Hello Melissa,
Two subjects isn't really enough to run randomise with - hence only 2 permutations are possible.
Kind Regards
Matthew
> Hello Matthew,
>
> I found some error messages on several drD.o files:
>
> randomise options: -i temp.dr2/dr_stage2_ic0000 -o temp.dr2/dr_stage3_ic0000 -m temp.dr2/mask -d GLM_design2.mat -t GLM_design2.con -n 5000 -T -V
> Loading Data:
> Data loaded
> 2 permutations required for exhaustive test of t-test 1
> Doing all 2 unique permutations
> Starting permutation 1 (Unpermuted data)
> Warning: The unpermuted statistic image for the current image contains no positive values, and cannot be processed with TFCE. A blank output image will be created.
> Starting permutation 2
> Critical Value for: temp.dr2/dr_stage3_ic0000_tfce_corrp_tstat1 is: 0
> 2 permutations required for exhaustive test of t-test 2
> Doing all 2 unique permutations
> Starting permutation 1 (Unpermuted data)
> Warning: The unpermuted statistic image for the current image contains no positive values, and cannot be processed with TFCE. A blank output image will be created.
> Starting permutation 2
> Critical Value for: temp.dr2/dr_stage3_ic0000_tfce_corrp_tstat2 is: 0
> Finished, exiting.
>
> How should I change my glm set up?
>
> This is my GLM_design.con:
> /ContrastName1 mean
> /ContrastName2 group1>group2
> /NumWaves 2
> /NumContrasts 2
> /PPheights 1.000000e+00 1.000000e+00
> /RequiredEffect 5128651.500 5128651.500
>
> /Matrix
> 1.000000e+00 -1.000000e+00
> -1.000000e+00 1.000000e+00
>
>
> This is my GLM_design.mat:
> /NumWaves 2
> /NumPoints 2
> /PPheights 1.000000e+00 1.000000e+00
>
> /Matrix
> 1.000000e+00 0.000000e+00
> 0.000000e+00 1.000000e+00
>
>
> I only used two subjects.
>
> Thank you so much in advance,
>
> -Melissa H.
> ________________________________________
> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Matthew Webster <[log in to unmask]>
> Sent: Friday, March 4, 2016 11:03 AM
> To: [log in to unmask]
> Subject: Re: [FSL] fslnvols error from dual regression
>
> Hello Melissa,
> The melodic_IC file will be in resampled MNI152 space ( the default in the GUI is 4mm ). To save space you can just resample the MNI brain to the same resolution, and view images on top of that. It is unlikely that all the corrp files from an analysis will be blank - it is suggestive that something may have gone wrong. I would look carefully through the logs in the log directory, especially if any error logs ( *.e* ) are not empty.
>
> Kind Regards
> Matthew
>
>> Hello Matthew,
>>
>> It did work after putting the full path to the melodic_IC. However, I was trying to run this command:
>>
>> fslview $<full path>/groupmelodic.ica/MNI152_T1_2mm_brain <full path>groupmelodic.ica/melodic_IC -l "Red-Yellow" -b 4,15
>> (had to put the MNI file on my own folder because it said unreadable from the FSLDIR)
>>
>> I got a message error that says: "All overlays must have same dimensions as the base image"
>> So, I decided to flirt the melodic_IC.nii.gz although it used MNI152_T1_2mm_brain during the Melodic process. After this, I was able to open both and tried to add dr_stage3_ic0000_tfce_corrp_tstat1.nii.gz but I got the same message: All overlays must have same dimensions as the base image." Should I flirt the dr_stage3 files as well? I tried and got an error message..
>>
>> Did I miss a step between the melodic process and dual_regression... or after dual_regression..?
>>
>> Also, when I tried to check the corrected p-values.. using:
>> for i in dr_stage3_ic00??_tfce_corrp_tstat?.nii.gz ; do
>> echo $i `fslstats $i -R`
>> done
>>
>> This was the output:
>> dr_stage3_ic0000_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
>> dr_stage3_ic0000_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
>> dr_stage3_ic0001_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
>> dr_stage3_ic0001_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
>> dr_stage3_ic0002_tfce_corrp_tstat1.nii.gz 0.000000 0.000000
>> dr_stage3_ic0002_tfce_corrp_tstat2.nii.gz 0.000000 0.000000
>> It was the same all the way to component 29!.. Is this normal??
>>
>> Thank you so much in advance,
>>
>> -Melissa H.
>>
>> ________________________________________
>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Matthew Webster <[log in to unmask]>
>> Sent: Tuesday, March 1, 2016 10:56 AM
>> To: [log in to unmask]
>> Subject: Re: [FSL] fslnvols error from dual regression
>>
>> Hello Melissa,
>> Can you let me know the full command line call you used, also can you confirm that in the directory you are running dual_regression in, groupmelodic.ica/melodic_IC.nii.gz exists?
>>
>> Kind Regards
>> Matthew
>>> Hello,
>>>
>>> I finally got all the permissions to work on fsl on this computer. However, I'm now getting a new error message after running dual regression:
>>>
>>> DrC.e: Could not open matrix file groupmelodic.ica/melodic_IC
>>> ols_dof: Error in determining the trace, resorting to basic calculation
>>> ols_dof: Error in determining the trace, resorting to basic calculation
>>> Cannot open volume temp.dr.xfm/dr_stage2_subject00000 for reading!
>>>
>>> I got dr_stage1 and dr_stage2 files but not 3. Also, when setting up my GLM con and mat, I didn't get any "filest.txt" so I created it myself using the filtered_func_data from the "reg_standard" folder. I also tried to flirtered the filtered func data to see if I get any diff output but still the same error..
>>>
>>> What could I be doing wrong?
>>>
>>> Thank you in advance,
>>>
>>> -Melissa H.
>>>
>>> ________________________________________
>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Niels Bergsland <[log in to unmask]>
>>> Sent: Wednesday, February 24, 2016 3:21 PM
>>> To: [log in to unmask]
>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>
>>> Hi -
>>> I just meant that it was normal that you didn't get any output from
>>> running ". $FSLDIR/etc/fslconf/fsl.sh" That file should just set up
>>> your environment for using FSL.
>>>
>>> After you run that command, what happens if you try running in that terminal:
>>>
>>> /usr/share/fsl/5.0/bin/fslnvols
>>>
>>> On Wed, Feb 24, 2016 at 8:34 PM, Hidalgo, Melissa <[log in to unmask]> wrote:
>>>> Hello Niels,
>>>>
>>>> What does that mean? Is there a way that I can change the fsl.sh? or do you think I should re-install fsl, would that change anything?
>>>>
>>>> Thank you,
>>>>
>>>> -Melissa H.
>>>> ________________________________________
>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Niels Bergsland <[log in to unmask]>
>>>> Sent: Wednesday, February 24, 2016 1:05 PM
>>>> To: [log in to unmask]
>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>
>>>> Hi Melissa,
>>>> The first one is correct - it doesn't actually produce any output - it
>>>> just sets up some shell variables.
>>>>
>>>> The second gives you that error because it is looking for
>>>> "usr/share/fsl/5.0/etc/fslconf/fsl.sh" under your current working
>>>> directory.
>>>>
>>>> Best,
>>>> Niels
>>>>
>>>> On Wed, Feb 24, 2016 at 6:59 PM, Hidalgo, Melissa <[log in to unmask]> wrote:
>>>>> Hello Niels,
>>>>>
>>>>>
>>>>> If I put a space it doesnt do anything, the next line that appears is
>>>>> (blank): @detective:~$
>>>>>
>>>>> But, if I dont put the space then I get:
>>>>>
>>>>> bash: ./usr/share/fsl/5.0/etc/fslconf/fsl.sh: No such file or directory
>>>>>
>>>>>
>>>>> -Melissa H.
>>>>>
>>>>>
>>>>> ________________________________
>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Niels
>>>>> Bergsland <[log in to unmask]>
>>>>> Sent: Wednesday, February 24, 2016 12:53 PM
>>>>>
>>>>> To: [log in to unmask]
>>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>>
>>>>> Hi - Your command is missing a space.
>>>>>
>>>>> It should be:
>>>>>
>>>>> . $FSLDIR/etc/fslconf/fsl.sh
>>>>>
>>>>> not
>>>>>
>>>>> .$FSLDIR/etc/fslconf/fsl.sh
>>>>>
>>>>> Best,
>>>>> Niels
>>>>>
>>>>> On Wed, Feb 24, 2016 at 6:49 PM, Hidalgo, Melissa <[log in to unmask]> wrote:
>>>>>>
>>>>>> Hello Matthew,
>>>>>>
>>>>>>
>>>>>> So I'm putting this:
>>>>>>
>>>>>> .$FSLDIR/etc/fslconf/fsl.sh
>>>>>>
>>>>>>
>>>>>> and I'm getting this:
>>>>>>
>>>>>> bash: ./usr/share/fsl/5.0/etc/fslconf/fsl.sh: No such file or directory
>>>>>>
>>>>>>
>>>>>> Even though the fsl.sh file is there. Is there a way to change the script
>>>>>> of the fsl.sh file so that it reads the other fsl binary files?
>>>>>>
>>>>>>
>>>>>> Thank you so much,
>>>>>>
>>>>>>
>>>>>> -Melissa H.
>>>>>>
>>>>>>
>>>>>> ________________________________
>>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>>> Matthew Webster <[log in to unmask]>
>>>>>> Sent: Wednesday, February 24, 2016 10:54 AM
>>>>>>
>>>>>> To: [log in to unmask]
>>>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>>>
>>>>>> Hello Melissa,
>>>>>> If the GUI commands are running, then this is probably a configuration
>>>>>> problem with your shell - what is the result of sourcing the fsl config file
>>>>>> - e.g.
>>>>>>
>>>>>> . $FSLDIR/etc/fslconf/fsl.sh
>>>>>>
>>>>>> ( please note the '. ' in the command above is important ).
>>>>>>
>>>>>> Kind Regards
>>>>>> Matthew
>>>>>>
>>>>>> Hello Matthew,
>>>>>>
>>>>>> Unfortunately I dont have local administration rights, and the only
>>>>>> platform that I have available is Windows, in which I'm trying to run the
>>>>>> scripts through VNC viewer. The group melodic ICA analysis went fine
>>>>>> (through the GUI), and I got all my results. But it is the dual regression
>>>>>> script that is not working properly (on the terminal). Do you know of any
>>>>>> other way of doing dual regression (another script or through any GUI). I
>>>>>> was thinking I may have to change the FLSDIR to something else, because it
>>>>>> is not reading the file that is in there.
>>>>>>
>>>>>> Any help would be greatly appreciated,
>>>>>>
>>>>>> -Melissa H.
>>>>>>
>>>>>> ________________________________
>>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>>> Matthew Webster <[log in to unmask]>
>>>>>> Sent: Tuesday, February 23, 2016 11:59 AM
>>>>>> To: [log in to unmask]
>>>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>>>
>>>>>> Hello Melissa,
>>>>>> Unfortunately we have no experience with running the current version of
>>>>>> FSL under cygwin, as it is no longer a supported platform. Our recommended
>>>>>> solution for FSL on Windows is to use a virtual machine: please see
>>>>>>
>>>>>> https://urldefense.proofpoint.com/v2/url?u=http-3A__fsl.fmrib.ox.ac.uk_fsl_fslwiki_FslInstallation_Windows&d=CwIBaQ&c=y2w-uYmhgFWijp_IQN0DhA&r=eUjRRWNzkwKGsO9WQsZhZQ&m=OerBNu7fh-ZXtuuSh6bzx2pBpNkT8EZIv1F5wuuOU5w&s=KCZMbPnarD6DoJob0ZVqPayJpg3eCPMDFEDAAjBO3I4&e=
>>>>>>
>>>>>> for the full instructions on how to download and configure a VM for your
>>>>>> system.
>>>>>>
>>>>>> Files like fslmaths, fsl_glm are binary data rather than scripts, and no
>>>>>> attempt should be made to modify them with a text editor, as this could
>>>>>> easily corrupt the files.
>>>>>>
>>>>>> Hope this helps,
>>>>>> Kind Regards
>>>>>> Matthew
>>>>>>
>>>>>> Hello Matthew,
>>>>>>
>>>>>> I just use cigwin as the portal to get connected into VNC viewer and then
>>>>>> run the scripts on the terminals. No, this FSL version was downloaded and
>>>>>> adapted to this computer. Is there a way to reload these files, for
>>>>>> instance: flsmaths, fslnvolvs, fsl_glm, fslmerge, fslsplit? When I tried to
>>>>>> open one of these files through gedit, I get a message:
>>>>>>
>>>>>> "The file you opened has some invalid characters. If you continue editing
>>>>>> this file you could corrupt this document. You can also choose another
>>>>>> character encoding and try again."
>>>>>>
>>>>>> Is there a way to edit this files the right way in order to try to run
>>>>>> dual_regression again?
>>>>>>
>>>>>> Thank you,
>>>>>>
>>>>>> -Melissa H.
>>>>>>
>>>>>> ________________________________
>>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>>> Matthew Webster <[log in to unmask]>
>>>>>> Sent: Tuesday, February 23, 2016 11:07 AM
>>>>>> To: [log in to unmask]
>>>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>>>
>>>>>> Hello Melissa,
>>>>>> We haven't actively supported cygwin builds for some time - was this
>>>>>> version of FSL built from source locally?
>>>>>> Kind Regards
>>>>>> Matthew
>>>>>>
>>>>>> Good morning,
>>>>>>
>>>>>> 1) It is running on windows, within cigwin.
>>>>>> 2) When I try to run fslmaths, it says: fslmaths: error while loading
>>>>>> shared libraries: libnewimage.so: cannot open shared object file: No such
>>>>>> file or directory.
>>>>>>
>>>>>> The same type of error message I was getting before, what could this be?
>>>>>>
>>>>>> Thank you,
>>>>>>
>>>>>> -Melissa H.
>>>>>>
>>>>>> ________________________________
>>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>>> Matthew Webster <[log in to unmask]>
>>>>>> Sent: Tuesday, February 23, 2016 6:27 AM
>>>>>> To: [log in to unmask]
>>>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>>>
>>>>>> Hello,
>>>>>> The script is failing at the first call to an FSL binary, can you
>>>>>> confirm:
>>>>>>
>>>>>> 1) What platform is the VNC session running on ( the server, not the local
>>>>>> machine )
>>>>>> 2) On a terminal in the command line session, what happens if you run the
>>>>>> FSL command "fslmaths" with no other command line arguements?
>>>>>>
>>>>>> Kind Regards
>>>>>> Matthew
>>>>>>
>>>>>> Good afternoon,
>>>>>>
>>>>>> I'm running this through VNC viewer in a local machine. I believe there
>>>>>> might be something else I need to put on the script, since it is running
>>>>>> halfway, and then it just stops, I'm using FSL 5.0.
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> -Melissa H.
>>>>>>
>>>>>>
>>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>>> Mark Jenkinson <[log in to unmask]>
>>>>>> Sent: Monday, February 22, 2016 12:55 PM
>>>>>> To: [log in to unmask]
>>>>>> Subject: Re: [FSL] fslnvols error from dual regression
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> What system are you running this on and how did you install FSL?
>>>>>> It looks like they are incompatible, based on the first error: error while
>>>>>> loading shared libraries: libnewimage.so: cannot open shared object file: No
>>>>>> such file or directory.
>>>>>>
>>>>>> All the best,
>>>>>> Mark
>>>>>>
>>>>>> From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of
>>>>>> "Melissa H." <[log in to unmask]>
>>>>>> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
>>>>>> Date: Monday, 22 February 2016 17:25
>>>>>> To: "[log in to unmask]" <[log in to unmask]>
>>>>>> Subject: [FSL] fslnvols error from dual regression
>>>>>>
>>>>>> Good afternoon,
>>>>>>
>>>>>> I'm new to this program and I was trying to do group melodic ICA analysis.
>>>>>> I got my melodic_IC output. However, when I tried to do dual regression, I
>>>>>> got an error message:
>>>>>>
>>>>>> creating common mask
>>>>>> doing the dual regressions
>>>>>> sorting maps and running randomize
>>>>>> /usr/share/fsl/5.0/bin/fslnvols: error while loading shared libraries:
>>>>>> libnewimage.so: cannot open shared object file: No such file or directory.
>>>>>> /usr/lib/fsl/5.0/dual_regression: 171: [: -lt: argument expected
>>>>>> wc: temp.dr.4/scripts+logs/drD: No such file or directory
>>>>>>
>>>>>>
>>>>>> I believe the error is in the fslnvolvs, since it seems to be stopping at
>>>>>> that point. However, when I tried to open the files from scripts+logs, I
>>>>>> also get error messages within the files:
>>>>>>
>>>>>> drA: 2: Syntax error: EOF in backquote substitution
>>>>>>
>>>>>> drB: /usr/share/fsl/5.0/bin/fslmerge: error while loading shared
>>>>>> libraries: libnewimage.so: cannot open shared object file: No such file or
>>>>>> directory
>>>>>> /usr/share/fsl/5.0/bin/fslmaths: error while loading shared libraries:
>>>>>> libnewimage.so: cannot open shared object file: No such file or directory
>>>>>>
>>>>>> Any help would be greatly appreciated!
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Melissa H.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Niels Bergsland
>>>>> Integration Director
>>>>> Buffalo Neuroimaging Analysis Center
>>>>> 100 High St. Buffalo NY 14203
>>>>> [log in to unmask]
>>>>
>>>>
>>>>
>>>> --
>>>> Niels Bergsland
>>>> Integration Director
>>>> Buffalo Neuroimaging Analysis Center
>>>> 100 High St. Buffalo NY 14203
>>>> [log in to unmask]
>>>
>>>
>>>
>>> --
>>> Niels Bergsland
>>> Integration Director
>>> Buffalo Neuroimaging Analysis Center
>>> 100 High St. Buffalo NY 14203
>>> [log in to unmask]
|