Description:
Skipped all the BOLD images in session and complained that no BOLD images found
Call:
qunex hcp_fmri_volume \
--sessionsfolder="/myStudy/bids_data/sessions" \
--parelements="4" \
--sessions="/myStudy/bids_data/processing/batch.txt" \
--sessionids="042_1"
Logs:
# Generated by QuNex 1.3.2 [QIO] on 2025-10-03_15.13.26.609563
#
============================= LOG ================================
# Generated by QuNex 1.3.2 [QIO] on 2025-10-03_15.13.26.602406#
=================================================================
qunex hcp_fmri_volume \
--sessionsfolder="/myStudy/bids_data/sessions" \
--parelements="4" \
--sessions="/myStudy/bids_data/processing/batch_fmri_edit.txt" \
--sessionids="042_1"
=================================================================
Starting multiprocessing sessions in /myStudy/bids_data/processing/batch_fmri_edit.txt with a pool of 1 concurrent processes
------------------------------------------------------------
Session id: 042_1
[started on Friday, 03. October 2025 15:13:26]
Running HCP fMRI Volume pipeline [HCPStyleData] ...
---> PreFS results present.
---> FS results present.
---> PostFS results present.
---> Looking for spin echo fieldmap set images [PA/AP].
... identified folder SpinEchoFieldMap1
... phase positive PA spin echo fieldmap image present
... phase negative AP spin echo fieldmap image present
... TOPUP configuration file present
Skipping the following BOLD images:
... bold1 [task-FUNCMUX6_run-01_bold]
... bold2 [task-FUNCMUX6_run-02_bold]
... bold3 [task-FUNCMUX6_run-03_bold]
... bold4 [task-FUNCMUX6_run-04_bold]
---> ERROR: No BOLD images found for session 042_1! Check your data or the contents of the batch file.
Running 1 BOLD images in parallel
HCP fMRIVolume completed on Friday, 03. October 2025 15:13:26
------------------------------------------------------------
---> Final report for command hcp_fmri_volume
... 042_1 ---> HCP fMRI Volume: bolds 1, 2, 3, 4 skipped
---> Successful completion of all tasks
** For an example of how to report an issue, please refer to this post.
the relevant part in the batch.txt file:
session: 042_1
subject: 042
bids: /myStudy/bids_data/sessions/042_1/bids
raw_data: /myStudy/bids_data/sessions/042_1/nii
hcp: /myStudy/bids_data/sessions/042_1/hcp
hcpready: true
1 :T1w :T1w: se(1)
2 :T2w :T2w: se(1)
3 :SE-FM-PA :dir-fpe_epi: EchoSpacing(0.000624): se(1): phenc¶
4 :SE-FM-AP :dir-rpe_epi: EchoSpacing(0.000624): se(1): phenc(AP)
5 :bold1: task-FUNCMUX6_run-01_bold: EchoSpacing(0.000732): se(1): phenc¶
6 :bold2: task-FUNCMUX6_run-02_bold: EchoSpacing(0.000732): se(1): phenc¶
7 :bold3: task-FUNCMUX6_run-03_bold: EchoSpacing(0.000732): se(1): phenc¶
8 :bold4: task-FUNCMUX6_run-04_bold: EchoSpacing(0.000732): se(1): phenc¶
#NOTE: the above symbol ¶ was generated by enclosing “PA” in () on this form, not in the original text file, in which “PA” is enclosed by parentheses.
tree structure of the input data
/sessions/042_1/hcp/042_1/unprocessed/
├── BOLD_1_PA
│ ├── 042_1_BOLD_1_PA.json
│ └── 042_1_BOLD_1_PA.nii.gz
├── BOLD_2_PA
│ ├── 042_1_BOLD_2_PA.json
│ └── 042_1_BOLD_2_PA.nii.gz
├── BOLD_3_PA
│ ├── 042_1_BOLD_3_PA.json
│ └── 042_1_BOLD_3_PA.nii.gz
├── BOLD_4_PA
│ ├── 042_1_BOLD_4_PA.json
│ └── 042_1_BOLD_4_PA.nii.gz
├── SpinEchoFieldMap1
│ ├── 042_1_BOLD_AP_SB_SE.json
│ ├── 042_1_BOLD_AP_SB_SE.nii.gz
│ ├── 042_1_BOLD_PA_SB_SE.json
│ └── 042_1_BOLD_PA_SB_SE.nii.gz
├── T1w
│ ├── 042_1_T1w_MPR1.json
│ └── 042_1_T1w_MPR1.nii.gz
└── T2w
├── 042_1_T2w_SPC1.json
└── 042_1_T2w_SPC1.nii.gz
7 directories, 16 files
Hi Xhao!
Looks like a bug. Let me investigate this over the next couple of days and get back to you.
If possible, is this perhaps a session you could upload (raw data) so I can reproduce this behavior and figure out what is going on? Thanks! That would be super helpful as we never saw such behavior before.
Also, does the same happen for other sessions in the study or is this an outlier?
Best, Jure
Thank you for the quick response, Jure! Please download the above-mentioned data from:
The same error also occurred when running another test dataset from the same study (I only ran two datasets so far with the new version of QuNex), even though both datasets were processed successfully two years ago using QuNex version 0.94.
By the way, when I ran the create_session_info process on the datasets, the new version of QuNex split the topup fieldmap from GE into two separate field maps, labeled se(1) and se(2) (as shown in the session_hcp.txt file copied below). It then reported a missing file for the required pair of topups. To resolve this, I manually changed all instances of se(2) to se(1) in the session_hcp.txt file and merged the two folders — SpinEchoFieldMap1 and SpinEchoFieldMap2 — in the /hcp/_042_1/unprocessed/ directory. This allowed me to successfully run the PreFreeSurfer stage.
The content of the generated “session_hcp.txt” file:
session: 042_1
subject: 042
bids: /myStudy/bids_data/sessions/042_1/bids
raw_data: /myStudy/bids_data/sessions/042_1/nii
hcp: /myStudy/bids_data/sessions/042_1/hcp
hcpready: true
1 :T1w :T1w: se(1)
2 :T2w :T2w: se(1)
3 :SE-FM-PA :dir-fpe_epi: EchoSpacing(0.000624): se(1): phenc¶
4 :SE-FM-AP :dir-rpe_epi: EchoSpacing(0.000624): se(2): phenc(AP)
5 :bold1:task-FUNCMUX6_run-01_bold: EchoSpacing(0.000732): se(2): phenc¶
6 :bold2:task-FUNCMUX6_run-02_bold: EchoSpacing(0.000732): se(2): phenc¶
7 :bold3:task-FUNCMUX6_run-03_bold: EchoSpacing(0.000732): se(2): phenc¶
8 :bold4:task-FUNCMUX6_run-04_bold: EchoSpacing(0.000732): se(2): phenc¶
Thanks again.
Best,
Xuejun
Hi,
Thanks for the dataset.
Yes, it seems like we have 2 issues here. The first one is the non-detection of BOLDs and the second one is the onboarding issue. The SE pair should be labeled as se(1) and then applied correctly to all other images.
Let me test and debug both issues simultaneously. Should be easy to do since I have the data. I will try to wrap this up this week/start of the next week. And then release a new version that patches everything as well.
Best, Jure
Great. Thank you again, Jure!
Hi,
The issue with onboarding is that your directions are not named in the completely standard way. The standard is to have dir-AP/PA or dir-LR/RL. You use naming dir-fpe and dir-bpe which is technically supported by BIDS but not exactly standard.
I already coded support for this less standard naming convention to QuNex. It will be released in the next version. I am now processing the data so I can get it to BOLD steps. Will keep you posted here.
Best, Jure
Can you also please upload your mapping and parameters/batch file, so I can use the same inputs as you. Thanks!
Best, Jure
Sure. Please find attached files.
parameters.txt (4.3 KB)
mapping.txt (598 Bytes)
batch.txt (5.3 KB)
Hi,
The mistake here is quite silly. Inside the parameters.txt and then as a consequence also batch.txt you have _bolds: rest. This means that only BOLDs that are marked as rest will be processed. Since your bolds are all task the results is obvious 
. After setting _bolds: all things worked fine. Note that I also had to change _hcp_unwarp_dir as you do not have a default set. I set it to y but I am not sure if that is OK, set as needed. Below is the script that I used and that worked and the mapping file I used. I also added some QC steps in between so I could perform some sanitcy checks that things indeed work. The mapping file is similar to yours but a bit cleaned up.
forum_bids.sh (3.3 KB)
mapping.txt (247 Bytes)
Note that I am using release candidate for QuNex 1.4.0, which already supports dir-fpe and dir-bpe naming in BIDS. The current released version 1.3.6 does not support that, so you need to change dir in the filename to AP/PA if you want to do everything in an end-to-end manner without manual edits of batch/session files.
Best, Jure
Thanks for finding the mistake, Jure! I made only minimal modifications to the sample parameter file “HCPA001_parameters.txt”. Interestingly, the same parameters worked fine with QuNex version 0.94, so I didn’t suspect they could be causing the problem. A detailed explanation of how to set those parameters, including the allowed values and how they affect processing, would be very helpful. As an example, I’ve seen the _bolds parameter set to “all” or “rest.” Does it also support a “task” option? Where can a user find this kind of information?
Anyway, thanks again for your help.
Best,
Xuejun
Yes, task is also a valid option. Or a custom tag in that manner. Most of this information can be found in the official documentation: Background — QuNex documentation. There you have general pages as well as specific documentation for every command and all of its parameters.
If something is missing or not clear, let us know via this forum and we will expand it.
Best, Jure