Hi All,
I have a maybe weird question, but I'm hoping there might be a good answer...
I have a few nipype pipelines at this point, mainly connecting modular workflows to other modular workflows. Is there a clever way of maybe treating a workflow like a mapnode?
I ask because the pipeline I'm working on now can only accommodate 1 functional/resting state run at a time. I would like to push each functional run through the same workflow in some portions and then bring the outputs of those common workflows together for something like a combined registration step after the preprocessing is done. This would be easy if I could specify something like an iterfield for a workflow to treat it like a mapnode.
Anyway, any thoughts and feedback are much appreciated. This is more of a theoretical how-to kind of question than an immediate please-help-me-now question.
My initial thought and current solution will be to just instantiate the common workflow as many times as I have functional runs.
Thanks Some More (for such an awesome project),
~Eric