help
help > RE: importing existing connectivity matrices in CONN
Aug 22, 2022 08:08 AM | chirokoff
RE: importing existing connectivity matrices in CONN
Hello,
I am currently trying to import connectivity matrices analyzed via an other software as described in the following post. I managed to create the .mtx.nii but when I tried to run the conn_module using :
"conn_module( 'glm' , ...
'data', Files.mtx.nii, ...
'design_matrix', X, ...
'contrast_between', C, ...
'contrast_within', M, ...
'folder', folder);
"
I get the following error message :" Too many input argument".
Using :
"conn_module( 'glm' , ...
'Files.mtx.nii', ...
'X', ...
'C', ...
'M', ...
'folder')
"
Fix the bug but results in another error "Incorrect dimensions (X and C1 should have the same number of columns)" (X and C does have the same numbers of colums).
I managed to run the function anyway using :
"
conn_module_glm(X, Files.mtx.nii; C, M, folder)
"
But then i get :
"
Warning: Matrix is singular to working precision.
> In spm_inv (line 30)
In spm_reml (line 100)
In spm_est_non_sphericity (line 281)
In spm_spm (line 433)
In conn_module_glm (line 151)
Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND =
-1.000000e+00.
> In spm_inv (line 30)
In spm_reml (line 103)
In spm_est_non_sphericity (line 281)
In spm_spm (line 433)
In conn_module_glm (line 151)
Error using spm_dx>expv (line 279)
The requested tolerance is too high.
Error in spm_dx (line 102)
dx = expv(1,J,x);
Error in spm_reml (line 145)
dh = spm_dx(dFdhh,dFdh,{t});
Error in spm_est_non_sphericity (line 281)
[V,h] = spm_reml(Cy,xX.X,xVi.Vi);
Error in spm_spm (line 433)
[xVi, am] = spm_est_non_sphericity(SPM);
Error in conn_module_glm (line 151)
SPM=spm_spm(SPM);
"
Does someone know how to fix this ?
Thank you a lot for your help,
Valentine
Originally posted by Alfonso Nieto-Castanon:
I am currently trying to import connectivity matrices analyzed via an other software as described in the following post. I managed to create the .mtx.nii but when I tried to run the conn_module using :
"conn_module( 'glm' , ...
'data', Files.mtx.nii, ...
'design_matrix', X, ...
'contrast_between', C, ...
'contrast_within', M, ...
'folder', folder);
"
I get the following error message :" Too many input argument".
Using :
"conn_module( 'glm' , ...
'Files.mtx.nii', ...
'X', ...
'C', ...
'M', ...
'folder')
"
Fix the bug but results in another error "Incorrect dimensions (X and C1 should have the same number of columns)" (X and C does have the same numbers of colums).
I managed to run the function anyway using :
"
conn_module_glm(X, Files.mtx.nii; C, M, folder)
"
But then i get :
"
Warning: Matrix is singular to working precision.
> In spm_inv (line 30)
In spm_reml (line 100)
In spm_est_non_sphericity (line 281)
In spm_spm (line 433)
In conn_module_glm (line 151)
Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND =
-1.000000e+00.
> In spm_inv (line 30)
In spm_reml (line 103)
In spm_est_non_sphericity (line 281)
In spm_spm (line 433)
In conn_module_glm (line 151)
Error using spm_dx>expv (line 279)
The requested tolerance is too high.
Error in spm_dx (line 102)
dx = expv(1,J,x);
Error in spm_reml (line 145)
dh = spm_dx(dFdhh,dFdh,{t});
Error in spm_est_non_sphericity (line 281)
[V,h] = spm_reml(Cy,xX.X,xVi.Vi);
Error in spm_spm (line 433)
[xVi, am] = spm_est_non_sphericity(SPM);
Error in conn_module_glm (line 151)
SPM=spm_spm(SPM);
"
Does someone know how to fix this ?
Thank you a lot for your help,
Valentine
Originally posted by Alfonso Nieto-Castanon:
Hi
Jaron,
Yes, that is perfectly fine, to run second-level analyses of arbitrary ROI-to-ROI connectivity matrices you may:
1) convert first your matrix data to NIFTI format, for example using something like:
conn_mtx_write('/folder/mydatafile.mtx.nii', Z);
where Z is your ROI-by-ROI-by-subjects 3D matrix containing the connectivity values for each subject. You may also include the ROI labels as well as the ROI spatial locations info to the conn_mtx_write command if you wish, see "help conn_mtx_write" for additional details
2) run your desired group-level analyses, for example using something like:
conn_module( 'glm' , ...
'data', '/folder/mydatafile.mtx.nii', ...
'design_matrix', X, ...
'contrast_between', C, ...
'contrast_within', M, ...
'folder', '/folder/myresults' );
where X is your design matrix, and C/M are the between- and within- subjects contrast, respectively (see https://web.conn-toolbox.org/fmri-method... for details)
and 3) after that, you may display the second-level results using the standard CONN gui options at any time using the syntax:
conn_display /folder/myresults
Hope this helps
Alfonso
Originally posted by Jaron Kupers:
Yes, that is perfectly fine, to run second-level analyses of arbitrary ROI-to-ROI connectivity matrices you may:
1) convert first your matrix data to NIFTI format, for example using something like:
conn_mtx_write('/folder/mydatafile.mtx.nii', Z);
where Z is your ROI-by-ROI-by-subjects 3D matrix containing the connectivity values for each subject. You may also include the ROI labels as well as the ROI spatial locations info to the conn_mtx_write command if you wish, see "help conn_mtx_write" for additional details
2) run your desired group-level analyses, for example using something like:
conn_module( 'glm' , ...
'data', '/folder/mydatafile.mtx.nii', ...
'design_matrix', X, ...
'contrast_between', C, ...
'contrast_within', M, ...
'folder', '/folder/myresults' );
where X is your design matrix, and C/M are the between- and within- subjects contrast, respectively (see https://web.conn-toolbox.org/fmri-method... for details)
and 3) after that, you may display the second-level results using the standard CONN gui options at any time using the syntax:
conn_display /folder/myresults
Hope this helps
Alfonso
Originally posted by Jaron Kupers:
Hello,
We are new to CONN. We would like to use CONN for the visualization of already analyzed resting-state fMRI results. We have already processed rs-fmri data, and constructed matrices of ROI connectivity and used in NBS. We are happy with the results, but not with the visualization tools offered in NBS. We were wondering therefore whether it Is possible to import only the connectivity matrices together with the mni labels and coordinates (in text files), hence without being forced to redo the preprocessing steps in CONN?
Thanks.
jaron
We are new to CONN. We would like to use CONN for the visualization of already analyzed resting-state fMRI results. We have already processed rs-fmri data, and constructed matrices of ROI connectivity and used in NBS. We are happy with the results, but not with the visualization tools offered in NBS. We were wondering therefore whether it Is possible to import only the connectivity matrices together with the mni labels and coordinates (in text files), hence without being forced to redo the preprocessing steps in CONN?
Thanks.
jaron
Threaded View
| Title | Author | Date |
|---|---|---|
| Jaron Kupers | Jan 7, 2021 | |
| Alfonso Nieto-Castanon | Jan 7, 2021 | |
| chirokoff | Aug 22, 2022 | |
| Alfonso Nieto-Castanon | Aug 29, 2022 | |
| chirokoff | Aug 31, 2022 | |
| Alfonso Nieto-Castanon | Aug 31, 2022 | |
| Samantha Baldi | Jul 14, 2022 | |
| Alfonso Nieto-Castanon | Jul 19, 2022 | |
| Jaron Kupers | Jan 8, 2021 | |
| Alfonso Nieto-Castanon | Jan 15, 2021 | |
| Jaron Kupers | Jan 21, 2021 | |
| Alfonso Nieto-Castanon | Jan 11, 2021 | |
| Jaron Kupers | Jan 14, 2021 | |