We have been working on timecourse data that has had the prestim removed (at the mkcontrast-sess level). The timecourses obtained with roisummary-sess do not seem to correspond to the timecourses plotted with tksurfer. The discrepancy is two-fold: the peak latency and the overall timecourse shape. What could be the source of this discrepancy? This is particularly obvious for the ALC (alcohol) and PLAC (placebo) conditions - their peak activity is at the same timepoint in tksurfer but at different points in the text output (as shown in the Excel file) Included are: - The original text ouput file - Excel file with the text output along with the graph, - screen shots of the tksurfer timecourses and activity - on the left side is the timecourse of the activity with prestim removed and on the right is the timecourse with all the points We have also included some of the relevant script commands we used throughout the analysis.
The entire working directory can be found here...
/space/veles/3/users/sheeva/TEST-S51
This is the analysis I used... analysis.info
analysis SNfir-test TR 2 designtype event-related nconditions 4 parname para.para fsd bold_SN funcstem fmcsm8 tpexclude
-timewindow 18 -prestim 4 -polyfit 2 -TER 2 -nskip 0 -fwhm 0 -extreg mcextreg -nextreg 3
example of mkcont command...removing the prestim
entire script can be found here.../space/veles/3/users/sheeva/TEST-S51/mkcont-sess.csh
setenv aa 1 setenv cc 0 echo '0 0 0 1 1 1 1 1 1' | mkcontrast-sess -analysis SNfir-test -contrast CvFix-RMPS -a $aa -c $cc -rmprestim
stxgrinder-sess -analysis SNfir-test -contrast CvFix-RMPS -sf 51-SN-intragp
Finally, here is the csh that I used for making the timecourses....
setenv SUBJECTS_DIR /space/perun/3/users/cylix/Subjects
foreach roi (lh-AC-med)
foreach num (51) foreach ltr (a j)
mri_label2label --srclabel /space/veles/3/users/sheeva/fMRIanal/fsavg-labels/RMPS/surface/LH/${roi}.label --srcsubject fsaverage --trgsubject sub${num}_recon --trglabel sub${num}_recon/label/${roi}.label --hemi lh --regmethod surface
func2roi-sess -roidef ${roi}.label -labelfile /space/veles/3/users/sheeva/fMRIanal/fsavg-labels/RMPS/surface/LH/${roi}.label -d . -s sub${num}${ltr} -analysis SNfir-test
roisummary-sess -analysis SNfir-test -roidef ${roi}.label -d . -sf SN51-a-j-RMPS -sumfile TEST-S51aj/Unmasked/LH/${roi}-unmasked.txt
end end end
Thank you Elizabeth
