help
489 Subscribers 
help > Error following completing setup
Showing 1-7 of 7 posts
Feb 5, 2021 02:02 PM | Pavan Bains - UCL
Error following completing setup
Hi all,
I am using CONN for the first time. I have completed the setup and preprocessing successfully. I clicked 'done' to move to the next step, denoising. After that, the following error popped up:
ERROR DESCRIPTION:
Index in position 2 exceeds array bounds.
Error in conn (line 5874)
CONN_h.menus.m_preproc.y.data=[CONN_h.menus.m_preproc.y.data(:,CONN_gui.refs.surf.default2reduced) tempdata(:,CONN_gui.refs.surf.default2reduced)];
Error in conn (line 5747)
if ~conn_projectmanager('ispending')&&~ispending, conn gui_preproc;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.3Gb available
I am unsure of how I can proceed. Help would be greatly appreciated.
Thanks,
Pavan
I am using CONN for the first time. I have completed the setup and preprocessing successfully. I clicked 'done' to move to the next step, denoising. After that, the following error popped up:
ERROR DESCRIPTION:
Index in position 2 exceeds array bounds.
Error in conn (line 5874)
CONN_h.menus.m_preproc.y.data=[CONN_h.menus.m_preproc.y.data(:,CONN_gui.refs.surf.default2reduced) tempdata(:,CONN_gui.refs.surf.default2reduced)];
Error in conn (line 5747)
if ~conn_projectmanager('ispending')&&~ispending, conn gui_preproc;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.3Gb available
I am unsure of how I can proceed. Help would be greatly appreciated.
Thanks,
Pavan
Feb 10, 2021 09:02 AM | Pavan Bains - UCL
RE: Error following completing setup
Some more information to help with solving the problem:
I am using version 20b. The analysis is to be ran in native space.
T1.mgz files from freesurfer were entered into the structural tab, coregistered functional files into the functional tab, native ROIs and native GM + WM + CSF into the ROI tab, and rp_ files as first level covariates. The data were already preprocessed, so only ART-based scrubbing (with conservative settings) was conducted in preprocessing. In options, the analysis space was set as "volume: same as functionals" and the mask as the implicit (subject-specific) mask.
I have tried changing the options in a couple ways but it has made no difference and the same error is appearing.
I'd appreciate any help :)
Pavan
I am using version 20b. The analysis is to be ran in native space.
T1.mgz files from freesurfer were entered into the structural tab, coregistered functional files into the functional tab, native ROIs and native GM + WM + CSF into the ROI tab, and rp_ files as first level covariates. The data were already preprocessed, so only ART-based scrubbing (with conservative settings) was conducted in preprocessing. In options, the analysis space was set as "volume: same as functionals" and the mask as the implicit (subject-specific) mask.
I have tried changing the options in a couple ways but it has made no difference and the same error is appearing.
I'd appreciate any help :)
Pavan
Feb 12, 2021 12:02 PM | Alfonso Nieto-Castanon - Boston University
RE: Error following completing setup
Hi Pavan,
That is curious, the error indicates that (for some reason) CONN is interpreting your data as being 'surface-based' even if it does not match the fsaverage dimensions. Could you please send me your file named /data/DATA_Subject001_Session001.mat (just the .mat file, do not need the .matc files) to see if that helps me better understand what might be the issue here?
Thanks
Alfonso
Originally posted by Pavan Bains:
That is curious, the error indicates that (for some reason) CONN is interpreting your data as being 'surface-based' even if it does not match the fsaverage dimensions. Could you please send me your file named /data/DATA_Subject001_Session001.mat (just the .mat file, do not need the .matc files) to see if that helps me better understand what might be the issue here?
Thanks
Alfonso
Originally posted by Pavan Bains:
Hi all,
I am using CONN for the first time. I have completed the setup and preprocessing successfully. I clicked 'done' to move to the next step, denoising. After that, the following error popped up:
ERROR DESCRIPTION:
Index in position 2 exceeds array bounds.
Error in conn (line 5874)
CONN_h.menus.m_preproc.y.data=[CONN_h.menus.m_preproc.y.data(:,CONN_gui.refs.surf.default2reduced) tempdata(:,CONN_gui.refs.surf.default2reduced)];
Error in conn (line 5747)
if ~conn_projectmanager('ispending')&&~ispending, conn gui_preproc;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.3Gb available
I am unsure of how I can proceed. Help would be greatly appreciated.
Thanks,
Pavan
I am using CONN for the first time. I have completed the setup and preprocessing successfully. I clicked 'done' to move to the next step, denoising. After that, the following error popped up:
ERROR DESCRIPTION:
Index in position 2 exceeds array bounds.
Error in conn (line 5874)
CONN_h.menus.m_preproc.y.data=[CONN_h.menus.m_preproc.y.data(:,CONN_gui.refs.surf.default2reduced) tempdata(:,CONN_gui.refs.surf.default2reduced)];
Error in conn (line 5747)
if ~conn_projectmanager('ispending')&&~ispending, conn gui_preproc;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.3Gb available
I am unsure of how I can proceed. Help would be greatly appreciated.
Thanks,
Pavan
Feb 12, 2021 01:02 PM | Pavan Bains - UCL
RE: Error following completing setup
Hi Alfonso,
Thanks so much for your response. I also suspected this was the case, so I just had a look at the data I put into CONN. When I first ran the preprocessing, I included the "functional resampling of functional data at the location of freesurfer subject-specific structural cortical surface" step. I thought maybe the error was to do with this, so I saved a copy of the analysis and tried again without including that step, but the same error occurred. However, upon inspecting my functional files, I realised that they were changed from .nii to .surf.nii files in the preprocessing step, and any subsequent attempts were using those files. So, I reverted back to the original files and it seemed to work.
I was able to proceed to denoising, however when I tried to denoise I received the following error:
ERROR DESCRIPTION:
Error using conn_process (line 1224)
Empty data (samples=405,voxels=0) in file /Volumes/My Passport for Mac/MSc Project/Pilot/LC_rsfc_PILOT5/data/DATA_Subject001_Session001.mat
Error in conn_process (line 30)
case {'preprocessing_gui','denoising_gui'}, conn_disp(['CONN: RUNNING DENOISING STEP']); conn_process([1.5,2,6:9],varargin{:});
Error in conn (line 6388)
else conn_process('denoising_gui'); ispending=false;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.5Gb available
spm @ /Users/pavanbains/Documents/UCL/Neuroimaging/spm12
conn @ /Users/pavanbains/Documents/UCL/conn
It seems to be an issue with the DATA_Subject001_Session001.mat file, which I have attached. Thanks so much!
Kind Regards,
Pavan
Thanks so much for your response. I also suspected this was the case, so I just had a look at the data I put into CONN. When I first ran the preprocessing, I included the "functional resampling of functional data at the location of freesurfer subject-specific structural cortical surface" step. I thought maybe the error was to do with this, so I saved a copy of the analysis and tried again without including that step, but the same error occurred. However, upon inspecting my functional files, I realised that they were changed from .nii to .surf.nii files in the preprocessing step, and any subsequent attempts were using those files. So, I reverted back to the original files and it seemed to work.
I was able to proceed to denoising, however when I tried to denoise I received the following error:
ERROR DESCRIPTION:
Error using conn_process (line 1224)
Empty data (samples=405,voxels=0) in file /Volumes/My Passport for Mac/MSc Project/Pilot/LC_rsfc_PILOT5/data/DATA_Subject001_Session001.mat
Error in conn_process (line 30)
case {'preprocessing_gui','denoising_gui'}, conn_disp(['CONN: RUNNING DENOISING STEP']); conn_process([1.5,2,6:9],varargin{:});
Error in conn (line 6388)
else conn_process('denoising_gui'); ispending=false;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.5Gb available
spm @ /Users/pavanbains/Documents/UCL/Neuroimaging/spm12
conn @ /Users/pavanbains/Documents/UCL/conn
It seems to be an issue with the DATA_Subject001_Session001.mat file, which I have attached. Thanks so much!
Kind Regards,
Pavan
Feb 12, 2021 07:02 PM | Alfonso Nieto-Castanon - Boston University
RE: Error following completing setup
Hi Pavan,
This seems unrelated to the previous issue, as the "empty data" error indicates that there are no voxels at all in your analysis mask (specified in Setup.Options). If I recall correctly you were using the "implicit-masking" option? If so, the mask is derived using SPM convention (voxels with average BOLD values above 80% of the global signal), so one possible issue that would break this procedure and perhaps result in no supra-threshold voxels at all in your mask would be if your data was already denoised (in particular if the average BOLD signal was zero or negative). Could you perhaps double-check your functional data to see if the range of BOLD signal values looks fine? (e.g. in the Setup.Functional tab, click anywhere in the BOLD image display and hover over the timeseries display below to see the range of BOLD signal values -typically with values in the hundreds/thousand-).
Best
Alfonso
Originally posted by Pavan Bains:
This seems unrelated to the previous issue, as the "empty data" error indicates that there are no voxels at all in your analysis mask (specified in Setup.Options). If I recall correctly you were using the "implicit-masking" option? If so, the mask is derived using SPM convention (voxels with average BOLD values above 80% of the global signal), so one possible issue that would break this procedure and perhaps result in no supra-threshold voxels at all in your mask would be if your data was already denoised (in particular if the average BOLD signal was zero or negative). Could you perhaps double-check your functional data to see if the range of BOLD signal values looks fine? (e.g. in the Setup.Functional tab, click anywhere in the BOLD image display and hover over the timeseries display below to see the range of BOLD signal values -typically with values in the hundreds/thousand-).
Best
Alfonso
Originally posted by Pavan Bains:
Hi Alfonso,
Thanks so much for your response. I also suspected this was the case, so I just had a look at the data I put into CONN. When I first ran the preprocessing, I included the "functional resampling of functional data at the location of freesurfer subject-specific structural cortical surface" step. I thought maybe the error was to do with this, so I saved a copy of the analysis and tried again without including that step, but the same error occurred. However, upon inspecting my functional files, I realised that they were changed from .nii to .surf.nii files in the preprocessing step, and any subsequent attempts were using those files. So, I reverted back to the original files and it seemed to work.
I was able to proceed to denoising, however when I tried to denoise I received the following error:
ERROR DESCRIPTION:
Error using conn_process (line 1224)
Empty data (samples=405,voxels=0) in file /Volumes/My Passport for Mac/MSc Project/Pilot/LC_rsfc_PILOT5/data/DATA_Subject001_Session001.mat
Error in conn_process (line 30)
case {'preprocessing_gui','denoising_gui'}, conn_disp(['CONN: RUNNING DENOISING STEP']); conn_process([1.5,2,6:9],varargin{:});
Error in conn (line 6388)
else conn_process('denoising_gui'); ispending=false;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.5Gb available
spm @ /Users/pavanbains/Documents/UCL/Neuroimaging/spm12
conn @ /Users/pavanbains/Documents/UCL/conn
It seems to be an issue with the DATA_Subject001_Session001.mat file, which I have attached. Thanks so much!
Kind Regards,
Pavan
Thanks so much for your response. I also suspected this was the case, so I just had a look at the data I put into CONN. When I first ran the preprocessing, I included the "functional resampling of functional data at the location of freesurfer subject-specific structural cortical surface" step. I thought maybe the error was to do with this, so I saved a copy of the analysis and tried again without including that step, but the same error occurred. However, upon inspecting my functional files, I realised that they were changed from .nii to .surf.nii files in the preprocessing step, and any subsequent attempts were using those files. So, I reverted back to the original files and it seemed to work.
I was able to proceed to denoising, however when I tried to denoise I received the following error:
ERROR DESCRIPTION:
Error using conn_process (line 1224)
Empty data (samples=405,voxels=0) in file /Volumes/My Passport for Mac/MSc Project/Pilot/LC_rsfc_PILOT5/data/DATA_Subject001_Session001.mat
Error in conn_process (line 30)
case {'preprocessing_gui','denoising_gui'}, conn_disp(['CONN: RUNNING DENOISING STEP']); conn_process([1.5,2,6:9],varargin{:});
Error in conn (line 6388)
else conn_process('denoising_gui'); ispending=false;
Error in conn_menumanager (line 121)
feval(CONN_MM.MENU{n0}.callback{n1}{1},CONN_MM.MENU{n0}.callback{n1}{2:end});
CONN20.b
SPM12 + DAiSS DEM FieldMap MEEGtools
Matlab v.2020b
project: CONN20.b
storage: 3892.5Gb available
spm @ /Users/pavanbains/Documents/UCL/Neuroimaging/spm12
conn @ /Users/pavanbains/Documents/UCL/conn
It seems to be an issue with the DATA_Subject001_Session001.mat file, which I have attached. Thanks so much!
Kind Regards,
Pavan
Feb 13, 2021 12:02 PM | Pavan Bains - UCL
RE: Error following completing setup
Hi Alfonso,
Thanks for your response! I did what you recommended (checked BOLD signal values) and they appeared to be quite low (around 0, ranged from around -3 to 4). I have attached a screenshot for clarity. I checked this for all 6 participants and found a similar trend.
The resting-state data was derived by removing the task-related BOLD signal from the data, leaving task residuals. Could this cause a problem?
Thanks so much for your help.
Kind Regards,
Pavan
Thanks for your response! I did what you recommended (checked BOLD signal values) and they appeared to be quite low (around 0, ranged from around -3 to 4). I have attached a screenshot for clarity. I checked this for all 6 participants and found a similar trend.
The resting-state data was derived by removing the task-related BOLD signal from the data, leaving task residuals. Could this cause a problem?
Thanks so much for your help.
Kind Regards,
Pavan
Feb 13, 2021 05:02 PM | Alfonso Nieto-Castanon - Boston University
RE: Error following completing setup
Hi Pavan,
Yes, that is definitely the issue here. If you are importing residual BOLD signals, very likely with average 0, then the "implicit-masking" option in CONN or SPM is not going to work at all (as that relies solely on the average BOLD signal values to differentiate voxels within vs. outside of the mask) so you simply need to provide an explicit mask instead (e.g. it may be possible to create one directly from your data, as from your plots it appears that outside voxels may already being labeled with 0's?)
Of course another approach is to simply enter the original data (without removing the task-related BOLD signal), and have task-effects removed as usual during the denoising step
Hope this helps
Alfonso
Originally posted by Pavan Bains:
Yes, that is definitely the issue here. If you are importing residual BOLD signals, very likely with average 0, then the "implicit-masking" option in CONN or SPM is not going to work at all (as that relies solely on the average BOLD signal values to differentiate voxels within vs. outside of the mask) so you simply need to provide an explicit mask instead (e.g. it may be possible to create one directly from your data, as from your plots it appears that outside voxels may already being labeled with 0's?)
Of course another approach is to simply enter the original data (without removing the task-related BOLD signal), and have task-effects removed as usual during the denoising step
Hope this helps
Alfonso
Originally posted by Pavan Bains:
Hi Alfonso,
Thanks for your response! I did what you recommended (checked BOLD signal values) and they appeared to be quite low (around 0, ranged from around -3 to 4). I have attached a screenshot for clarity. I checked this for all 6 participants and found a similar trend.
The resting-state data was derived by removing the task-related BOLD signal from the data, leaving task residuals. Could this cause a problem?
Thanks so much for your help.
Kind Regards,
Pavan
Thanks for your response! I did what you recommended (checked BOLD signal values) and they appeared to be quite low (around 0, ranged from around -3 to 4). I have attached a screenshot for clarity. I checked this for all 6 participants and found a similar trend.
The resting-state data was derived by removing the task-related BOLD signal from the data, leaving task residuals. Could this cause a problem?
Thanks so much for your help.
Kind Regards,
Pavan