MNEPython

MNE-Python is an open-source library for processing magnetoencephalography (MEG) and electroencephalography (EEG) data in Python. It provides an end-to-end workflow for data import, preprocessing, analysis, and visualization, with a focus on reproducibility and interoperability with other scientific Python tools. The project is widely used in research and education to perform both sensor-space and source-space brain data analyses.

The library organizes data in core structures such as Raw for continuous data, Epochs for segmented trials, and Evoked for averaged responses. It supports reading data from common MEG/EEG formats (notably FIF from Elekta/Neuromag) and provides interfaces to convert and export data to other formats. MNE-Python also integrates with MRI data and anatomical models, enabling coregistration and forward modeling for source localization.

Analytical capabilities cover a broad range of steps in MEG/EEG analysis. Preprocessing includes filtering, resampling, artifact detection and removal (including ICA-based ocular and cardiac artifacts), and epoching with baseline correction. Time-domain analyses yield event-related potentials and fields, while time–frequency analysis uses Morlet wavelets and multitaper methods. Sensor-space analyses encompass averaging, topographic visualization, and functional connectivity measures. In source space, users can compute forward models (e.g., BEM), apply inverse solutions such as minimum-norm estimates, dSPM, and related methods, and visualize source activity. The package also supports statistical testing, including nonparametric permutation tests with cluster-based corrections.

MNE-Python is built on NumPy, SciPy, and Matplotlib, and interoperates with NiBabel, Nilearn, and other neuroscience tools. It is licensed under BSD-3-Clause and maintained by an active open-source community, with extensive documentation, tutorials, and examples to support users across levels of expertise.