alfnie/conn

CONN toolbox, development version (conn-toolbox.org)

conn's Issues

Do GLM routines account for temporal autocorrelation?

I colleague asked me about the linear modelling tools in CONN, and I was trying to figure out if CONN's GLM tools account for temporal autocorrelation in fMRI time series. I studied https://github.com/alfnie/conn/blob/master/conn_glm.m & https://github.com/alfnie/conn/blob/master/conn_glmunivariate.m but as far I can tell these GLMs are always estimated with OLS (and not GLS with whitening for time series autocorrelation).

If this is correct, isn't there a problem with the stderr's, T-values and P-values when these models are fit to first-level fMRI time series data?

Importing field maps from BIDS dataset

Dear Alfonso,

I am using conn21a to create a new project by importing data from BIDS dataset. The dataset has successfully been validated with BIDS validator, and it contains "anat" (T1w scan), "fmap" (magnitude1, magnitude2, and phasediff), and "func" (rest-bold), all files in compressed *nii.gz format.

During import, conn stops and throws the error below.
It seems it happens when dealing with field map files. I tried using the development version and also conn20b with the same data. The import was done "without errors", however the derivatives/fmap ends up with only one file: magnitude1.nii

Do you have any idea why this might be happening?

Looking forward to hearing from you.
Thank you in advance.
Best,
João

ERROR DESCRIPTION:

Error using file2mat
Cant open file.

Error in file_array/subsref>multifile2mat (line 179)
val(cc(i)+1:cc(i+1)) = file2mat(obj,int64(1),int64(x(y==i)));

Error in file_array/subsref>subfun (line 99)
t = multifile2mat(sobj,varargin{:});

Error in file_array/subsref (line 65)
t = subfun(sobj,args{:});

Error in nifti/subsref>rec (line 221)
t = subsref(t,subs(2:end));

Error in nifti/subsref>rec (line 234)
ti = rec(ti,subs(2:end));

Error in nifti/subsref (line 45)
varargout = rec(opt,subs);

Error in spm_file_merge (line 148)
ni.dat(:,:,:,i) = N(i).dat(:,:,:,ind(i,1),ind(i,2));

Error in conn_fileutils (line 386)
else spm_file_merge(varargin{:});

Error in conn_importvol2bids (line 183)
conn_fileutils('spm_file_merge',a,out);

Error in conn_importbids (line 119)
[nill,nill,nV]=conn_importvol2bids(char(fnames),nsub,nses,'fmap','fmap',[],[],[],[],true);

Error in conn (line 6106)
if bids_dofunc, [ok,err]=conn_importbids(info.dataset_select.func.data.file,'type','functional','subjects',nsubs,'subjects_id',bids_subj_id,'nset',nset,'localcopy',localcopy,'copytoderiv',copytoderiv); ERR=[ERR err]; end
CONN21.a
SPM12
Matlab v.2019b
project: CONN21.a
storage: 14108.2Gb available

spm @ /SCRATCH/software/toolboxes/spm12
conn @ /SCRATCH/software/toolboxes/conn21a

Multiple slice timings for multiple files/subjects

Hi,
I am trying to specify multiple slice timings for multiple files within one project. In the documentation of the conn_batch script, it is specified: "note: use cell array for subject-specific vectors".
Here:

conn/conn_batch.m

Line 418 in 47e2980

% Setup.preprocessing.sliceorder : (functional_slicetime) acquisition order (vector of indexes; 1=first slice in image; note: use cell

When I try to specify the batch.Setup.preprocessing.sliceorder value as a cell (e.g. {[1 2 3 4], [1 3 2 4]}}) there is a line in conn_setup_preproc.m that flattens the whole timing data.

conn/conn_setup_preproc.m

Line 303 in 47e2980

if iscell(sliceorder), sliceorder=[sliceorder{:}]; end

When commenting this line, the script allows to manualy specify individual slicetimings perfectly fine.

Also, the way the code seems to interpret the sliceorder cell array later on, seems like each subject's slice timing info will be stored in an array called sliceorder, while all the subjects/files will be stored in an array called sliceorder_all. So I think it doesn't make sense to flatten the cell array as it is done at the referenced section.

fmriprep import issue

Hi,

Just wanted to report a bug when importing an fmriprep dataset. An extra .nii is appended to all file names when copied to the derivatives/conn directory, see the image below (sub names have been deliberately cut off). Functionally, it still works, but is inconvenient, especially because any other files you want to include (such as events) also have to have that same naming convention if conn is to find them automatically.

Best,
Steven

Unable to compute non-parametric statistics

conn_module('glm', 'data', {sub1.mtx.nii, sub2.mtx.nii...}, 'design_matrix', [1 0; 1 0; 0 1; 0 1;...], 'analysistype', 1)

I attempted to use GLM module of CONN to analysis my handcrafted ROI-ROI functional connectivity matrices (246 x 246). However, I found that the toolbox could only perform cluster-level p-FDR correctition (MVPA ominbus test). When I selected other non-parametric method like NBS, a message box popped up and said that unable to compute non-parametric statistics. Are there any ways to solve this problem ?

fmriprep-confound not including "echo"

Hi,
I just wanted to inform you that I had some problems importing the confounds.tsv files from a fmriprep-dataset.
This is a name of an example bold-file:
sub-01_ses-1_task-rest_run-1_echo-1_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
and the corresponding confounds file is:
sub-01_ses-1_task-rest_run-1_desc-confounds_timeseries.tsv

The conn-toolbox couldn't recognize it, because of the missing "_echo-1" in the tsv. After I manually renamed the tsv to:
sub-01_ses-1_task-rest_run-1_echo-1_desc-confounds_timeseries.tsv
everything worked fine!!

Best,
Karl

conn_v2v comments

Hi, I'm trying to port the connectivity measures from the conn-toolbox to Python. I believe conn_v2v could be a good starting point, however the file provides few comments to explain how it works. Would it be possible to shed some light on this and explain the overall logic or suggest how best to analyze it to understand the actual execution flows for the respective metrics and how they map to those listed here? Also, I have the impression that conn_v2v covers mostly the metrics listed as "Network measures (voxel-level)" on the website. Would there be another file to cover the Seed-based Connectivity measures? Thank you in advance for your consideration. Best regards, -- Stanislaw