The command looks correct, though I would recommend passing a mask to it (use the mask.mgh in the glmfit folder). What are the thresholds it comes up with?
On 11/13/2017 04:41 PM, Hossein Rejali wrote:
Hi,
Sorry I am very new to this field, forgive me for ignorance. I am trying to do multiple comparison correction on my data set that I ran through *mri_glmfit.* I am currently using *mri_fdr* as a means for multiple comparison correction. The problem I am having now is that after running *mri_fdr*, the output surface overlays are completely zero. This seems very unlikely from the clusters I had, well at least hopefully unlikely.
Below I have posted part of the code for my analysis that is calling the *mri_fdr *function:
mri_fdr --fdr $thresh --${sign} --i $SUBJECTS_DIR/Group_Analysis/R1_projdist0.${dist}.fwhm${fwhm}/GLM_Output/${hemi}.Output/Diff-con-scz-R1/sig.mgh nomask $SUBJECTS_DIR/Group_Analysis/R1_projdist0.${dist}.fwhm${fwhm}/GLM_Output/${hemi}.Output/Corrected/FDR-Output-${sign}-thresh${thresh}.mgh
Sorry for the mess but essentially, I am running though a couple of for loops that runs through variable values posted below:
for*thresh* in 0.5 0.1 0.05 0.01 for*sign* in neg pos for*hemi* in lh rh for*fwhm* in15 20 for*dist *in 2
I am hoping there is something wrong with the way I am calling mri_fdr, any help would be greatly appreciated.
Kind Regards,
Hossein Rejali
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer