Rosalia,
You seem to be strangely resistant to advice - this makes it quite difficult (read impossible) to sort out what probably is quite a trivial matter.
As Mark already suggested, can you please send the output of
ls -al dual_regression
and the output of
pwd
in the directory you're trying to run this from. Maybe there is someone local who knows a bit of linux and who you could ask to help you do this?
Also, when you did the chmod a+x dual_regression command, was there an error output?
hth
Christian
On 2 Jul 2011, at 15:28, Rosalia Dacosta Aguayo wrote:
>
>
> 2011/7/2 Rosalia Dacosta Aguayo <[log in to unmask]>
> Message is bash:dual_regression command not found....
>
> It's always the same...dual_regression script is in the same directory as all the other documents: .IC, .mat, .con,....but the message is allways the same and
> I can't start dual_regression for my 6 components. I have choice 6 components, I want to do dual_regression with every component to see my results but I have been working all the day and I haven't find the problem...incredible.
>
> Please, some helping, some light...
>
> Best regards,
>
> Rosalia.
>
> ---------- Forwarded message ----------
> From: Anderson Winkler <[log in to unmask]>
> Date: 2011/7/2
> Subject: Re: [FSL] Some help with ICA: Please, I am a little desperate
> To: [log in to unmask]
>
>
> Dear Rosalia,
>
> Downloading the script from the FMRIB website and making it executable works for me. See below the outputs I see in the terminal screen.
> Are you following all the descriptions given in the official page? http://www.fmrib.ox.ac.uk/analysis/dualreg
> I'm not sure what else might be going on there...
>
> Anderson
>
>
> ander@volans:~ $ wget http://www.fmrib.ox.ac.uk/analysis/dualreg/dual_regression
> --2011-07-02 14:33:48-- http://www.fmrib.ox.ac.uk/analysis/dualreg/dual_regression
> Resolving www.fmrib.ox.ac.uk... 163.1.84.10
> Connecting to www.fmrib.ox.ac.uk|163.1.84.10|:80... connected.
> HTTP request sent, awaiting response... 200 OK
> Length: 4183 (4.1K) [text/plain]
> Saving to: `dual_regression'
>
> 100%[===============================================================================================================================================>] 4,183 --.-K/s in 0.07s
>
> 2011-07-02 14:33:48 (58.5 KB/s) - `dual_regression' saved [4183/4183]
>
> ander@volans:~ $ chmod a+x dual_regression
> ander@volans:~ $ ./dual_regression
>
> dual_regression v0.5 (beta)
>
> ***NOTE*** ORDER OF COMMAND-LINE ARGUMENTS IS DIFFERENT FROM PREVIOUS VERSION
>
> Usage: dual_regression <group_IC_maps> <des_norm> <design.mat> <design.con> <n_perm> <output_directory> <input1> <input2> <input3> .........
> e.g. dual_regression groupICA.gica/groupmelodic.ica/melodic_IC 1 design.mat design.con 500 grot `cat groupICA.gica/.filelist`
>
> <group_IC_maps_4D> 4D image containing spatial IC maps (melodic_IC) from the whole-group ICA analysis
> <des_norm> 0 or 1 (1 is recommended). Whether to variance-normalise the timecourses used as the stage-2 regressors
> <design.mat> Design matrix for final cross-subject modelling with randomise
> <design.con> Design contrasts for final cross-subject modelling with randomise
> <n_perm> Number of permutations for randomise; set to 1 for just raw tstat output, set to 0 to not run randomise at all.
> <output_directory> This directory will be created to hold all output and logfiles
> <input1> <input2> ... List all subjects' preprocessed, standard-space 4D datasets
>
> <design.mat> <design.con> can be replaced with just
> -1 for group-mean (one-group t-test) modelling.
> If you need to add other randomise option then just edit the line after "EDIT HERE" below
>
>
>
>
>
> On 07/02/2011 02:17 PM, Rosalia Dacosta Aguayo wrote:
>> Dear Anderson,
>>
>> When I write the command ./dual_regression the message is that dual_regression not found. But I have dual_regression script in the same directory...I don't understand nothing. I made executable the script by the command: chmod a+x dual_regression and I think nothing happens....Could you attach me some script for dual_regression, maybe my script is the problem....
>>
>> 2011/7/2 Anderson Winkler <[log in to unmask]>
>> Dear Rosalia,
>>
>> In addition to what Wolf said, in your syntax, where you put `cat .filelist´ it should instead be `cat .filelist` (the difference is the ` symbol when ending the command).
>>
>> If you have difficulty finding this ` symbol in your keyboard, and if your system is fairly recent, you can safely replace this part for $(cat .filelist). The $() does the same as the ` delimiter.
>>
>> Hope this helps!
>>
>> All the best,
>>
>> Anderson
>>
>>
>> On 07/02/2011 07:33 AM, wolf zinke wrote:
>>> Hi
>>>
>>> Just to be sure, all files AND the dual_regression script are in the same directory?
>>>
>>> If you use ./ before a command, it just tells your system that this command is found in the current working directory. If, however, your script is stored in another directory, you need to specify the path to the command, or add this directory to the known paths. For example if the script is saved in /home/aguayo/dualreg, you have to run /home/aguayo/dualreg/dual_regression from the directory, where you have got your data, and everything will be fine.
>>> If you use this script more often, it would be more convenient to include the directory to your known paths, but how to do that depends on the system you are using. I specified a method valid for the BASH shell, if you add this to your ./bash_profile file found in your home directory, this path will be added whenever you log into your system. Otherwise you have to do it manually every time.
>>>
>>> I hope I could clarify it a bit, and sorry that I havent been clear about the meaning of ./' before.
>>>
>>> good luck,
>>> wolf
>>>
>>> On 02/07/11 13:19, Rosalia Dacosta Aguayo wrote:
>>>> Hi,
>>>>
>>>> Thank you for answer my mail. I have all the files in the same directory, but when I open a terminal for executing dual_regression....if I just look for dual_regression the message is that: command don't found (I think the problem is in the programme dual_regression...but I don't know how to resolve it.
>>>>
>>>> If a open the terminal and I just put ./dual_regression, it tells me about the root and the different commands...I do this but, the message is: dual_regression not found (maybe it isn't executable) and that "denegate acces to design.mat and design.com....
>>>>
>>>> I am lost :((
>>>>
>>>>
>>>> 2011/7/2 wolf zinke <[log in to unmask]>
>>>> Hi,
>>>>> Where is the problem?
>>>> Do you still get the same error message? What exactly is the output you get on the screen after issuing this command? Are the other files in the same directory? Otherwise you need to add the path to these files, otherwise the command looks at the wrong places.
>>>>
>>>> If you type just ./dual_regression do you get the help information or an error.
>>>>
>>>> Right now the information provided is a bit sparse, so it is difficult for me to see, where the problem is. Maybe if you are a bit patient, someone else will give a try. But ideally you give information about the error, this makes help a lot easier ;)
>>>>
>>>> good luck,
>>>> wolf
>>>>
>>>>
>>>>>
>>>>> In the same directory where I have dual_regression programme I write the following command: ./dual_regression salience.nii.gz 1 design.mat design.con 5000 analisis_salience `cat .filelist´
>>>>>
>>>>> 2011/7/2 Rosalia Dacosta Aguayo <[log in to unmask]>
>>>>>
>>>>> Thank you!
>>>>>
>>>>> I am going to try it and I will tell you if this works :))
>>>>>
>>>>> Rosalia.
>>>>>
>>>>> 2011/7/2 wolf zinke <[log in to unmask]>
>>>>> Hi,
>>>>>
>>>>> You need to add the path where you saved this function to your known paths, or if you want to call it from the directory, use: ./dual_regression
>>>>>
>>>>>
>>>>> Adding the path:
>>>>>
>>>>> PATH=your_directory_here:${PATH}
>>>>> export PATH
>>>>>
>>>>> I hope this helps,
>>>>> wolf
>>>>>
>>>>>
>>>>>
>>>>> On 02/07/11 12:23, Rosalia Dacosta Aguayo wrote:
>>>>>> Hi Andreas,
>>>>>>
>>>>>> I have problem with a programme called dual_regression. I must download and execute the programme: chmod a+x dual_regression. But when I do this, it's seems it doesn't work well, because when I try to do the dual_regression, the Linux doesn't find it, even calling it in the directory where it is...
>>>>>>
>>>>>> Thank you!
>>>>>>
>>>>>> 2011/7/2 Andreas Bartsch <[log in to unmask]>
>>>>>> Hi,
>>>>>>
>>>>>> if you explain the problems, we might be able to help you.
>>>>>> Cheers,
>>>>>> Andreas
>>>>>> ________________________________________
>>>>>> Von: FSL - FMRIB's Software Library [[log in to unmask]] im Auftrag von Rosalia Dacosta Aguayo [[log in to unmask]]
>>>>>> Gesendet: Samstag, 2. Juli 2011 11:20
>>>>>> An: [log in to unmask]
>>>>>> Betreff: [FSL] Some help with ICA
>>>>>>
>>>>>> Hello,
>>>>>>
>>>>>> Anyone can help me about dual_regression with ICA and the following steps. I have some problems.
>>>>>>
>>>>>> Thank you.
>>>>>>
>>>>>> Rosalia.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>
>>
>
>
>
