Hello experts,
Just want to make sure I am on the right track with my statistics. I am comparing 2 groups rs-fMRI data with Melodic, have completed the post analysis with randomize with 5000 permutations, comparing only those with high IQ vs low IQ, to see what if any differences are found any of the group rs-fMRI networks. My question is, since I am querying all of the networks to see if any of the results are significant, is it necessary to feed my corrp randomize output files through FDR? Or should the pstat files be fed into FDR. If it is corrp, then would the following command be correct:
fdr -i dr_state3_ic_0003_tfce_corrp_tstat1.nii.gz --oneminusp -o fdr_ic_0003_corrp -q 0.05
Thank you,
Varina
|