[RESOLVED] Import HCP-YA released data


I’d like to process the diffusion data from HCP-YA S1200 release.

In specific, I hope to use the released preprocessed diffusion data (100206/T1w/Diffusion) or bedpostx results (100206/T1w/Diffusion.bedpostx), and run dwi_probtrackx_dense_gpu + dwi_parcellate to derive the structural connectomes.

However, I’m not sure how should I use import_hcp or map_hcp_data to import the data into QuNex. Could you help with this? Thank you!

Zhen-Qi Liu


I just successfully used the command below to import HCPYA data into a QuNex study:

# import hcp
qunex import_hcp \
	--sessionsfolder="/gpfs/project/fas/n3/Studies/QuNexAcceptTest/Results/jd2528/qunexdev/QuNex_acceptance_MB_HCPYA_20220401_050424/sessions" \
	--inbox="/gpfs/project/fas/n3/Studies/QuNexAcceptTest/RawData/QuNex_acceptance_MB_HCPYA" \
	--sessions="100206_3T" \
	--action="copy" \
	--overwrite="yes" \
	--archive="leave" \
	--nameformat="(?P<subject_id>[^/]+?)/unprocessed/(?P<session_name>.*?)/(?P<data>.*)" \

Note the different name format. A different pattern needs to be used to properly extract subject info from the folder structure.

Cheers, Jure.

Hi, did the above solution resolve your issue?

Hi, thank you for the reply. It works perfectly to import the unprocessed data.

According to this newer post, I also did setup_hcp. However, I’m not sure how to properly do step 3: move the processed data to the hcp folder. Do I just move the released T1w and MNINonLinear folders for each subject? Anything needs special attention?

Thank you!


I believe that copying the released T1w and MNINonLinear folders in the <study>/sessions/<session id>/hcp/<session id> folder should suffice. Please, try it out for a single session and let us know if it worked ok.

All the best, Grega


To update, I have been using the following sequence of commands to derive the structural connectomes.

  • create_study, import_hcp (custom nameformat suggested above), setup_hcp, create_batch
  • bash commands to (1) soft link {SUBJ_ID}/MNINonLinear and {SUBJ_ID}/T1w folders, and (2) rename {SUBJ_ID} to {SUBJ_ID}_3T
  • dwi_pre_tractography, dwi_probtrackx_dense_gpu, dwi_parcellate

The pipeline runs without error (thanks for the great tool!). I’m currently working to check the output connectomes for correctness.

In general, as I’m using QuNex in a pretty black-box way to import and process the data, my concern is if qunex correctly handled the linked data (since HCP-YA official release was long ago and there might be potential pipeline changes). For example, MNI vs T1w spaces, resolutions, ROI masks, etc.

Please let me know if there’s anything I should pay special attention to. I’ll also update my findings in this thread.

Thank you!

Hi, thank you for the kind words. It is hard to say whether everything you did is OK, but the fact that the processing finished successfully is a strong indicator that you did. If you were to import some files erroneously or if some files were missing the processing would not be able to complete and QuNex commands would report errors. The best way to keep track of what is going on are the QuNex logs. To give you a definitive answer whether everything went fine in your processing we would need to have access to your QuNex study and its logs.

Please keep us posted regarding the progress of your processing and we will help out to the best of our abilities.