Back to the main page.
Bug 532 - implement a ft_crossfreqanalysis function
| Status | CLOSED WONTFIX |
| Reported | 2011-03-15 14:20:00 +0100 |
| Modified | 2014-01-29 13:28:35 +0100 |
| Product: | FieldTrip |
| Component: | core |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Windows |
| Importance: | P1 enhancement |
| Assigned to: | Eelke Spaak |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Eelke Spaak - 2011-03-15 14:20:34 +0100
A FieldTrip implementation for cross-frequency (e.g., power-to-power coupling, phase-amplitude coupling) analysis might be nice. Angela Onslow has written a nice small PAC matlab toolbox, from which we might be able to adapt some algorithms. See http://www.cs.bris.ac.uk/Research/MachineLearning/pac/ for the toolbox (Onslow, Bogacz, & Jones, 2011, Progress in Biophysics and Molecular Biology 102:49-57)
Eelke Spaak - 2011-05-04 14:43:04 +0200
Mathilde is actually working on something like this, before you start working on it, it would be good to discuss, maybe with the three of us.
Johanna - 2011-05-04 14:45:47 +0200
Yes, I've just begun discussing with Mathilde also, which is why I assigned myself this bug, but of course welcome input if you wanted to do this bug instead. :-) We can the three of us meet.
Roemer van der Meij - 2011-11-30 16:50:08 +0100
Hey, just spotted this bug, wonderful! It's been a while so it seems :). How's it going?
Eelke Spaak - 2011-12-01 10:43:23 +0100
Robert does not want it :) The main argument being that there are very many methods of (especially) phase-amplitude coupling out there, and us implementing a subset of them would be taken as us endorsing that particular subset. I kind of see the point there, although I still do think it would be nice to have such a function.
Johanna - 2011-12-01 10:54:57 +0100
I haven't been doing much CFC analysis lately....so maybe better if someone else (Eelke or Roemer?) takes this bug who is more actively working on it? I also see Robert's point about not endorsing any one particular option. However, would be nice to have first a function with around 5 options (whichever are used most by the person writing the code!) and then more can be added later as individual requests or expertise come in. Or maybe this bug is resovled->won't-fix ?
Roemer van der Meij - 2011-12-01 12:08:40 +0100
Hmmm, well, both make sense I guess. The not endorsing a particular option and just implementing a few. Not looking forward to bookkeeping issues though :O. And it will be a looot of work to implement properly.
Johanna - 2012-01-31 14:28:33 +0100
Hi Eelke, I think it makes most sense if I re-assign this bug to you. I'm simply not working on this type of analysis now or in the forseeable future, and haven't done much of it in the past either. I will leave it to you to decide and discuss with others for whether/how this should be implemented.
Eelke Spaak - 2012-01-31 14:59:34 +0100
OK I'll just keep this in my list of things labeled 'might do'.
Eelke Spaak - 2012-09-21 11:29:54 +0200
This is a won't fix, lot of work and Robert does not want it.