I am using mne_bids_pipeline and have successfully run my data set up to the end of step 4 - sensor space analysis. I have questions about using Maxfilter for realignment of data due to head movement (1) within and (2) across participants. I am trying to understand whether I can do this within the pipeline and the most appropriate way I should do it/ what aspects of the scripts I need to modify.
Using mne BIDS naming conventions/ structure, in my dataset, participants have 6 tasks, each of which has a single run.
(1) I want to compare and potentially combine data across the 6 tasks and therefore would like to realign data to a single task within participants. In the pipeline there is the option to define mf_reference_run but as it stands this requires specification of a run number when I want to specify a task.
(2) I would like to do sensor-level group analysis of the magnitude of evoked responses and therefore want to realign data across participants. In the past the way I did this was to identify a participant with the most average head position (based on coordinates of all participants) then align all participants to that position (for sensor-analysis only; I use original for source analysis).
Given my goal of using the pipeline as much as possible in early stages of my data processing, please do you have any suggestions as to the best way to achieve these steps? I can see that (2) is perhaps best performed after/outside of the pipeline, but I feel like (1) should be possible within the pipeline.
With thanks for advice and thoughts.
- MNE version: 1.2.1
- operating system: Centos 7