[Mne_analysis] Group level sensor space permutation cluster t-test on n channels?

[Peter Goodin]

Dear Denis,

Thank you very much for the code snippet and neighbour definitions advice! As usual the problem was simply user error. My evoked data was being imported and placed into an array of arrays instead of a single multi-dimensional array, so the array object properties (specifically x.ndim) wasn't being assigned.

Peter
__________________________
Peter Goodin,
BSc (Hons), Ph.D Candidate.

Brain and Psychological Sciences Research Centre (BPsych)
Swinburne University,
Hawthorn, Vic, 3122
http://www.swinburne.edu.au/swinburneresearchers/index.php?fuseaction=profile&pid=4149

Monash Alfred Psychiatry Research Centre (MAPrc)
Level 4, 607 St Kilda Road,
Melbourne 3004
________________________________
From: mne_analysis-bounces at nmr.mgh.harvard.edu [mne_analysis-bounces at nmr.mgh.harvard.edu] on behalf of Denis-Alexander Engemann [denis.engemann at gmail.com]
Sent: Sunday, 8 June 2014 11:35 PM
To: Discussion and support forum for the users of MNE Software
Subject: Re: [Mne_analysis] Group level sensor space permutation cluster t-test on n channels?

Hi Peter,

On Sun, Jun 8, 2014 at 3:01 PM, Peter Goodin <pgoodin at swin.edu.au<mailto:pgoodin at swin.edu.au>> wrote:
Hi MNE list,

I'm having a bit of trouble on getting group level sensor space stats going. I've run similar in Fieldtrip and just create an array (or matrix in matlab speak) consisting of participants x channels x samples, and looking through the examples it seems we need to work directly with the array data too. But none of the functions I've used seem to accept anything other than a 2d array (3d for TF) and all the examples seem to either be single channel or one sample t-tests.

you can use all  `mne.stats.permutation_cluster_XXX` functions. They accept mutli-dimensional arrays as input. Irrespective of the cluster-forming stat used (t-test, f-test) they will perform a cluster permutation test.
For sensor space data you first have to assemble a corresponding matrix / array.

X = np.array([e.data for e in m_evoeds])

To yield meaningful results you need to specify a connectivity structure, e.g., based on the fieldtrip neighbor definitions.
C.f. https://github.com/mne-tools/mne-python/pull/1306 which we recently added.

The input matrix is then expected to be of size n_subjects, n_times, n_channels.

HTH,
Denis

Are group level stats available currently or does something need to be written / cobbled together to get them to work?

Thanks for any help,

Peter

__________________________
Peter Goodin,
BSc (Hons), Ph.D Candidate.

Brain and Psychological Sciences Research Centre (BPsych)
Swinburne University,
Hawthorn, Vic, 3122
http://www.swinburne.edu.au/swinburneresearchers/index.php?fuseaction=profile&pid=4149

Monash Alfred Psychiatry Research Centre (MAPrc)
Level 4, 607 St Kilda Road,
Melbourne 3004

_______________________________________________
Mne_analysis mailing list
Mne_analysis at nmr.mgh.harvard.edu<mailto:Mne_analysis at nmr.mgh.harvard.edu>
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis


The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20140609/ea726024/attachment.html