External Email - Use Caution
I'm using version 6.0.0.
2020년 8월 20일 (목) 오전 12:08, Douglas N. Greve dgreve@mgh.harvard.edu님이 작성:
what version of fs are you using?
On 8/19/2020 9:03 AM, 정현우 wrote:
External Email - Use CautionHi Douglas,
Thank you for your response. I followed your suggestions and additionally tried using Matlab instead of Octave for FsPalm analysis, but I still failed to get the results. Adding $FREESURFER_HOME/fsfast/toolbox to my matlab path removed the first error output <error: 'fast_p2z' undefined near line 1 column 11>, but I still get the other error output <error: cannot find expected palm output "fsp_clustere_tstat_fwep.mgz" - check for run_palm.m failures'>.
Do you have any suggestions for this problem?
Thanks, Hyunwoo
2020년 8월 13일 (목) 오후 10:37, Douglas N. Greve dgreve@mgh.harvard.edu님이 작성:
On 8/12/2020 2:44 AM, 정현우 wrote:
External Email - Use CautionHello Freesurfer Developers,
I'm trying to do permutation simulation on 50 subjects to find brain region clusters of which cortical thicknesses have significant association with clinical parameters while controlling for age and sex. I tried this via 'mri_glmfit-sim' but it seems that permutation analysis could not be done with non-orthogonal variables in Freesurfer. I copied the command I used and the error output below.
mri_glmfit-sim --glmdir NK_IES_avoidance_correlation.rh.glmdir --perm 1000 1.3 abs --cwp 0.05 --2spaces
ERROR: design matrix is not orthogonal, cannot be used with permutation. If this something you really want to do, run with --perm-force
What version of FS are you using? If Version 6, then make sure you download the patch. See "If You're Not in an organized course" from https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/MultipleComparisonsV6.0...
So I tried using PALM, which allows permutation analysis with non-orthogonal variables according to a post in the Freesurfer Archive. I installed PALM, FsPalm and Octave on my computer and ran the command below, but I ran into another error output (copied below) that I do not quite understand.
You need to add $FREESURFER_HOME/fsfast/toolbox to your matlab path
fspalm --glmdir NK_IES_avoidance_correlation.rh.glmdir --cft 1.3 --twotail --name perm.NK_IES_avoidance_correlation.rh --iters 1000 --2spaces --cwp .05 --octave
preparing a palm subdirectory in /media/sjkim/hd2/subject_data/glm/NK_IES_avoidance_correlation.rh.glmdir/perm.NK_IES_avoidance_correlation.rh found NK_IES_avoidance_correlation.rh.glmdir/surface file - this is a surface-based analysis using the glmfit input file /media/sjkim/hd2/subject_data/glm/NK_IES_avoidance_correlation.rh.10.mgh using area file /media/sjkim/hd2/subject_data/fsaverage/surf/rh.white.avg.area.mgh palm prep complete - running the generated matlab script... error: 'fast_p2z' undefined near line 1 column 11 error: cannot find expected palm output "fsp_clustere_tstat_fwep.mgz" - check for run_palm.m failures
I copied the matlab script 'run_palm.m' below.
hemi = 'rh'; input = '/media/sjkim/hd2/subject_data/glm/NK_IES_avoidance_correlation.rh.10.mgh'; maskfile = '/media/sjkim/hd2/subject_data/glm/NK_IES_avoidance_correlation.rh.glmdir/mask.mgh'; surffile = '/media/sjkim/hd2/subject_data/fsaverage/surf/rh.white'; areafile = '/media/sjkim/hd2/subject_data/fsaverage/surf/rh.white.avg.area.mgh';
%% Changing threshold for two-tailed test cft = 1.3 + 0.301; pthresh = 10^-cft; zthresh = fast_p2z(pthresh); iters = 1000;
zthreshstr = sprintf('%f',zthresh); itersstr = sprintf('%d',iters);
palm('-i',input,'-m',maskfile,'-d','design.mat','-t','design.con','-logp',... '-n',itersstr,'-C',zthreshstr,'-o','fsp','-twotail','-s',surffile, areafile); return;
Do you have any suggestions on how I could figure out this problem? Of note, I am using Freesurfer v6.0 and Ubuntu 18.04.4 LTS platform.
Thanks in advance, Hyunwoo Jeong
Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer