Back to the main page.

Bug 2696 - provide documentation on the applicability of ft_networkanalysis

Reported 2014-09-22 13:20:00 +0200
Modified 2019-08-10 12:33:10 +0200
Product: FieldTrip
Component: documentation
Version: unspecified
Hardware: PC
Operating System: Mac OS
Importance: P5 normal
Assigned to: Tzvetan Popov
Depends on:
See also:

Tzvetan Popov - 2014-09-22 13:20:14 +0200

provide a short tutorial illustrating SOME ways of using graph theory in fieldtrip

Tzvetan Popov - 2014-10-17 13:43:04 +0200

Created attachment 670 node degree with dics and pcc filters

Tzvetan Popov - 2014-10-17 13:47:14 +0200

I updated the pipeline according to the suggestions from the last team meeting. I realized from this bug: that there will be 3 node degree maps for each of the dip directions, correct? I've attached a figure illustrating the result. I wonder why it deviates so dramatically?

Robert Oostenveld - 2014-10-17 14:10:24 +0200

after ft_sourceanalysis with pcc you would do ft_sourcedescriptives with projectmom=yes. That projects the source level data along the dipole orientation with maximal variance.

Tzvetan Popov - 2014-10-17 15:17:20 +0200

(In reply to Robert Oostenveld from comment #3) Yes I tried that, but I got stuck with an error: In an assignment A(I) = B, the number of elements in B and I must be the same. Error in ft_sourcedescriptives (line 487) source.avg.pow(i) = powmethodfun(source.avg.csd{i}(dipsel,dipsel)); need to look it up further.

Tzvetan Popov - 2014-11-05 14:38:46 +0100

In the ft meeting today we discussed: * add the possibility to plot 3D spaghetti accidents also in the plotting tutorial as a subheading of "plotting source level data". * make several notes throughout the tutorial for instance on the fact that special leakage is a problem more on the sensor level yet present also on source * biases of power, trial number etc. on connectivity estimates * the icoh producing false positives with signal sources that can be modeled with rotating dipoles such as the heart. * end up the tutorial with Summary part and suggested further reading. For instance include Schoffelen et al. HBM. * matlab or non FieldTrip code should be avoided- consider the creation of node.node and edge.edge file within existing ft functionality. * spoon feed the user by introducing breaks in both code sections and text sections. * introduce a section where the motivation behind choosing 'pcc' instead of 'dics' is made accessible to the reader. So far there are two instances of 'pcc'- the common filter tutorial and the networkanalysis one.

Tzvetan Popov - 2014-11-05 14:45:34 +0100

(In reply to Tzvetan Popov from comment #5) * make the motivation behind splitting the data into high and low alpha as accessible to the reader as it gets, i.e. make the analog to eyes open eyes closed where there are two conditions to begin with etc.

Robert Oostenveld - 2015-02-13 13:47:57 +0100

I just moved the zip file on the ftp from pub/fieldtrip to pub/fieldtrip/tutorial. Subject_rest should be renamed to SubjectRest (for the ds, hdr and img inside the zip file).

Tzvetan Popov - 2015-02-14 10:36:01 +0100

(In reply to Robert Oostenveld from comment #7) I made the changes and uploaded the new zip file.

Tzvetan Popov - 2016-02-24 09:22:58 +0100


Robert Oostenveld - 2019-08-10 12:33:10 +0200

This closes a whole series of bugs that have been resolved (either FIXED/WONTFIX/INVALID) for quite some time. If you disagree, please file a new issue on