sswarper2 failed Fatal Signal 2 (SIGINT) received

AFNI Message Board
1

AFNI version info (afni -ver): AFNI version=AFNI_25.0.06 (Feb 12 2025) [64-bit]
Hello there,
I am running afni on Ubuntu. I am trying to do some basic preprocessing and somehow my codes don't work (they work on other lab member's device). They stop at "*** Coarse pass begins ***" every time. Hopefully I can get some feedbacks. Thanks!

code text  
#!/bin/bash

raw_dir="/mnt/c/Users/hulas/ds003721-download"
processed_dir="/mnt/c/Users/hulas/ds003721-download/processed"

mkdir -p "$processed_dir/SSwarper"

subjects=("sub-04")

for subj in "${subjects[@]}"
do
    mkdir -p "$processed_dir/sswarper/$subj"
    dir_in="$raw_dir/$subj/anat"
    dir_out="$processed_dir/sswarper/$subj"

    sswarper2 \
    -input "$dir_in/${subj}_acq-highres_T1w_deface.nii" \
    -base MNI152_2009_template_SSW.nii.gz \
    -deoblique \
    -subid "$subj" \
    -odir "$dir_out"

done

below is the error message

++ 3dAllineate: AFNI version=AFNI_25.0.06 (Feb 12 2025) [64-bit]
++ Authored by: Zhark the Registrator
++ lpa+ parameters: hel=0.40 mi=0.00 nmi=0.20 crA=0.40 ov=0.40 [to be zeroed at Final iteration]
++ Source dataset: /mnt/c/Users/hulas/ds003721-download/processed/sswarper/sub-04/anatUAC.sub-04.nii
++ Base dataset:   /home/jieronglin/abin/MNI152_2009_template_SSW.nii.gz
++ Loading datasets into memory
++ Local correlation: blok type = 'TOHD(5.17699)'
++ lpa+ parameters: hel=0.40 mi=0.00 nmi=0.20 crA=0.40 ov=0.40 [to be zeroed at Final iteration]
*+ WARNING: center of mass shifts (-cmass) are turned off, but would be TERRIBLY large!
*+ WARNING:   - at least one is more than 50% of search range
 +        -cmass x y z shifts =    0.400  -29.829  -61.208
 +  shift search range is +/- =   75.435   86.349   73.830
 +                                  0.5%    34.5%    82.9%
++ master dataset for output = base
++ OpenMP thread count = 15
++ ======= Allineation of 1 sub-bricks using Local Pearson Abs + Others =======
++ *** Coarse pass begins ***
^C
Fatal Signal 2 (SIGINT) received
      GA_get_warped_overlap
     GA_scalar_costfun
    GA_scalar_fitter
   mri_genalign_scalar_ransetup
  3dAllineate
 Bottom of Debug Stack
** AFNI version = AFNI_25.0.06  Compile date = Feb 12 2025
** [[Precompiled binary linux_ubuntu_16_64: Feb 12 2025]]
** Program Death **
2

Howdy-

OK, I ran that same command on my computer (using a different dataset), and did not see that issue.

My guess is that what is happening is that there is not great overlap, and that is causing woe. In your output directory, there should be a subdirectory called ssw_align_hist* for QC images as the program works. You should have 1-2 images there of the initial overlap. What do those images look like?

Alternatively, if even those were not created (but I imagine they should be), what is the output of:

@djunct_overlap_check \
     -ulay "$dir_in/${subj}_acq-highres_T1w_deface.nii" \
     -olay MNI152_2009_template_SSW.nii.gz

? (where you might need to add the path for the MNI*SSW.nii.gz dset; I don't think that program looks automatically for it in special directories like the alignment command does.) If there are two images output, could you please output both?

thanks,
pt

3

Thank you for the reply! All the generated output images seem to be misfit to the atlases. I even tried other atlases but still getting similar results. Please see the screen shot below:

image
image1449×578 114 KB

J

4

Hi, those initial overlaps are not wonderful, but I am not sure why they would be deeply problematic either.

I will send you an email about checking about the data.

--pt

5

This suggests that running @Align_Centers might be the way to go. The program probably ran out of RAM and crashed.

-rick