hcp fix multi-run error

[Austin Cooper]

Hello!

I'm noticing that my FIX processing seems to be stopping short in terms of outputting all expected files. 

The particular error I'm seeing is:

Fri Oct 10 10:07:23 EDT 2025:hcp_fix_multi_run: Running MELODIC located at: /cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/melodic
Fri Oct 10 10:07:23 EDT 2025:hcp_fix_multi_run: melodic_cmd: /cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/melodic -i all_fMRI_data_hp0_vnts -o all_fMRI_data_hp0.ica/filtered_func_data.ica --nobet --report --Oall --tr=0.937000 --vn --dim=328 -m all_fMRI_data_brain_mask
Fri Oct 10 11:03:34 EDT 2025:hcp_fix_multi_run: melodic has been run: return_code = 0
Fri Oct 10 11:03:47 EDT 2025:hcp_fix_multi_run: Running FIX
Fri Oct 10 11:03:47 EDT 2025:hcp_fix_multi_run: using training data file: /project/6001995/coopapap/scripts/HCPpipelines-master/fix/training_files/HCP_Style_Single_Multirun_Dedrift.RData
Fri Oct 10 11:03:47 EDT 2025:hcp_fix_multi_run: fix_cmd: /project/6001995/coopapap/scripts/HCPpipelines-master/fix/fix all_fMRI_data_hp0.ica /project/6001995/coopapap/scripts/HCPpipelines-master/fix/training_files/HCP_Style_Single_Multirun_Dedrift.RData 10
Fri Oct 10 11:03:47 EDT 2025:hcp_fix_multi_run: Check all_fMRI_data_hp0.ica/fix/logMatlab.txt for log output from feature extraction
Fri Oct 10 11:03:47 EDT 2025:hcp_fix_multi_run: Check all_fMRI_data_hp0.ica/.fix_2b_predict.log for log output from component classification
Fri Oct 10 11:03:47 EDT 2025:hcp_fix_multi_run: Check all_fMRI_data_hp0.ica/.fix.log for log output from cleanup stage
FIX Feature extraction for Melodic output directory: all_fMRI_data_hp0.ica
 create edge masks
 run FAST
 registration of standard space masks
 extract features
FIX Classifying components in Melodic directory: all_fMRI_data_hp0.ica using training file: /project/6001995/coopapap/scripts/HCPpipelines-master/fix/training_files/HCP_Style_Single_Multirun_Dedrift.RData and threshold 10
No valid labelling file specified
Fri Oct 10 11:50:41 EDT 2025:hcp_fix_multi_run: Done running FIX
Could not find a supported file with prefix "all_fMRI_data_hp0.ica/filtered_func_data_clean"
Fri Oct 10 11:50:41 EDT 2025:hcp_fix_multi_run:  While running '/project/6001995/coopapap/scripts/HCPpipelines-master/ICAFIX/hcp_fix_multi_run --fmri-names=/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_TASK1_AP/tfMRI_TASK1_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_TASK2_AP/tfMRI_TASK2_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_TASK3_AP/tfMRI_TASK3_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_TASK4_AP/tfMRI_TASK4_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/rfMRI_REST1_AP/rfMRI_REST1_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_MOVIE1_AP/tfMRI_MOVIE1_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_MOVIE2_AP/tfMRI_MOVIE2_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_MOVIE3_AP/tfMRI_MOVIE3_AP@/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/tfMRI_MOVIE4_AP/tfMRI_MOVIE4_AP --high-pass=0 --concat-fmri-name=/scratch/coopapap/data/MGH/HCPNaming/glaucoma/15_scan1/MNINonLinear/Results/all_fMRI_data/all_fMRI_data --motion-regression=FALSE --training-file=HCP_Style_Single_Multirun_Dedrift.RData --fix-threshold=10 --delete-intermediates=FALSE --config= --processing-mode=HCPStyleData':
Fri Oct 10 11:50:41 EDT 2025:hcp_fix_multi_run: ERROR: '/cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/immv' command failed with return code: 1
Fri Oct 10 11:50:41 EDT 2025:hcp_fix_multi_run: ERROR: '/cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/immv' command failed with return code: 1

===> ERROR: Command returned with nonzero exit code
---------------------------------------------------
         script: hcp_fix_multi_run
stopped at line: 909
           call: $FSLDIR/bin/immv ${concatfmrihp}.ica/filtered_func_data_clean ${concatfmrihp}_clean
  expanded call: /cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/immv all_fMRI_data_hp0.ica/filtered_func_data_clean all_fMRI_data_hp0_clean
       hostname: rl21703
      exit code: 1
---------------------------------------------------

===> Aborting execution!

and I do notice that the subjects I have successfully processed in the past had a file named "prefiltered_func_data_mcf_conf_hp_clean.nii.gz" in their .../MNINonLinear/Results/all_fMRI_data/all_fMRI_data_hp0.ica/mc folder, whilst this subject does not have that file. 

I wonder if this file's absence is to do with the "No valid labelling file specified" that I see in the output log? 

A glimpse into the parameters I have set for this processing step are:

fMRINames="tfMRI_TASK1_AP@tfMRI_TASK2_AP@tfMRI_TASK3_AP@tfMRI_TASK4_AP@rfMRI_REST1_AP@tfMRI_MOVIE1_AP@tfMRI_MOVIE2_AP@tfMRI_MOVIE3_AP@tfMRI_MOVIE4_AP"

ConcatNames="all_fMRI_data"

bandpass=0 #changed from 2000 to 0 since it apparently reduces the computational load

domot=FALSE

MRTrainingData=HCP_Style_Single_Multirun_Dedrift.RData

SRTrainingData=HCP_hp2000.RData

I wonder if this is enough information for you to provide a pointer/solution?

[Tim Coalson]

We recommend using pyfix instead of legacy fix when feasible (use a recent version of FSL 6 and leave FSL_FIXDIR unset).  "No valid labelling file specified" may mean that legacy R-based fix failed at an earlier step, and it tends to put the actual error message into some other log file instead of reporting it to the user.

Tim

--
You received this message because you are subscribed to the Google Groups "HCP-Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hcp-users+...@humanconnectome.org.
To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/34f4ed84-b210-4dd6-b4a0-e00c7fddbb41n%40humanconnectome.org.

[Austin Cooper]

Thanks Tim. The FSL module I'm using is fsl/6.0.7.18 with r/4.3.1. I also have easy access to fsl/6.0.7.7. I believe I used fsl/6.0.1 when running this with another compute cluster, and perhaps this module is more adept for this pipeline? Any insights?

Next, I see that  FSL_FIXDIR is set to FSL_FIXDIR=$HCPPIPEDIR/fix in the setupHCPPipeline script... should I remove this line?

I also see this type of warning come up in many different locations of the output text file:

Detected locale "C" with character encoding "ANSI_X3.4-1968", which is not UTF-8.
Qt depends on a UTF-8 locale, and has switched to "C.UTF-8" instead.
If this causes problems, reconfigure your locale. See the locale(1) manual for more information.

Any pointers regarding these would be greatly appreciated!

[Tim Coalson]

Commenting out the FSL_FIXDIR line should switch to python fix, yes.

That warning is harmless.  QT6 apparently disagrees with "LC_ALL=C", which we have set for reproducibility purposes.  We may change that to "LC_ALL=C.UTF-8" at some point to reduce the warning messages.

Tim