FBA population_template error: UID contains whitespace

Hi,all
I finished all steps of FBA followed this using 2 patients data at two different time (time1:patient01a,patient02a;time2:patient01b,patient02b).
But when I choose 23 patients data at two different time(46 total),I met this problem when I want to finish Step 9.

How can I solve this problem?

2X2 data USES ANTS(Bias field correction in here) but the 23X2 data is not used.

Thanks,
Silver

@rsmith
How do you think this problem can be solved?I’m looking forward to your advice. :grin:

Thanks,
Silver

Hi Silver,

This looks like a bug in population_template. Could you please rerun population_template with the -info option and report the output before initial alignment of images. Also, what is the output of ls -la ../template/mask_intput?

Hi Max,
I am running again without ANTs to test if this bug is still existing.I am running dwi2fod(Step 7 in this page) when I finished this step I will show population_template(Step 9) with the -info option to you.

I don’t what “report the output before initial alignment of images. Also, what is the output of ls -la …/template/mask_intput?” exactly mean.And,"…/template/mask_input" is each data’s dwi_mask_upsampled.mif.


population_template uses some logic to match images to masks and internally generates filenames using unique identifiers of the input images. Something in that process seems to go wrong, so I’d need to know the file names of your input masks (using ls -la) and how population_template interprets them which is shown with the -info flag. For instance:

Ok, I think I know the reason for the error. Until this fix is merged, make sure your input images and masks do not contain whitespace characters in the uniquely identifying portions of the filename. As shown in the above screenshot, whitespace in the directory name (mask 2 or in the common file prefix (mask ) or post-fix (.mif.gz) are fine but using mask subject 3.mif.gz leads to an error. If that’s not the reason, please let me know.

Hi Max,
I showed “how population_template interprets them which is shown with the -info flag” (image 1)and “the file names of my input masks(using ls -la)” (image 2)

image 1

image2:

Edit: Because I’m wondering if ANTs is the cause of this,I reran the the first eight steps again (except for the third),but this question still exists.The picture above shows the result of my re-run.The picture I showed 2 days ago includes ANTs.

Hi Max,
I want to apologize to you for my carelessness.

In this image,‘03_06 ’ should be ‘03_06’.I change the name and the program can continue to run.Thank you very much for your patient answer.

Thanks,
Silver

1 Like