Hello,
I’m looking to use the ClustSim option in 3dttest++ to do some corrections for partial axial coverage resting state 7T data. I keep getting an error that says “dataset grid too big–must recompile to use shorts as indexes” and then the program crashes. Do you think this has to do with the fact that this is 7T data?
My command was:
3dttest++ -setA AMY_CEN {insert subjects here} -mask func_mask.nii -ClustSim -covariates Q_scores_female.1D -prefix AMY_CEN_seed_female_ClustSimTest.nii
*The AMY_CEN_seed_female_ClustSimTest.nii file is output and has the covariate data. I also see some of the intermediate .sdat files and have the *.5percent.txt file too.
The terminal output is:
++ 3dttest++: AFNI version=AFNI_17.2.09 (Aug 22 2017) [64-bit]
++ Authored by: Zhark++
++ option -setA :: processing as LONG form (label label dset label dset …)
++ 3051791 voxels in -mask dataset
++ Number of -Clustsim threads set to 10
- Default clustsim prefix set to ‘TT.8ZkNZHzA25g’
++ Covariates file: 3 columns, each with 21 rows - Covariate column labels: ID;IUS;STAI
- Found 2 numeric columns: IUS STAI
++ loading covariates
++ loading -setA datasets
*+ WARNING: 3dttest++ -setA :: 133 vectors are constant
++ Memory usage now = 641,076,188 (641 million [mega])
++ t-testing:0123456789.0123456789.0123456789.0123456789.0123456789.! - skipped 133 voxels completely for having constant data
++ ---------- End of analyses – freeing workspaces ----------
++ Memory usage now = 1,874,289,848 (1.9 billion [giga])
++ Creating FDR curves in output dataset
++ Smallest FDR q [1 AMY_CEN_Zscr] = 1.15845e-09
*+ WARNING: Smallest FDR q [3 AMY_CEN_IUS_Zscr] = 0.918285 ==> few true single voxel detections
*+ WARNING: Smallest FDR q [5 AMY_CEN_STAI_Zscr] = 0.999897 ==> few true single voxel detections - Added 3 FDR curves to dataset
++ Output dataset ./AMY_CEN_seed_female_ClustSimTest.nii
++ Output dataset ./TT.8ZkNZHzA25g.resid.nii
++ cprefix[0] = ./AMY_CEN_seed_female_ClustSimTest.nii
++ ================ Starting -ClustSim calculations ================ - === temporary files will have prefix TT.8ZkNZHzA25g ===
- === running 10 -randomsign jobs (1000 iterations per job) ===
- === creating 61,035,820,000 (61 billion [giga]) bytes of pseudo-data in .sdat files ===
- — 3dClustSim reads .sdat files to compute cluster-threshold statistics —
- — there is 67,662,475,264 (68 billion [giga]) bytes of memory on your system —
*+ WARNING: — runs may be slow (or crash) due to memory needs
—
++ 3dttest++: AFNI version=AFNI_17.2.09 (Aug 22 2017) [64-bit]
++ Authored by: Zhark++
++ 3051791 voxels in -mask dataset
++ Covariates file: 3 columns, each with 21 rows - Covariate column labels: ID;IUS;STAI
- Found 2 numeric columns: IUS STAI
++ option -setA :: processing as LONG form (label label dset label dset …)
++ random seeds are 1637391400 1951550665
++ loading covariates
++ opened file ./TT.8ZkNZHzA25g.0000.sdat for output
++ loading -setA datasets
*+ WARNING: 3dttest++ -setA :: 133 vectors are constant
++ Memory usage now = 406,866,408 (407 million [mega])
++ t-test randomsign:0123456789.0123456789.0123456789.0123456789.0123456789.!
++ saving main effect t-stat MIN/MAX values in ./TT.8ZkNZHzA25g.0000.minmax.1D
++ output short-ized file ./TT.8ZkNZHzA25g.0000.sdat - ===== simulation jobs have finished (22557.1 s elapsed) =====
++ Global % FPR points for simulated z-stats:
4.967 = 1-sided 9% FPR A
5.164 = 2-sided 9% FPR A
5.005 = 1-sided 8% FPR A
5.196 = 2-sided 8% FPR A
5.046 = 1-sided 7% FPR A
5.234 = 2-sided 7% FPR A
5.088 = 1-sided 6% FPR A
5.269 = 2-sided 6% FPR A
5.137 = 1-sided 5% FPR A
5.313 = 2-sided 5% FPR A
5.196 = 1-sided 4% FPR A
5.370 = 2-sided 4% FPR A
5.269 = 1-sided 3% FPR A
5.433 = 2-sided 3% FPR A
5.370 = 1-sided 2% FPR A
5.513 = 2-sided 2% FPR A
5.513 = 1-sided 1% FPR A
5.671 = 2-sided 1% FPR A -
[above results also in file TT.8ZkNZHzA25g.A.5percent.txt] - ===== starting 3dClustSim A: elapsed = 22603.1 s =====
++ 3dClustSim: AFNI version=AFNI_17.2.09 (Aug 22 2017) [64-bit]
++ Authored by: RW Cox and BD Ward
++ Loading -insdat datasets
** FATAL ERROR: dataset grid too big – must recompile to use shorts as indexes
** Program compile date = Aug 22 2017
** FATAL ERROR: ===== 3dClustSim command failed :-((( =====
** Program compile date = Aug 22 2017
Thanks for your help!
Carissa