Header_source trait of a _MergeInputSpec instance fmriprep error

Hi all - we are trying to process the unprocessed HCP data through fmriprep. We have converted a sample subject using hcp2bids. Here is the tree for the sample dataset. As you’ll see, some of the files are symlinks and some aren’t. This was built into hcp2bids.

hcp_test_bids
├── dataset_description.json
├── sub-132118
│   ├── anat
│   │   ├── sub-132118_run-01_T1w.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/T1w_MPR2/132118_3T_T1w_MPR2.nii.gz
│   │   ├── sub-132118_run-01_T2w.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/T2w_SPC1/132118_3T_T2w_SPC1.nii.gz
│   │   ├── sub-132118_run-02_T1w.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/T1w_MPR1/132118_3T_T1w_MPR1.nii.gz
│   │   └── sub-132118_run-02_T2w.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/T2w_SPC2/132118_3T_T2w_SPC2.nii.gz
│   ├── fmap
│   │   ├── sub-132118_dir-1_epi.json
│   │   ├── sub-132118_dir-1_epi.nii.gz
│   │   ├── sub-132118_dir-2_epi.json
│   │   ├── sub-132118_dir-2_epi.nii.gz
│   │   ├── sub-132118_dir-3_epi.json
│   │   ├── sub-132118_dir-3_epi.nii.gz
│   │   ├── sub-132118_dir-4_epi.json
│   │   ├── sub-132118_dir-4_epi.nii.gz
│   │   ├── sub-132118_dir-5_epi.json
│   │   ├── sub-132118_dir-5_epi.nii.gz
│   │   ├── sub-132118_dir-6_epi.json
│   │   ├── sub-132118_dir-6_epi.nii.gz
│   │   ├── sub-132118_dir-7_epi.json
│   │   ├── sub-132118_dir-7_epi.nii.gz
│   │   ├── sub-132118_dir-8_epi.json
│   │   ├── sub-132118_dir-8_epi.nii.gz
│   │   ├── sub-132118_run-01_magnitude1.nii.gz
│   │   ├── sub-132118_run-01_magnitude2.nii.gz
│   │   ├── sub-132118_run-01_phasediff.json
│   │   ├── sub-132118_run-01_phasediff.nii.gz
│   │   ├── sub-132118_run-02_magnitude1.nii.gz
│   │   ├── sub-132118_run-02_magnitude2.nii.gz
│   │   ├── sub-132118_run-02_phasediff.json
│   │   ├── sub-132118_run-02_phasediff.nii.gz
│   │   ├── sub-132118_run-03_magnitude1.nii.gz
│   │   ├── sub-132118_run-03_magnitude2.nii.gz
│   │   ├── sub-132118_run-03_phasediff.json
│   │   ├── sub-132118_run-03_phasediff.nii.gz
│   │   ├── sub-132118_run-04_magnitude1.nii.gz
│   │   ├── sub-132118_run-04_magnitude2.nii.gz
│   │   ├── sub-132118_run-04_phasediff.json
│   │   └── sub-132118_run-04_phasediff.nii.gz
│   └── func
│       ├── sub-132118_task-rest_acq-lr_run-01_bold.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST1_LR/132118_3T_rfMRI_REST1_LR.nii.gz
│       ├── sub-132118_task-rest_acq-lr_run-01_sbref.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST1_LR/132118_3T_rfMRI_REST1_LR_SBRef.nii.gz
│       ├── sub-132118_task-rest_acq-lr_run-02_bold.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST2_LR/132118_3T_rfMRI_REST2_LR.nii.gz
│       ├── sub-132118_task-rest_acq-lr_run-02_sbref.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST2_LR/132118_3T_rfMRI_REST2_LR_SBRef.nii.gz
│       ├── sub-132118_task-rest_acq-rl_run-01_bold.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST1_RL/132118_3T_rfMRI_REST1_RL.nii.gz
│       ├── sub-132118_task-rest_acq-rl_run-01_sbref.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST1_RL/132118_3T_rfMRI_REST1_RL_SBRef.nii.gz
│       ├── sub-132118_task-rest_acq-rl_run-02_bold.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST2_RL/132118_3T_rfMRI_REST2_RL.nii.gz
│       └── sub-132118_task-rest_acq-rl_run-02_sbref.nii.gz -> /gpfs/gsfs6/users/HCP/hcp_test/132118/unprocessed/3T/rfMRI_REST2_RL/132118_3T_rfMRI_REST2_RL_SBRef.nii.gz
├── task-rest_acq-lr_bold.json
├── task-rest_acq-lr_sbref.json
├── task-rest_acq-rl_bold.json
└── task-rest_acq-rl_sbref.json

4 directories, 49 files

The sample dataset passes the validator in the browser but fails when using a singularity image (v1.6.0) on our HPC:

	1: [ERR] Internal error. SOME VALIDATION STEPS MAY NOT HAVE OCCURRED (code: 0 - INTERNAL ERROR)
		.undefined
			Evidence: Error: Could not load HED schema version "Latest" from remote repository - "Error: getaddrinfo EAI_AGAIN raw.githubusercontent.com".
    at /src/node_modules/hed-validator/utils/schema.js:40:13
    at tryCallOne (/src/node_modules/promise/lib/core.js:37:12)
    at /src/node_modules/promise/lib/core.js:123:15
    at flush (/src/node_modules/asap/raw.js:50:29)
    at processTicksAndRejections (internal/process/task_queues.js:76:11)

	Please visit https://neurostars.org/search?q=INTERNAL ERROR for existing conversations about this issue.

	2: [ERR] 'IntendedFor' field needs to point to an existing file. (code: 37 - INTENDED_FOR)
		./sub-132118/fmap/sub-132118_dir-1_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-rl_run-01_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-2_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-rl_run-01_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-3_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-lr_run-01_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-4_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-lr_run-02_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-5_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-lr_run-01_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-6_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-rl_run-02_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-7_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-lr_run-02_bold.nii.gz
		./sub-132118/fmap/sub-132118_dir-8_epi.nii.gz
			Evidence: func/sub-132118_task-rest_acq-rl_run-02_bold.nii.gz
		./sub-132118/fmap/sub-132118_run-01_phasediff.nii.gz
			Evidence: anat/sub-132118_run-01_T1w.nii.gz
		./sub-132118/fmap/sub-132118_run-02_phasediff.nii.gz
			Evidence: anat/sub-132118_run-02_T2w.nii.gz
		... and 2 more files having this issue (Use --verbose to see them all).

	Please visit https://neurostars.org/search?q=INTENDED_FOR for existing conversations about this issue.

	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 https://neurostars.org/search?q=README_FILE_MISSING for existing conversations about this issue.

	2: [WARN] The Authors field of dataset_description.json should contain an array of fields - with one author per field. This was triggered because there are no authors, which will make DOI registration from dataset metadata impossible. (code: 113 - NO_AUTHORS)

	Please visit https://neurostars.org/search?q=NO_AUTHORS for existing conversations about this issue.


        Summary:                 Available Tasks:        Available Modalities:
        37 Files, 42.25MB        rest                    epi
        1 - Subject                                      magnitude1
        1 - Session                                      magnitude2
                                                         phasediff


	If you have any questions, please post on https://neurostars.org/tags/bids.

We decided to push through to get fmriprep running and are hitting an error. Below is the fmriprep command, version 20.2.0. I have doubled check that we have all the available resources needed for this processing.

singularity run --cleanenv                                                      \
	-B $HOME/.cache/templateflow:/templateflow         							\
	-B /data/HCP/hcp_test_bids:/in                                              \
	-B /data/HCP/fmriprep_test:/out                                             \
	-B /data/HCP/fmriprep_test/wrk:/wrk                                         \
	/data/MLDSST/containers/fmriprep-20.2.0.simg /in /out participant           \
	--participant_label sub-132118                                              \
	-w /wrk                                                                     \
	--nthreads $SLURM_CPUS_PER_TASK                                             \
	--mem_mb $SLURM_MEM_PER_NODE                                                \
	--fs-license-file /data/BrainBlocks/code/license.txt                        \
	--output-spaces fsaverage6 fsLR MNI152NLin2009cAsym                         \
	--cifti-output                                                              \
	--skip-bids-validation                                                      \
	--notrack                                                                   \
	--omp-nthreads 8                                                            \
	--use-aroma                                                                 \
	--error-on-aroma-warnings

And here is the error:

/usr/local/miniconda/lib/python3.7/site-packages/bids/layout/validation.py:46: UserWarning: The ability to pass arguments to BIDSLayout that control indexing is likely to be removed in future; possibly as early as PyBIDS 0.14. This includes the `config_filename`, `ignore`, `force_index`, and `index_metadata` arguments. The recommended usage pattern is to initialize a new BIDSLayoutIndexer with these arguments, and pass it to the BIDSLayout via the `indexer` argument.
  warnings.warn("The ability to pass arguments to BIDSLayout that control "
You are using fMRIPrep-20.2.0, and a newer version of fMRIPrep is available: 20.2.1.
Please check out our documentation about how and when to upgrade:
https://fmriprep.readthedocs.io/en/latest/faq.html#upgrading
210422-15:18:28,290 cli WARNING:
	 Per-process threads (--omp-nthreads=8) exceed total threads (--nthreads/--n_cpus=2)
210422-15:18:35,116 nipype.workflow WARNING:
	 Previous output generated by version 0+unknown found.
210422-15:18:35,163 nipype.workflow IMPORTANT:

    Running fMRIPREP version 20.2.0:
      * BIDS dataset path: /in.
      * Participant list: ['132118'].
      * Run identifier: 20210422-151824_4c3074cb-2378-4557-92a9-41776d33eca6.
      * Output spaces: fsaverage:den-41k fsLR MNI152NLin2009cAsym:res-native.
      * Pre-run FreeSurfer's SUBJECTS_DIR: /out/freesurfer.
210422-15:18:36,196 nipype.workflow INFO:
	 Using single-band reference file(s) sub-132118_task-rest_acq-lr_run-01_sbref.nii.gz.
210422-15:18:37,156 nipype.workflow INFO:
	 Using single-band reference file(s) sub-132118_task-rest_acq-lr_run-02_sbref.nii.gz.
210422-15:18:37,782 nipype.workflow INFO:
	 Using single-band reference file(s) sub-132118_task-rest_acq-rl_run-01_sbref.nii.gz.
210422-15:18:38,434 nipype.workflow INFO:
	 Using single-band reference file(s) sub-132118_task-rest_acq-rl_run-02_sbref.nii.gz.
210422-15:18:42,544 nipype.workflow INFO:
	 fMRIPrep workflow graph with 1413 nodes built successfully.
210422-15:18:55,955 nipype.workflow IMPORTANT:
	 fMRIPrep started!
210422-15:19:23,2 nipype.workflow CRITICAL:
	 fMRIPrep failed: The 'header_source' trait of a _MergeInputSpec instance must be a pathlike object or string representing an existing file, but a value of '/in/sub-132118/func/sub-132118_task-rest_acq-lr_run-01_bold.nii.gz' <class 'str'> was specified.
  1. Do you have any idea why the validator might be giving different results between the browser and the singularity image? The browser is running from the same machine as the container.
  2. What issues might cause the fmriprep error?

Thanks in advance for your help!

Bump. Does anyone have suggestions of where to look to troubleshoot the following error?

fMRIPrep failed: The 'header_source' trait of a _MergeInputSpec instance must be a pathlike object or string representing an existing file, but a value of '/in/sub-132118/func/sub-132118_task-rest_acq-lr_run-01_bold.nii.gz' <class 'str'> was specified.
1 Like

I would start by trying 2 things:

1 Fix the “IntendedFor” fields, not only to help the BIDS-validator finish but also so fMRIPrep can use the fieldmaps for SDC. I can share a short script in python that does that, would you be interested?
2 Update the bids validator singularity image to the most recent version (1.7.1 at this time).

Best,
Steven

Thanks, I’ll check with an updated validator image.
And sure, I’ll happily look at a script to fix the intendedFor field, I’m not sure what’s wrong with ours so this’ll be helpful.

To be expected, but I got the same validator output errors from version 1.7.1 as 1.6.0.

Please DM me your email and I’ll send you the jupyter notebook for adding intendedfor fields. Neurostars does not allow me to upload these file types here.

1 Like

I worked with Jessica yesterday and we discovered the symbolic links were the problem. Though singularity didn’t mind reading the symbolic links being mapped into the container with the -B argument, singularity didn’t know where the destinations were pointing because the symbolic link destinations were unknown to the singularity container. Adding in the additional singularity -B argument mapping in the destination of the symbolic links resolved the issue. This is likely why the online validator was fine to pull and read from the filesystem, but the singularity on-system validator did not work.

2 Likes