Hi,
All doubly-hyphenated options in FSL use the syntax:
--optionname=value
as opposed to single hyphenated options which use the syntax:
-o value
So just replace the space with an equals sign in your last example.
All the best,
Mark
On 8 Sep 2011, at 13:01, Boris Bernhardt wrote:
> Hi Mark
>
> thank you very much for your fast response.
>
> Okay, to follow your advice I concatenated the bvars of my subjects (myTest.bvars) and generated a dummy design matrix of the same n (myTest_design.mat).
>
> Running the command
> first_utils --usebvars --vertexAnalysis -d myTest_design.mat -useReconMNI -i myTest.bvars -o myTestOut.vtk
> indeed does a mvglm at each vertex and saves a vtk file with statistical outputs which I can inspect in fslview.
>
> However, when I add the --saveVertices option,
> first_utils --usebvars --vertexAnalysis -d myTest_design.mat -useReconMNI -i myTest.bvars -o myTestOut.vtk --saveVertices
> or
> first_utils --usebvars --vertexAnalysis -d myTest_design.mat -useReconMNI -i myTest.bvars -o myTestOut.vtk --saveVertices myTestOutOut
>
> my shell prints out the first_utils help and tells me:
> --saveVertices: missing non-optional argument!
>
> What did I do wrong?
>
> Thank you very much,
> Boris
>
>
> On 2011-09-08, at 12:50 PM, Mark Jenkinson wrote:
>
>> Dear Boris,
>>
>> Run the vertex analysis command as you would normally do (i.e. first_utils ...)
>> but add the option --saveVertices. This saves a matrix of the vertex coordinates
>> that go into the statistical analysis. I think this is what you want.
>>
>> All the best,
>> Mark
>>
>>
>> On 8 Sep 2011, at 11:28, Boris Bernhardt wrote:
>>
>>> Hi,
>>>
>>> having processed a T1-weighted dataset using your nice function run_first_all, I would now like to perform vertex-based analysis of several subcortical structures in a different software (e.g. Matlab).
>>>
>>> I have understood that the .vtk files of each structure contain the corresponding mesh coordinates in native space, and that there is already a point-correspondence across subjects. Going though previous posts (see e.g. http://tinyurl.com/3krawon), however, I have read that it may not be optimal to directly load and analyze the coordinates from the vtks (or linearly registered versions of it in MNI space) into Matlab, and then run my stats.
>>>
>>> Is there thus a recommended way to post-process and convert the FIRST .bvar (or .vtk) outputs into ascii files containing suitable vertex coordinates for each subject that follow either the --useReconNative and --useRigidAlign option from first_utils, or alternatively the --useReconMNI flag.
>>>
>>> I hope my question makes some sense.
>>> Many thanks for your help,
>>> Boris
>>>
>>>
>>> ---
>>> Boris Bernhardt
>>> Max Planck Institute for Human Cognitive and Brain Sciences
>>> Department of Social Neuroscience
>>> Stephanstr. 1a, 04103 Leipzig, Germany
>>>
>>> p: +(49) 341 9940 2658
>>> e: [log in to unmask]
>>> http://www.cbs.mpg.de/~bernhardt
>>>
>>>
>>>
>>>
>>>
>>>
>
> ---
> Boris Bernhardt
> Max Planck Institute for Human Cognitive and Brain Sciences
> Department of Social Neuroscience
> Stephanstr. 1a, 04103 Leipzig, Germany
>
> p: +(49) 341 9940 2658
> e: [log in to unmask]
> http://www.cbs.mpg.de/~bernhardt
>
|