OK, I think I've fixed it. I've checked the fix into dev which should be live tomorrow. When you make the analysis, add -per-session to the cmd line. doug
Douglas N Greve wrote:
Hi Tapsya, I've verified the problem and am working on a fix. doug
tapsya@nmr.mgh.harvard.edu wrote:
Hi Doug,
It seems that there are still problems related to using -per-session. Preproc-sess now finishes apparently fine (correct file names are generated, i.e., fmc.*) but the analyses again defaults to using -per-run at mkanalysis-sess.
Specifically, mkanalysis-sess creates an analysis.info file (attached) where the parameters mcstem and nuisreg still point to -per-run (mcstem fmcpr and nuisreg mcprextreg 3). Consequently the rest of the analysis fails. I have attached the analysis.info file in the email
Thank you for your help again!
-Tapsya
Details:
Machine: adapt nmr-dev-env (vs 5.1)
preproc-sess -s avml07 -surface faverage lhrh -mni305 -per-session -so siemens -fsd bold -fwhm 5
set ana_BASIC_lh = ISIXTR_BERT_ERFIRsm5pf5tpefsub_18cat_lh
mkanalysis-sess -analysis $ana_BASIC_lh -surface fsaverage lh -fwhm 5 -TR 1.15 -paradigm AVML_BRERT_ISIXTR_001_18cat.par -event-related -nconditions 18 -fir 2.3 18.4 -mcextreg -polyfit 5 -runlistfile rlf_ISIXTR_BERT.txt -stc siemens -force -fsd bold -tpef tpef.txt
I just ran this co
mmand and it completed successfully
preproc-sess -s avml07 -mni305 -per-session -so siemens -fsd bold -fwhm
5
Is this what you ran? I'm also running it to generate the surfaces now,
but it does not seem to be complaining.
doug
Douglas N Greve wrote:
I'm currently running preproc-sess -s avml07 -mni305 -per-session -so siemens -fsd bold -fwhm 5 and it seems to be going ok, I won't know until tomorrow whether this
is really working or not
doug raij@nmr.mgh.harvard.edu wrote:
Still no luck. mc-sess is stubbornly using -per-run despite that the
preproc-sess requests per-sess. This may be where things get on the wrong
track. -Tommi
> Sorry, what I meant was for you to change the name and then run the >
stream as usual.
> doug > raij@nmr.mgh.harvard.edu wrote: > >> Done. I also made a backup of the registration. The entire >> directory at >> /space/adapt/1/users/raij/AVISI_FSFAST5.1/ >> is a copy of the actual data residing elsewhere (apart from the >>
Freesurfer
>> recons) so feel free to change anything you need. -Tommi >> >>> I think I've found the problem, but it's going to be a bit >>> difficult to >>> fix. Can you change the name of your registration file to >>> register.dof6.dat and remove write permissions so that it does >>> not >>> get >>> overwritten? >>> doug >>> raij@nmr.mgh.harvard.edu wrote: >>> >>>> Yes we have. It exits with the same error as below. -Tommi >>>> >>>>> Have you tried specifying your regfile with -regfile? >>>>> doug >>>>> raij@nmr.mgh.harvard.edu wrote: >>>>> >>>>>> Hi Doug, >>>>>> Thank you for your help today! I tried running preproc-sess >>>>>>
again
>>>>>> but >>>>>> it >>>>>> again exits with per-session related errors. However, the >>>>>> errors >>>>>>
are
>>>>>> different than before so I think the changes you made were a >>>>>>
step to
>>>>>> the >>>>>> right direction. This time >>>>>> preproc-sess \ >>>>>> -sf subjectname \ >>>>>> -df sessdir \ >>>>>> -mni305 \ >>>>>> -per-session \ >>>>>> -sliceorder siemens \ >>>>>> -fsd bold \ >>>>>> -fwhm 5 \ >>>>>> -surface $SUBJECT lhrh \ >>>>>> -noreg \ # I want to use my manually adjusted >>>>>> bold/register.dof6.dat >>>>>> file >>>>>> (1) mc-sess output is still saved with the per-run stem (pr) >>>>>>
despite
>>>>>> that >>>>>> we defined -per-session. >>>>>> (2) stc-sess output is still saved with the per-run stem (pr) >>>>>>
despite
>>>>>> that >>>>>> we defined -per-session. >>>>>> (3) rawfunc2surf-sess defaults to -per-run and then expects to >>>>>> find >>>>>> instem >>>>>> fmcpr.siemens >>>>>> (4) ERROR: cannot find >>>>>> /autofs/space/adapt_001/users/raij/AVISI_FSFAST5.1/avml07/bold/002/register.dof6.dat. >>>>>> >>>>>>
Apparently the script looks for the registration file inside the run
>>>>>> directory (as it should for per-run analysis) instead of >>>>>> the fsd >>>>>>
<bold>
>>>>>> where it should reside for a -per-session analysis. This >>>>>> forces >>>>>> the >>>>>> script >>>>>> to exit. >>>>>> I changed permissions so that you can write in the study >>>>>> directory. >>>>>> For details please see below. >>>>>> Thanks! >>>>>> -Tommi & Tapsya >>>>>> on machine avml >>>>>> nmr-dev-env (vs 5.1) >>>>>> cd /space/adapt/1/users/raij/AVISI_FSFAST5.1/avml07 >>>>>> setenv SUBJECTS_DIR /space/adapt/1/users/tapsya/subject_MRIs >>>>>>
setenv SUBJECT avml07
>>>>>> logs in /space/adapt/1/users/raij/AVISI_FSFAST5.1/avml07/log >>>>>>
preproc-sess log file
>>>>>> $Id: preproc-sess,v 1.48 2011/11/14 17:41:35 greve Exp $ raij >>>>>> setenv FREESURFER_HOME /usr/local/freesurfer/dev >>>>>> setenv SUBJECTS_DIR /space/adapt/1/users/tapsya/subject_MRIs >>>>>>
Linux avml 2.6.32-131.12.1.el6.x86_64 #1 SMP Mon Sep 26
>>>>>> 10:13:43 BST >>>>>> 2011 >>>>>> x86_64 x86_64 x86_64 GNU/Linux >>>>>> /autofs/space/adapt_001/users/raij/AVISI_FSFAST5.1/avml07 >>>>>>
/usr/local/freesurfer/dev/fsfast/bin/preproc-sess
>>>>>> -sf subjectname -df sessdir -mni305 -per-session -sliceorder >>>>>>
siemens
>>>>>> -fsd >>>>>> bold -fwhm 5 -surface avml07 lhrh -noreg >>>>>> Mon Nov 14 20:07:57 EST 2011 >>>>>> instem f >>>>>> mc 1 f fmcpr >>>>>> stc 1 fmcpr fmcpr.siemens >>>>>> sm 0 >>>>>> mask 1 brain >>>>>> ---------------------------- Original Message >>>>>> ---------------------------- >>>>>> Subject: [FreeSurfer]: preproc-sess error >>>>>> From: tapsya@nmr.mgh.harvard.edu >>>>>> Date: Thu, November 10, 2011 1:41 pm >>>>>> To: freesurfer@nmr.mgh.harvard.edu >>>>>> Cc: raij@nmr.mgh.harvard.edu >>>>>> -------------------------------------------------------------------------- >>>>>> >>>>>>
Dear Doug,
>>>>>> We have encountered errors running preproc-sess (FSFAST v5.1) >>>>>>
command
>>>>>> with >>>>>> the flag "-per-session". The command line is >>>>>> on machine adapt (same result on machine avml) >>>>>> cd /space/adapt/1/users/raij/AVISI_FSFAST5.1/avml07 >>>>>> setenv SUBJECTS_DIR /space/adapt/1/users/tapsya/subject_MRIs >>>>>>
setenv SUBJECT avml07
>>>>>> nmr-std-env (=FS 5.1) >>>>>> preproc-sess \ >>>>>> -sf subjectname >>>>>> -df sessdir \ >>>>>> -mni305 \ >>>>>> -per-session \ >>>>>> -sliceorder siemens \ >>>>>> -fsd bold \ >>>>>> -fwhm 5 \ >>>>>> The error message: >>>>>> ... >>>>>> stc-sess Done >>>>>> avml07 To MNI305 ---------------- >>>>>> rawfunc2tal-sess -fwhm 5 -s avml07 -d >>>>>> /autofs/space/adapt_001/users/raij/AVISI_FSFAST5.1 -fsd bold >>>>>>
-stc
>>>>>> siemens -update -subcort-mask >>>>>> ---------------- >>>>>> 1/1 avml07 >>>>>> 1/1 avml07 002 ---------------- >>>>>> "ERROR: could not determine file for >>>>>> /autofs/space/adapt_001/users/raij/AVISI_FSFAST5.1/avml07/bold/002/fmcpr.siemens" >>>>>> >>>>>>
Hence it looks like rawfunc2tal-sess is searching for -per-run files (stem fmcpr*) rather than -per-session files (stem fmc*). We tried to take control over the stems and added the following lines:
>>>>>> -mcin f \ >>>>>> -mcout fmc \ >>>>>> -stcin fmc \ >>>>>> -stcout fmcstc \ >>>>>> -smin fmcstc \ >>>>>> -smout fmcstcsm5 \ >>>>>> OR (modern naming conventions): >>>>>> -mcin f >>>>>> -mcout fmc >>>>>> -stcin fmc >>>>>> -stcout fmc.siemens \ >>>>>> -smin fmc.siemens \ >>>>>> -smout fmc.siemens.sm5.mni305.2mm \ >>>>>> However, these resulted in the same error. >>>>>> We also tried this on a more recent data set and received the >>>>>> same >>>>>> error. >>>>>> In summary, it appears that rawfunc2tal-sess (and in fact also >>>>>>
rawfunc2surf-sess) expect the stem fmcpr even when we use per-session
>>>>>> (where they should expect the stem fmc). >>>>>> Any help would be greatly appreciated! >>>>>> Best regards, >>>>>> Tapsya >>>>>> >>>>> -- >>>>> Douglas N. Greve, Ph.D. >>>>> MGH-NMR Center >>>>> greve@nmr.mgh.harvard.edu >>>>> Phone Number: 617-724-2358 >>>>> Fax: 617-726-7422 >>>>> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting >>>>> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html >>>>> >>> -- >>> Douglas N. Greve, Ph.D. >>> MGH-NMR Center >>> greve@nmr.mgh.harvard.edu >>> Phone Number: 617-724-2358 >>> Fax: 617-726-7422 >>> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting >>> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html >>> > -- > Douglas N. Greve, Ph.D. > MGH-NMR Center > greve@nmr.mgh.harvard.edu > Phone Number: 617-724-2358 > Fax: 617-726-7422 > Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting > FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html >
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer