The FLUX pipeline

FLUX: a pipeline for MEG analysis

Aim

Magnetoencephalography (MEG) allows for quantifying modulations of human neuronal activity on a millisecond time scale while also making it possible to estimate the location of the underlying neuronal sources. The technique relies heavily on signal processing and source modelling. To this end, there are several open sources toolboxes developed by the community. While these toolboxes are powerful as they provide a wealth of options for analyses, the many options also provide a challenge for reproducible research as well as for researchers new to the field. The FLUX pipeline aims to make the analyses steps and setting explicit for standard analysis done in cognitive neuroscience. The pipeline is derived from the Cogitate consortium addressing a set of concrete cognitive neuroscience questions. Specifically, the pipeline is defined for MNE Python (a Python toolbox) and FieldTrip (a Matlab toolbox) and a data set on visuospatial attention is used to illustrate the steps. The scripts are provided as notebooks implemented in Jupyter and Live Editor providing explanations, justifications and graphical outputs for the essential step. Furthermore, we also provide suggestions for text and parameter settings to be used in registrations and publications to improve replicability and facilitate pre-registrations. The FLUX pipeline is designed such that it can be used for education either in self-studies or guided workshops. We hope that the FLUX pipeline will strengthen the field of MEG by providing some standardization on the basic analysis steps and by aligning approaches across toolboxes. Furthermore, we also aim to support new researchers entering the field by providing education and training. The FLUX pipeline is not meant to be static; it will evolve with the development of the toolboxes and with new insight.

Target users

The target users are researchers new to MEG as well as more advanced users seeking to standardize their analyses. The FLUX pipeline is particular geared towards cognitive neuroscientists with an interest in task-based paradigms. The developers of the pipeline are cognitive neuroscientists with a strong interest in brain oscillations and multivariate approaches.

The data set

Download the ~4 Gb dataset from FigShare .  The raw data are composed of 4 files. The dataset is used in the step-by-step tutorials. It is from a simple paradigm on spatial attention. The participants are asked to attend left or right to a moving grating. The task and stimulation produce robust modulations in the alpha and gamma bands.

Recommendations on where to store the data are made in the tutorials below.

Standard Operation Procedure (SOP) for MEG data acquisition

We here share an example of a SOP for MEG data acquisition. It is derived from the Cogitate project and relevant parts can be adapted to specific studies and used for preregistrations.

The FLUX pipeline – MNE Python

The pipeline is implemented as Jupyter scripts maintained on a GitHub server

  1. Introduction and installation
  2. The dataset
  3. A first look at the data
  4. Artefacts annotations
  5. ICA for attenuating artefacts
  6. Extracting condition-specific trials
  7. Event-related fields
  8. Time-frequency representations of power 
  9. Classification using multivariate patterns analysis
  10. Source modelling using beamforming

The FLUX pipeline – FieldTrip

The pipeline is implemented as MAtlab Livescripts scripts maintained on a GitHub server 

 

  1. A first look at the data
  2. Extracting condition-specific trials
  3. Artefact rejection
  4. ICA for attenuating artefacts
  5. Event-related fields
  6. Time-frequency representations of power
  7. Source modelling using beamforming
CHBH black landscape

© 2020 Centre for Human Brain Health. All rights reserved.