Error in "native" EPI space processing (--output-spaces func)

Hello stars,

I’ve been trying to run fmriprep in the “native” EPI space. My understanding is that since version 1.4 this is available (# fMRIPrep: preprocessed data in native BOLD space) using the “–output-spaces func” argument.

I updated my docker version from 1.3 to 1.4 and tried:

fmriprep-docker $Fmri_preproc_in_folder $Fmri_preproc_out_folder participant --participant-label $Subj --fs-license-file $Fs_lic_file -vvv -w $work_dir --nthreads 2 --omp-nthreads 4 --mem-mb 14000 --low-mem --output-spaces func

However theres an error I can’t work out. I do not see anything of help in the workdir either.

The output from fmriprep is:

fmriprep-docker /Volumes/nfs/drorcohen/data/fmripreproc_test_bidsdata/ /Volumes/nfs/drorcohen/data/fmripreproc_test_output/ participant --participant-label sub-002 --fs-license-file /Volumes/nfs/drorcohen/data/fmripreproc_scripts/license.txt -vvv -w /Users/drorcohen/Documents/fmriprep_work_dir/ --nthreads 2 --omp-nthreads 4 --mem-mb 14000 --low-mem --output-spaces func
RUNNING: docker run --rm -it -e DOCKER_VERSION_8395080871=18.09.2 -v /Volumes/nfs/drorcohen/data/fmripreproc_scripts/license.txt:/opt/freesurfer/license.txt:ro -v /Volumes/nfs/drorcohen/data/fmripreproc_test_bidsdata:/data:ro -v /Volumes/nfs/drorcohen/data/fmripreproc_test_output:/out -v /Users/drorcohen/Documents/fmriprep_work_dir:/scratch poldracklab/fmriprep:1.4.1 /data /out participant --participant-label sub-002 -vvv --nthreads 2 --omp-nthreads 4 --mem-mb 14000 --low-mem --output-spaces func -w /scratch
Making sure the input data is BIDS compliant (warnings can be ignored in most cases).
1: [WARN] The recommended file /README is missing. See Section 03 (Modality agnostic files) of the BIDS specification. (code: 101 - README_FILE_MISSING)

Please visit for existing conversations about this issue.

    Summary:                 Available Tasks:        Available Modalities: 
    69 Files, 12.65GB        PView                   T1w                   
    3 - Subjects                                     bold                  
    3 - Sessions                                                           

If you have any questions, please post on

Although --fs-no-reconall was not set (i.e., FreeSurfer is to be run), no FreeSurfer output space (valid values are: fsaverage5, fsaverage, fsaverage6, fsnative) was selected. Adding default “fsaverage5” to the list of output spaces.
190717-03:00:24,353 nipype.workflow WARNING:
Per-process threads (–omp-nthreads=4) exceed total threads (–nthreads/–n_cpus=2)
190717-03:00:24,357 nipype.workflow IMPORTANT:

Running fMRIPREP version 1.4.1:
  * BIDS dataset path: /data.
  * Participant list: ['002'].
  * Run identifier: 20190717-030024_1ca7ecb2-88bc-41eb-9630-ee08975544eb.

190717-03:00:25,579 nipype.workflow IMPORTANT:
Creating bold processing workflow for “/data/sub-002/ses-1/func/sub-002_ses-1_task-PView_acq-TR1000_run-1_bold.nii.gz” (0.42 GB / 630 TRs). Memory resampled/largemem=1.67/4.29 GB.
190717-03:00:25,579 nipype.workflow IMPORTANT:
No single-band-reference found for sub-002_ses-1_task-PView_acq-TR1000_run-1_bold.nii.gz
190717-03:00:27,155 nipype.workflow IMPORTANT:
Slice-timing correction will be included.
Process Process-2:
Traceback (most recent call last):
File “/usr/local/miniconda/lib/python3.7/multiprocessing/”, line 297, in _bootstrap
File “/usr/local/miniconda/lib/python3.7/multiprocessing/”, line 99, in run
self._target(*self._args, **self._kwargs)
File “/usr/local/miniconda/lib/python3.7/site-packages/fmriprep/cli/”, line 610, in build_workflow
File “/usr/local/miniconda/lib/python3.7/site-packages/fmriprep/workflows/”, line 259, in init_fmriprep_wf
File “/usr/local/miniconda/lib/python3.7/site-packages/fmriprep/workflows/”, line 609, in init_single_subject_wf
File “/usr/local/miniconda/lib/python3.7/site-packages/fmriprep/workflows/bold/”, line 551, in init_func_preproc_wf
if ‘MNI152NLin2009cAsym’ in volume_std_spaces else next(iter(volume_std_spaces))
fMRIPrep: Please report errors to

Any help appreciated.


I have managed a workaround.

The issue is solved by adding MNI152NLin2009cAsym as one of the output spaces. Seems closely related to this, which is on the 1.4.2 milestone, so this may get fixed in the next version

That is correct, alternatively you could use the pre-release 1.5.0rc1, although as a release candidate it can be not stable enough.

1 Like