Hi Mark,
thank you for the response. So the tdof_* files pretty much look like a full brain mask and it's all non-zero numbers (the same across). Now, I do not have a dof file (other than the tdof_*), might that be the problem? So two things: (1) can I generate it from the tdof_t* files; or (2) how should I modify flameo to output the dof file ( don't see an obvious option from the flameo help)?
(Note, I've done everything manually because I needed to vary some parameters systematically, so I don't quite have the folder structure to allow running FSL from the GUI..)
flameo \\
--cope=combined_cope.nii.gz --varcope=combined_varcope.nii.gz --mask=std_mask.nii.gz \\
--dm=second_level.mat --tc=second_level.con --cs=second_level.grp \\
--logdir=flame1_cope_w_varcope_log --runmode=flame1 -v
thank you for the help!
Roman
Date: Fri, 7 Sep 2012 05:59:05 +0100
From:
[log in to unmask]Subject: Re: [FSL] ttologp error: DOF cannot be zero or negative
To:
[log in to unmask]Dear Roman,
Have you looked at the dof values in your file to see if any of them are zero or negative?
This is most likely what is causing the problem with ttologp.
All the best,
Mark
Dear all,
I am trying to run FDR thresholding, following the description on the website, but I keep getting an error on ttologp; namely it complains that
"stdtr domain error
DOF cannot be zero or negative!"
Now, I am using the command (almost) as written on the website (including backquotes for `cat dof`) -- bar for a loop (though I get the same error if I run it one at a time)
> ttologp -logpout logp${i} varcope${i}.nii.gz cope${i}.nii.gz `cat dof`
I am wondering whether this has to do with the fact that I ran all the analysis from the command prompt (using flameo). The folder it created includes the usual 'pe*' 'cope*' tdof_t*' 'tstat*' 'varcope*' 'weights1' 'zflame1lowerstats*' and 'zstat*'
Did I forget to create some file?
thank you for any input you might have,
Cheers
Roman
