Hi,

The *_L_Hipp_first.bvars matches all files in the directory that
end with L_Hipp_first.bvars, so if there are more than 3 files
like this then this is not the command you should be running.

For a simple test I recommend doing 5 (three might be too
few to be reliable) and writing out the names by hand.  That is:
  concat_bvars L_Hipp_all.bvars name1_L_Hipp_first.bvars name2_L_Hipp_first.bvars name3_L_Hipp_first.bvars name4_L_Hipp_first.bvars name5_L_Hipp_first.bvars 
where name1, name2, etc should be changed to match your
specific files.

If this creates a file called L_Hipp_all.bvars (remove the old
one before running the concat_bvars command so that
you are sure) then it should be fine.

Before running the first_utils command, make sure that
you create a design matrix with 5 rows and, just to be sure,
make sure that the second column has a zero mean.

Also, do you have a directory called shape_analysis inside
the directory you are running this command from?  If not
that will be a problem.

If it still doesn't work, please send us the error message
that you see and the result of "ls" in the directory.

All the best,
	Mark


On 31 Jan 2011, at 20:16, Charlotte Bernard wrote:

> Thanks for your answer.
> 
> In fact, I have tried to do the analysis on 3 subjects in order to understand how it works. I have used this line command:
> concat_bvars L_Hipp_all.bvars con*_L_Hipp_first.bvars dis*_L_Hipp_first.bvars
> I have modified it to adapt it to my sample:
> concat_bvars L_Hipp_all.bvars *_L_Hipp_first.bvars
> A L_Hipp_all.bvars has been created but I'm not sure the concatenation worked.
> How can I be sure of that?
> 
> Then I have used my design with 2 EVS: group and age
> EV group      EV age demean
> 1                0.8
> 1                -0.2
> 1                0.3
> 
> and I use 
> first_utils --vertexAnalysis --usebvars -i L_Hipp_all.bvars -d L_Hipp.mat -o shape_analysis/L_Hipp --useReconNative --useRigidAlign -v >& shape_analysis/L_Hipp.txt
> I have a error message which says: 
> cannot open the file
> Can't find the L_Hipp.txt
> 
> Could you tell me the line commands I have to use in order to work on my group? Those available on the FSL MRI ara adapted to comparison group! Can you tell me those who are adapted to study only ONE group?
> 
> Thanks
> Charlotte
> 
> 
> > Date: Mon, 31 Jan 2011 16:49:58 +0000
> > From: [log in to unmask]
> > Subject: Re: [FSL] Problem with FIRST vertex analysis
> > To: [log in to unmask]
> > 
> > Hi,
> > 
> > That is a lot of subjects.
> > How much memory (RAM) is available on your machine?
> > 
> > I would recommend that you try a subset of the data initially
> > to see if you have everything correct (try with 30 subjects).
> > If that works then try increasingly larger sets until you run
> > into the problem. At that point we should be able to figure 
> > out if it is a memory problem and if so, how much you need.
> > 
> > Also, I don't know what your "modifications" are - do you just
> > mean a different design matrix?
> > 
> > All the best,
> > Mark
> > 
> > 
> > On 31 Jan 2011, at 11:20, SUBSCRIBE FSL Anonymous wrote:
> > 
> > > Dear FSL users,
> > > 
> > > I would like to perform a vertex analysis with FIRST. I have 331 subjects. I would to study a correlation between cognitive performances and hippocampal volumes. I have tried to concatenate my bvars file. Anex file is created but I believe it doesn't work. In the secund step of the analysis, I can read a error meaage: cannot open the file... 
> > > I apply the command line with modifications because I only have one group. 
> > > Can somebody help me?
> > > 
> > > Thanks
> > > Charlotte
> > >