compute_ROI_stats.tcsh created without -regress_compute_tsnr_stats

AFNI Message Board
1

Hello,

I got a chunk computing TSNR in the brain in the process code generated by afni_proc.py. It may be created by -regress_compute_tsnr_stats, however, I did not give this option.

# compute TSNR stats for dset labels: brain
compute_ROI_stats.tcsh                                           \
    -out_dir    tsnr_stats_regress                               \
    -dset_ROI   mask_epi_anat.$subj+orig                         \
    -dset_data  TSNR.$subj+orig                                  \
    -rset_label brain                                            \
    -rval_list  1

It also returned an error as follows,

compute_ROI_stats.tcsh -out_dir tsnr_stats_regress -dset_ROI mask_epi_anat.sub-ep2168+orig -dset_data TSNR.sub-ep2168+orig -rset_label brain -rval_list 1
++ writing stats text to tsnr_stats_regress/stats_brain.txt
** failed to get depth extrema from command:
   3dExtrema -volume -nbest 1 -closure -partial -quiet -mask_file mask_epi_anat.sub-ep2168+orig<1> tsnr_stats_regress/depth_brain.nii.gz

   6 vals: 1 55.450 -2.72 7.93 8.00 1

I don't have this chunk in previous works, it appears after updating afni to the latest version.

Here is the afni_proc.py script. Could you please take a look at what is wrong?

# ==========================================================================
# script generated by the command:
#
# afni_proc.py -subj_id sub-ep2168 -out_dir                                                        \
#     /sub-ep2168/OC_dmUBLOCK_orig_dm                         \
#     -script                                                                                      \
#     /sub-ep2168/script/proc.sub-ep2168_OC_dmUBLOCK_orig_dm  \
#     -blocks despike tshift align volreg mask combine blur scale regress                          \
#     -radial_correlate_blocks tcat volreg -blip_forward_dset                                      \
#     /sub-ep2168/fmap/sub-ep2168_dir-AP_epi.nii.gz                            \
#     -blip_reverse_dset                                                                           \
#     /sub-ep2168/fmap/sub-ep2168_dir-PA_epi.nii.gz                            \
#     -copy_anat                                                                                   \
#     /sub-ep2168/anat/anatSS.sub-ep2168.nii                  \
#     -anat_has_skull no -anat_follower anat_w_skull anat                                          \
#     /sub-ep2168/anat/sub-ep2168_T1w.nii.gz                                   \
#     -anat_follower_ROI FS_REN_epi epi                                                            \
#     /sub-ep2168/SUMA/aparc.a2009s+aseg_REN_all.nii.gz \
#     -dsets_me_echo                                                                               \
#     /sub-ep2168/func/sub-ep2168_task-ied_run-2_echo-1_bold.nii.gz            \
#     -dsets_me_echo                                                                               \
#     /sub-ep2168/func/sub-ep2168_task-ied_run-2_echo-2_bold.nii.gz            \
#     -dsets_me_echo                                                                               \
#     /sub-ep2168/func/sub-ep2168_task-ied_run-2_echo-3_bold.nii.gz            \
#     -reg_echo 2 \
#     -echo_times 11.4 24.69 37.98 \
#     -combine_method OC                                  \
#     -tcat_remove_first_trs 2 \
#     -tshift_interp -wsinc9 \
#     -volreg_align_to     MIN_OUTLIER \
#     -volreg_align_e2a \
#     -volreg_post_vr_allin yes                                      \
#     -volreg_pvra_base_index MIN_OUTLIER \
#     -volreg_warp_final_interp wsinc5                         \
#     -align_opts_aea \
#     -check_flip -cost lpc+ZZ -giant_move -AddEdge                                \
#     -align_unifize_epi local \
#     -mask_epi_anat yes \
#     -mask_opts_automask -clfrac 0.1 -dilate 1 \
#     -blur_in_mask yes \
#     -blur_size 6 \
#     -blur_to_fwhm                                   \
#     -regress_motion_per_run \
#     -regress_censor_motion 0.3                                           \
#     -regress_censor_outliers 0.05 \
#     -regress_apply_mot_types demean deriv                          \
#     -html_review_style pythonic

Thanks,
Zhengchen

2

Hi Zhengchen,

Sorry about that, I did not initially realize that 3dExtrema might do this. I put it a fix for this a few days ago, but we have not run a build since then.

Just to be sure, are you using build_afni.py (what is in ~/abin/AFNI_version.txt)?

We will make a new package/tag available tonight. But if you are using build_afni.py, you can update now (by including "-git_tag NONE").

Thanks,

  • rick
3

Hi Rick,

Thank you so much. Usually, I use @update.afni.binaries to update AFNI. It is good to know there will be a fix. No rush, I can easily remove this small chunk in the proc script, it runs without error.

Thanks,
Zhengchen