE-BIDS to perform the conversion. In a sense is a wrapper that allows to do conversions from the command line.

Basic Usage

Terminal User Interface (sovatui)

sovabids ships an experimental terminal user interface (TUI) called sovatui. It lets you run a conversion without writing any command line code, guiding you through the full workflow in four tabs: Setup, Rules, Mappings, and Convert. It is optional (install it with the [tui] extra shown below); the CLI and Python API are the primary, fully supported interfaces.

Installation for TUI usage

This will install sovabids with the terminal user interface dependencies.

pip install "sovabids[tui]"

Running the TUI

sovatui

See the TUI tutorial video for a walkthrough and its example.

Command Line Interface (CLI) entry-points

Use sovabids through its CLI entry-points as follows:

Installation

pip install sovabids

sovapply

Use the sovapply entry-point to produce a mapping file from a source path, an output bids root path and a rules filepath.

sovapply source_path bids_path rules_path

By default the mapping file made will have the following filepath:

bids_path/code/sovabids/mappings.yml

sovaconvert

Use the sovaconvert entry-point to convert the dataset given its mapping file.

sovaconvert mapping_file

Using the experimental web GUI

Installation for WEB GUI usage

This will install sovabids for usage with an experimental web gui.

pip install "sovabids[gui]"

See the WEB GUI tutorial video for a walkthrough and its example.

Using the experimental bidscoin plugin

For the experimental bidscoin plugin, install the sovabids fork of bidscoin manually:

pip install "git+https://github.com/yjmantilla/bidscoin.git@sovabids"

Follow the example at https://sovabids.readthedocs.io/en/latest/auto_examples/bidscoin_example.html to see how to use the plugin.

Installation for developers

Fork this repo and run:

git clone https://github.com/<gh-username>/sovabids.git
cd sovabids
pip install -e ".[dev]"

Contributing

Contributions are welcome. See the Contributing guide for how to set up a development environment, run the tests, and open a pull request. The guide also lives at .github/CONTRIBUTING.md.

To report a bug, open an issue on the issue tracker. For usage questions and general help, use GitHub Discussions. The project follows a Code of Conduct.

Funding

Acknowledgments

sovabids is developed with the help of the following entities:

Academic Works

• Poster for the Big Data Neuroscience Workshop 2022 (Austin, Texas)
• Poster for OHBM 2022 Anual Meeting
• Video for OHBM 2022 Anual Meeting
• Poster for the eResearch Australasia Conference 2021

What does sova means?

sova is a contraction of 'eso va' which mean 'that goes' in spanish.

Nevertheless the real usage by the original developers is just to convey the idea of :

we will make it happen, we dont know how, but we will