Print

Print
Hello Carole,
                          It is unusual to get stage3 results without drB completing as drB generates the mask image used by subsequent steps. I did notice from your logs that the owner of your output files is "root" rather than a more conventional username. Is this a deliberate choice? Do you have a mask.nii.gz image in your output directory?

Kind Regards

Matthew

Hi Matthew


Thanks very much for your help. I resolve a part of my problem it was an extension problem of my input image i forgot one point in my extension for tem16 an tem7.
so for my results, I obtained dr_stage3 for all my component but for the drB I have the same problem it can open the maskAll. So have you an idea for this problem

Best Regards

Carole
De : Matthew Webster <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Vendredi 22 mars 2013 16h46
Objet : Re: [FSL] [FSL] dual regression

Hi Carole,
                  Thank you for your upload, the first thing I noticed is that two of your inputs are failing in the first step:

more drA.e77941.77
Cannot open volume dualregressioninput/tem16 for reading!
more drA.e77941.69
Cannot open volume dualregressioninput/tem7 for reading!

It's probably a good idea to sort out these failures before looking at the later stage errors..

Kind Regards

Matthew


Hi Matthew
I send to you by the server (its oK for little size file) ls -al of my input and my ouput and scrips+logs
also a tar-file of scripts+logs
Thank for your help
Best Regards
Carole

De : Matthew Webster <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Vendredi 22 mars 2013 15h32
Objet : Re: [FSL] [FSL] dual regression

Hi Carole,
                  In that case can you try sending me directly a tar-file of just the log files, as well as the results of running 

ls -al

on both your input and output directories?

Kind Regards

Matthew
Hello Matthew

It is impossible to upload my file.I try on several computer and nothing upload!

two files 800M0 and 300Mo or one 1.23G

Best regard

Carole


De : Matthew Webster <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Mercredi 20 mars 2013 16h59
Objet : Re: [FSL] Aw: Re: [FSL] dual regression

Hello,
           can you tar up your input and output directories ( including all log files ) and upload it to:



we will then try to diagnose what is failing.

Kind Regards

Matthew
my dual_regression script is located user/local/bin

is executable rwxrwxr

and i calling it like this :
dual_regression groupmelodic.ica/melodic_IC.nii.gz 1 design.mat design.con 5000 dualregressionoutput2 dualregressioninput/sub3.nii.gz dualregressioninput/sub5.nii.gz dualregressioninput/sub6.nii.gz dualregressioninput/sub10.nii.gz dualregressioninput/sub11.nii.gz dualregressioninput/sub13.nii.gz dualregressioninput/sub16.nii.gz dualregressioninput/sub19.nii.gz dualregressioninput/sub20.nii.gz dualregressioninput/sub21.nii.gz dualregressioninput/sub26.nii.gz dualregressioninput/sub29.nii.gz dualregressioninput/sub31.nii.gz dualregressioninput/sub38.nii.gz dualregressioninput/sub43.nii.gz dualregressioninput/sub44.nii.gz dualregressioninput/sub47.nii.gz dualregressioninput/tem1.nii.gz dualregressioninput/tem2.nii.gz dualregressioninput/tem3.nii.gz dualregressioninput/tem4.nii.gz dualregressioninput/tem5.nii.gz dualregressioninput/tem6.nii.gz dualregressioninput/tem7.nii.gz dualregressioninput/tem8.nii.gz dualregressioninput/tem9.nii.gz dualregressioninput/tem10.nii.gz dualregressioninput/tem11.nii.gz dualregressioninput/tem12.nii.gz dualregressioninput/tem13.nii.gz dualregressioninput/tem14.nii.gz dualregressioninput/tem16.nii.gz dualregressioninput/tem17.nii.gz dualregressioninput/tem18.nii.gz dualregressioninput/tem19.nii.gz dualregressioninput/tem20.nii.gz dualregressioninput/tem21.nii.gz dualregressioninput/tem22.nii.gz dualregressioninput/tem23.nii.gz dualregressioninput/tem25.nii.gz dualregressioninput/tem26.nii.gz dualregressioninput/tem27.nii.gz dualregressioninput/tem28.nii.gz dualregressioninput/tem29.nii.gz dualregressioninput/tem30.nii.gz

thanks
De : Rosalia Dacosta Aguayo <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Mercredi 20 mars 2013 16h27
Objet : Re: [FSL] Aw: Re: [FSL] dual regression

Hi,
1. Where do you have the dual regression script?
2. Have you you make it executable? chmod a+x dual regression?
3. How are you calling it? when you are at the directory you should do ./dual regression melodic.nii.gz 1 design.mat design.con 5000 results 'cat .filelist'
The signs that go with cat .filelist should be inclinated to the right and there ir a space between cat and .filelist
Rosalia.
El 20/03/2013 16:18, "carole menuel" <[log in to unmask]> escribió:
I have just updating my FSL at 5.02. I just running dual_regression and Ihave always the same problems



De : Jeanette Mumford <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Mercredi 20 mars 2013 15h37
Objet : Re: [FSL] Aw: Re: [FSL] dual regression

I've looked on the list and it seems this problem has come up before and was possibly a python issue?  Try updating to the most recent version of FSL and see if that helps.  There were 2-3 patches after 5.0.0 was released and FSL runs much more smoothly with the latest version in my experience.

Cheers,
Jeanette

On Wed, Mar 20, 2013 at 8:52 AM, carole menuel <[log in to unmask]> wrote:
Hi

Thanks for your response
i am running dual_regression where my directory is located Documents/dualregressionextreme

dual_regression groupmelodic.ica/melodic_IC.nii.gz 1 design.mat design.con 5000 dualregressionoutput2 dualregressioninput/sub3.nii.gz dualregressioninput/sub5.nii.gz dualregressioninput/sub6.nii.gz dualregressioninput/sub10.nii.gz dualregressioninput/sub11.nii.gz dualregressioninput/sub13.nii.gz dualregressioninput/sub16.nii.gz dualregressioninput/sub19.nii.gz dualregressioninput/sub20.nii.gz dualregressioninput/sub21.nii.gz dualregressioninput/sub26.nii.gz dualregressioninput/sub29.nii.gz dualregressioninput/sub31.nii.gz dualregressioninput/sub38.nii.gz dualregressioninput/sub43.nii.gz dualregressioninput/sub44.nii.gz dualregressioninput/sub47.nii.gz dualregressioninput/tem1.nii.gz dualregressioninput/tem2.nii.gz dualregressioninput/tem3.nii.gz dualregressioninput/tem4.nii.gz dualregressioninput/tem5.nii.gz dualregressioninput/tem6.nii.gz dualregressioninput/tem7.nii.gz dualregressioninput/tem8.nii.gz dualregressioninput/tem9.nii.gz dualregressioninput/tem10.nii.gz dualregressioninput/tem11.nii.gz dualregressioninput/tem12.nii.gz dualregressioninput/tem13.nii.gz dualregressioninput/tem14.nii.gz dualregressioninput/tem16.nii.gz dualregressioninput/tem17.nii.gz dualregressioninput/tem18.nii.gz dualregressioninput/tem19.nii.gz dualregressioninput/tem20.nii.gz dualregressioninput/tem21.nii.gz dualregressioninput/tem22.nii.gz dualregressioninput/tem23.nii.gz dualregressioninput/tem25.nii.gz dualregressioninput/tem26.nii.gz dualregressioninput/tem27.nii.gz dualregressioninput/tem28.nii.gz dualregressioninput/tem29.nii.gz dualregressioninput/tem30.nii.gz

caro
De : Jeanette Mumford <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Mercredi 20 mars 2013 14h34

Objet : Re: [FSL] Aw: Re: [FSL] dual regression

Hi,

Looks like your design.grp has the correct number of lines, so I'm assuming the design.mat does (I think you pasted the wrong file, but that's okay I don't need to see your design.mat).

Does .filelist have the correct number of lines in it?  45.  

Although it may not matter (depending on where you're running the command from), it may help to type out the full path to the files you're calling in dual_regression.

Jeanette


On Wed, Mar 20, 2013 at 8:01 AM, carole menuel <[log in to unmask]> wrote:
FSL version 5.0.0
dual_regression groupICA.gica/groupmelodic.ica/melodic_IC
1 design.mat design.con 5000 output `cat groupICA.gica/.filelist`
thanks
Carole
De : Rosalia Dacosta Aguayo <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Mercredi 20 mars 2013 12h13

Objet : Re: [FSL] Aw: Re: [FSL] dual regression

Hi,
Which FSL version are you using?
Which command line are you using?
Rosalia.
El 20/03/2013 12:07, "carole menuel" <[log in to unmask]> escribió:
dear rosalia
i haven't used another computer my matrix and all my input data are in the same directory so i don't understand the problem of path...


De : Rosalia Dacosta Aguayo <[log in to unmask]>
À : [log in to unmask]
Envoyé le : Mercredi 20 mars 2013 12h04
Objet : Re: [FSL] Aw: Re: [FSL] dual regression

Hi,
I run melodic in one computer. Then, I copied all the output files to another computer to run dual regression. The problem is that the .filelist file had the path from the first computer and, therefore, I was having the same problems for running dual regression script.
Rosalia.
El 20/03/2013 11:58, "Lysiann Kalmbach" <[log in to unmask]> escribió:

Dear Rosalia, thank for your fast answer. What do you mean with you changed to PC- I think everything should be fine there :/
Gesendet: Mittwoch, 20. März 2013 um 11:48 Uhr
Von: "Rosalia Dacosta Aguayo" <[log in to unmask]>
An: [log in to unmask]
Betreff: Re: [FSL] dual regression
Hi,
I had the same error. Be sure that the .filelist has the correct PATH. I changed of pc and I forgot this and I had problems.
Hope this helps.
Rosalia.
El 20/03/2013 11:10, "carole menuel" <[log in to unmask]">[log in to unmask]> escribió:
Hi Jeanette

i send you my design matrix because I don't understand my problem perhaps you can find it
thanks for your help
Carole

design.con
/ContrastName1    group A > group B
/ContrastName2    group B > group A
/ContrastName3    group A mean
/ContrastName4    group B mean
/NumWaves    2
/NumContrasts    4
/PPheights        1.000000e+00    1.000000e+00    1.000000e+00    1.000000e+00
/RequiredEffect        1.287    1.287    1.015    0.791

/Matrix
1.000000e+00 -1.000000e+00
-1.000000e+00 1.000000e+00
1.000000e+00 0.000000e+00
0.000000e+00 1.000000e+00

design.grp

/NumWaves    1
/NumPoints    45

/Matrix
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2

De : Jeanette Mumford <[log in to unmask]">[log in to unmask]>
À : [log in to unmask]">[log in to unmask]
Envoyé le : Mardi 19 mars 2013 16h01
Objet : Re: [FSL] dual regression

Sorry, I missed your second error:
numbers of rows in design matrix doesn't match number of "time point" in input data!

You need to be sure the design matrix you are inputting has exactly the same number of rows as the number of data sets you are inputting into the dual regression script.  Sounds like you have 45 subjects, so be sure your design matrix has 45 rows and you are inputting exactly 45 data sets.  

Cheers,
Jeanette

On Tue, Mar 19, 2013 at 9:46 AM, SUBSCRIBE FSL caro <[log in to unmask]">[log in to unmask]> wrote:
Hi jeanette
Thanks for your response but my software works correctly so i think is not a problem of path because i already check it. Have you an other idea ?
For the design matrix problem have you a suggestion!
Thanks
Caro