Dear freesurfers,
I contact you since I ran a PCA with mri_glmfit. I do not manage to load the output v.mgh as an overlay as suggested in the documentation, which makes me wonder what is wrong. The output file or me loading it?
For clarification, I ran the command: mri_glmfit --surf fsaverage lh --y fmcpr.siemens.sm4.fsaverage.lh.nii.gz --pca --osgm --glmdir /Users/Florence/Documents/test_glm_pca to run the PCA in surface.
v.mgh is of size 1*163842(nb of vertices)*1*300 (number of components). On the one hand, the output seems to make sense (its size, and the values are centered around 0), on the other hand I do not understand why I do not see anything when I load it as an overlay in tksurfer, despite modifying the thresholds and color parameters etc. Or should I then save separately each spatial component contained in v.mgh, and load each of them as one overlay? I would also like to specify that my input data (fmcpr.siemens.sm4.fsaverage.lh.nii.gz) are of size 1*163842(nb of vertices)*1*300 (time points). This is the data from a single subject for now (before getting group data via mris_preproc) and I am not sure of whether it has the proper format. Indeed, it is specified that —y is the “path to input file with each frame being a separate input”. I do not understand what it means: does frame refer to one time point, and should I then have 300 input files (which seems a lot!)?
Thank you very much,
Florence