Thanks a lot. It seems obvious now, as it often happens when someone handles you the answer...

One final question:

I've noticed that MELODIC_gui will only attempt to perform CONCAT-ICA when all the timeseries being concatenated have strictly the same number of timepoints. Is there a way to perform Concat-ICA on timepoint-dimension different sessions? Is it discouraged for some reason?

Thanks again! 

El martes, 25 de marzo de 2014, Stephen Smith <[log in to unmask]> escribió:
Hi

Yes those templates are in 2mm MNI152 standard space as supplied with FSL.  To regress those against your standard-space datasets, you will need to either resample the PNAS10 into your space, or vice versa. Probably the easiest (if not fastest) thing to do is to rerun your preprocessing and registration of your datasets, selecting the 2mm MNI152 standard space.

Cheers.



On 25 Mar 2014, at 15:38, Juan - Ignacio Reculé Rivera <[log in to unmask]> wrote:

Dear FSL experts:

Please have my kindest regards.

The group I participate in is trying to replicate the Dual_regression methodology as implemented in FSL package for our analysis; we intend to use the results in PNAS's paper "Correspondence of The Brain's Functional Architecture during activation and rest" as the template for obtaining our subjects' timecourses for stage 2 of the Dual Regression (as suggested in FSL's lecture slides).

We have been this far unable to do so. We have succesfully ran Dual_regression using our own Concat-ICA derived templates, but have been uncapable of using the maps that have been kindly uploaded for the benefit of the neuroscientific community.

We hypothesized that maybe dual regression only allows templates that share the same FOV and Voxel Size that the subjects being regressed unto; if so, we imagine this same scenario has been tackled before, as many people probably have wanted to use the PNAS paper as a template and not necessarily having the same data size.

If we're correct on that asumption, is there a preferred method for resizing the subjects or the templates to get the match? We already tried using fslroi for obtaining the same FOV, but to no avail.

Thank you SO much, in advance, both for whatever light you might spark to our problem, and for your stimulating work in fMRI. 

Juan - Ignacio Reculé Rivera
Médico - MD
Residente de Psiquiatría UC
Psychiatry Resident, Pontifical Catholic University of Chile 


--
Juan - Ignacio Reculé Rivera
Médico
Residente de Psiquiatría UC



---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director,  Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford  OX3 9DU, UK
+44 (0) 1865 222726  (fax 222717)
[log in to unmask]    http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------

Stop the cultural destruction of Tibet