I recently ran into an error message from auto_warp.py saying "Error: Anatomical dset pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD not found." However, looking into the folder of "awpy", this file does exist in there. To demonstrate this issue, I was simply warping the "MNI152_T1_2009c+tlrc" to the template of "MNI_caez_N27+tlrc"; and pasting the script output below:
[lzh@master2 test_anat]$ auto_warp.py -base MNI_caez_N27+tlrc -input ./MNI152_T1_2009c+tlrc
/usr/local/AFNI/linux_xorg7_64/afnipy/db_mod.py:6822: SyntaxWarning: invalid escape sequence '\_'
"%s -prefix %s\_REML%s\n" \
#++ auto_warp.py version: 0.06
# Output directory /data/home3/lzh/temp5/test_anat/awpy/
#Script is running:
mkdir /data/home3/lzh/temp5/test_anat/awpy/
cd /data/home3/lzh/temp5/test_anat/awpy/
#Script is running (command trimmed):
3dcopy /data/home3/lzh/temp5/test_anat/MNI152_T1_2009c+tlrc ./anat.nii
++ 3dcopy: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
#Script is running (command trimmed):
3dUnifize -GM -input ./anat.nii -prefix ./anat.un.nii
++ 3dUnifize: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
+ Pre-processing: AD
+ Voxel loop: 0123456789.0123456789.0123456789.0123456789.0123456789.!UW[s1]Gm
++ Output dataset ./anat.un.nii
++ ===== Elapsed = 49.3 sec
#Script is running (command trimmed):
3dcopy /usr/local/AFNI/linux_xorg7_64/MNI_caez_N27+tlrc ./base.nii
++ 3dcopy: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
#++ Aligning /data/home3/lzh/temp5/test_anat/awpy/base.nii data to /data/home3/lzh/temp5/test_anat/awpy/anat.un.nii data
#Script is running (command trimmed):
@auto_tlrc -base ./base.nii -input ./anat.un.nii -suffix .aff -no_ss -no_pre -init_xform CENTER
Notice:
Input dataset is in +tlrc already, @auto_tlrc might fail.
If the space of ./anat.un.nii is incorrectly set (check with next command)
3dinfo -prefix -av_space ./anat.un.nii
you can reset it with something like:
3drefit -space ORIG ./anat.un.nii
then remove all temp files, and try again.
Copying NIFTI volume to ./anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA
++ 3dcopy: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
Performing center alignment with @Align_Centers
++ 3dcopy: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
++ 3drefit: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
++ Authored by: RW Cox
++ Processing AFNI dataset ./anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc
+ deoblique
++ 3drefit processed 1 datasets
++ 3drename: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
** THD_rename_dataset_files: old header anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+orig.HEAD doesn't exist!
** THD_rename_dataset_files: old header anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+acpc.HEAD doesn't exist!
++ THD_rename_dataset_files: rename anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.HEAD -> __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.HEAD
++ THD_rename_dataset_files: rename anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.BRIK -> __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.BRIK
++ 3dWarp: AFNI version=AFNI_24.2.02 (Aug 16 2024) [64-bit]
++ Authored by: RW Cox
++ anat already stripped, re-using
Error: Anatomical dset pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD not found.
#**ERROR Failed in affine step
** ERROR - script failed
But if I list the content of "awpy", the claimed missing file does exist:
[lzh@master2 test_anat]$ ls -l awpy/
total 145924
-rw-r--r--. 1 lzh lzh 8533509 Jan 19 10:57 anat.nii
-rw-r--r--. 1 lzh lzh 34123716 Jan 19 10:58 anat.un.nii
-rw-r--r--. 1 lzh lzh 244 Jan 19 10:58 anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft.1D
-rw-r--r--. 1 lzh lzh 34120084 Jan 19 10:58 anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.BRIK
-rw-r--r--. 1 lzh lzh 3036 Jan 19 10:58 anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD
-rw-r--r--. 1 lzh lzh 34120084 Jan 19 10:58 __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.BRIK
-rw-r--r--. 1 lzh lzh 3404 Jan 19 10:58 __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.HEAD
-rw-r--r--. 1 lzh lzh 4375336 Jan 19 10:58 base.nii
-rw-r--r--. 1 lzh lzh 34120084 Jan 19 10:58 pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.BRIK
-rw-r--r--. 1 lzh lzh 3858 Jan 19 10:58 pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD
I see this error/warning message in line 3 out the output text there:
"%s -prefix %s\_REML%s\n" \
I checked in the repository about that, and that was actually fixed in this PR on Sept 30, 2024.
I am running your command now to see if there are any issues still (my code version is modern enough to include that fix), and so far things are OK (into lev=4 of the warping).
I will let you know if any other error pops up, but I suspect if you update your local AFNI version, then this issue should be resolved.
--pt
ps: there is this warning in the text terminal, as well:
Input dataset is in +tlrc already, @auto_tlrc might fail.
... and i had wondered if this would present a problem, but we will see...
Thank you Paul, for your test and guidance! My system administrator has updated AFNI for me, but the same issue still persists. Could you please take another look at the following outputs? Thank you so much for your help!
Check AFNI version:
[lzh@master2 test_anat]$ which afni
/usr/local/afni_OpenMP/linux_openmp_64/afni
[lzh@master2 test_anat]$ afni -ver
Precompiled binary linux_openmp_64: Jan 14 2025 (Version AFNI_25.0.00 'Severus Alexander')
Replicate the issue:
[lzh@master2 test_anat]$ auto_warp.py -base MNI_caez_N27+tlrc -input ./MNI152_T1_2009c+tlrc
#++ auto_warp.py version: 0.06
# Output directory /data/home3/lzh/temp5/test_anat/awpy/
#Script is running:
mkdir /data/home3/lzh/temp5/test_anat/awpy/
cd /data/home3/lzh/temp5/test_anat/awpy/
#Script is running (command trimmed):
3dcopy /data/home3/lzh/temp5/test_anat/MNI152_T1_2009c+tlrc ./anat.nii
++ 3dcopy: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
#Script is running (command trimmed):
3dUnifize -GM -input ./anat.nii -prefix ./anat.un.nii
++ 3dUnifize: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
+ Pre-processing: ADV.....UW[s1]Gm
++ Output dataset ./anat.un.nii
++ ===== Elapsed = 10.5 sec
#Script is running (command trimmed):
3dcopy /usr/local/afni_OpenMP/linux_openmp_64/MNI_caez_N27+tlrc ./base.nii
++ 3dcopy: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
#++ Aligning /data/home3/lzh/temp5/test_anat/awpy/base.nii data to /data/home3/lzh/temp5/test_anat/awpy/anat.un.nii data
#Script is running (command trimmed):
@auto_tlrc -base ./base.nii -input ./anat.un.nii -suffix .aff -no_ss -no_pre -init_xform CENTER
Notice:
Input dataset is in +tlrc already, @auto_tlrc might fail.
If the space of ./anat.un.nii is incorrectly set (check with next command)
3dinfo -prefix -av_space ./anat.un.nii
you can reset it with something like:
3drefit -space ORIG ./anat.un.nii
then remove all temp files, and try again.
Copying NIFTI volume to ./anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA
++ 3dcopy: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
Performing center alignment with @Align_Centers
++ 3dcopy: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
++ 3drefit: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
++ Authored by: RW Cox
++ Processing AFNI dataset ./anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc
+ deoblique
++ 3drefit processed 1 datasets
++ 3drename: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
** THD_rename_dataset_files: old header anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+orig.HEAD doesn't exist!
** THD_rename_dataset_files: old header anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+acpc.HEAD doesn't exist!
++ THD_rename_dataset_files: rename anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.HEAD -> __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.HEAD
++ THD_rename_dataset_files: rename anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.BRIK -> __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.BRIK
++ 3dWarp: AFNI version=AFNI_25.0.00 (Jan 14 2025) [64-bit]
++ Authored by: RW Cox
++ anat already stripped, re-using
Error: Anatomical dset pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD not found.
#**ERROR Failed in affine step
** ERROR - script failed
Check the "awpy" folder:
[lzh@master2 test_anat]$ ls -l awpy/
total 145924
-rw-r--r--. 1 lzh lzh 8533509 Jan 25 10:52 anat.nii
-rw-r--r--. 1 lzh lzh 34123716 Jan 25 10:52 anat.un.nii
-rw-r--r--. 1 lzh lzh 244 Jan 25 10:52 anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft.1D
-rw-r--r--. 1 lzh lzh 34120084 Jan 25 10:52 anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.BRIK
-rw-r--r--. 1 lzh lzh 3039 Jan 25 10:52 anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD
-rw-r--r--. 1 lzh lzh 34120084 Jan 25 10:52 __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.BRIK
-rw-r--r--. 1 lzh lzh 3409 Jan 25 10:52 __ats_tmp__anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA_shft+tlrc.HEAD
-rw-r--r--. 1 lzh lzh 4375544 Jan 25 10:52 base.nii
-rw-r--r--. 1 lzh lzh 34120084 Jan 25 10:52 pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.BRIK
-rw-r--r--. 1 lzh lzh 3862 Jan 25 10:52 pre.anat.un_XYZ_OYmeqfZ3JqF7n9LNC7gDAA+tlrc.HEAD
I'm not seeing this problem either, but I have only a couple comments that might help:
This problem looks like it's trying to read a file before the file system has completely written it out from a previous step. We have seen problems like this pretty rarely, but it happens. Just in case, you might add this option to auto_warp.py - Run auto_warp.py with "-at_opts -keep_tmp" . You can also add a "sleep 2" to wait two seconds in the @auto_tlrc script just before the error.
OK_notice:
sleep 2
if ( @CheckForAfniDset ${anat_in} != 2 ) then
echo "Error: Anatomical dset ${anat_in} not found."
set estat = 1
goto END
endif
It looks like you want to align one template to another. The N27 template is the same as the one you use, but the MNi_2009c template we use more often is the MNI152_2009_template.nii.gz. Those templates are essentially the same except the latter one has been unifized, masked , and it's in NIFTI format. _SS version has additional volumes for masking. You might try that template instead. We no longer distribute that original version, but it is on our website, so I tried, and that works fine too.
Between these two "MNI" templates, you could try skipping the affine alignment completely; they should be close to start. With auto_warp.py, you would use -skip_affine. Otherwise, just use 3dQwarp to compute only the nonlinear part of the alignment.
For the transformations between the two templates, I think I had done this myself a while ago, but I can't find it on our website. I can redo it, and post the data, if that's useful.
Hope that helps. Sorry that this is more complicated than it should be.
The
National Institute of Mental Health (NIMH) is part of the National Institutes of
Health (NIH), a component of the U.S. Department of Health and Human
Services.