Hello Rick,
I also have a participant who has no "other" condition. In order for him to rely on the same model as the other participants, I set the times in the stim times file to -1 and *, added -allzero_OK and -GOFORIT 10. This way, the warning messages that appear are accepted (see below):
Script:
3dDeconvolve\
-input psc.ssm.tsm.cl.dsp.hpf.reg.ts.1+orig psc.ssm.tsm.cl.dsp.hpf.reg.ts.2+orig\
-polort 5\
-num_stimts 10\
-global_times\
-stim_times_AM2 1 onsettimes_pleasant_S10.txt 'BLOCK(18,1)'\
-stim_label 1 pleasant\
-stim_times_AM2 2 onsettimes_neutral_S10.txt 'BLOCK(18,1)'\
-stim_label 2 neutral\
-stim_times_AM2 3 onsettimes_unpleasant_S10.txt 'BLOCK(18,1)'\
-stim_label 3 unpleasant\
-stim_times 4 onsettimes_other_S10.txt 'BLOCK(18,1)'\
-stim_label 4 other\
-allzero_OK\
-GOFORIT 10\
-stim_file 5 mot[1]\
-stim_base 5\
-stim_label 5 Roll\
-stim_file 6 mot[2]\
-stim_base 6\
-stim_label 6 Pitch\
-stim_file 7 mot[3]\
-stim_base 7\
-stim_label 7 Yaw\
-stim_file 8 mot[4]\
-stim_base 8\
-stim_label 8 dS\
-stim_file 9 mot[5]\
-stim_base 9\
-stim_label 9 dL\
-stim_file 10 mot[6]\
-stim_base 10\
-stim_label 10 dP\
-num_glt 6\
-gltsym 'SYM: +pleasant[0] -neutral[0]'\
-glt_label 1 pleasant_vs_neutral\
-gltsym 'SYM: +pleasant[0] -unpleasant[0]'\
-glt_label 2 pleasant_vs_unpleasant\
-gltsym 'SYM: +neutral[0] -unpleasant[0]'\
-glt_label 3 neutral_vs_unpleasant\
-gltsym 'SYM: +pleasant[1] -neutral[1]'\
-glt_label 4 pleasant_vs_neutral_modulation\
-gltsym 'SYM: +pleasant[1] -unpleasant[1]'\
-glt_label 5 pleasant_vs_unpleasant_modulation\
-gltsym 'SYM: +neutral[1] -unpleasant[1]'\
-glt_label 6 neutral_vs_unpleasant_modulation\
-fout\
-rout\
-tout\
-fitts fitts_modulation_S10\
-x1D Xmat.x1D\
-bucket decon_modulation_S10\
Messages:
++ '-stim_times_AM2 1 onsettimes_pleasant_S10.txt' has 1 auxiliary values per time point
++ '-stim_times_AM2 1 onsettimes_pleasant_S10.txt' will have 2 regressors
++ '-stim_times_AM2 2 onsettimes_neutral_S10.txt' has 1 auxiliary values per time point
++ '-stim_times_AM2 2 onsettimes_neutral_S10.txt' will have 2 regressors
++ '-stim_times_AM2 3 onsettimes_unpleasant_S10.txt' has 1 auxiliary values per time point
++ '-stim_times_AM2 3 onsettimes_unpleasant_S10.txt' will have 2 regressors
*+ WARNING: '-stim_times 4' didn't read any good times from file 'onsettimes_other_S10.txt'
++ 3dDeconvolve: AFNI version=AFNI_2010_10_19_1028 (Jan 3 2011) [64-bit]
++ Authored by: B. Douglas Ward, et al.
++ current memory malloc-ated = 113,559 bytes (about 114 thousand)
++ loading dataset psc.ssm.tsm.cl.dsp.hpf.reg.ts.1+orig psc.ssm.tsm.cl.dsp.hpf.reg.ts.2+orig
++ current memory malloc-ated = 275,373,871 bytes (about 275 million)
++ Auto-catenated datasets start at: 0 210
++ STAT automask has 22643 voxels (out of 163840 = 13.8%)
++ Input polort=5; Longest run=630.0 s; Recommended minimum polort=5 ++ OK ++
++ -stim_times using TR=3 s for stimulus timing conversion
++ -stim_times using TR=3 s for any -iresp output datasets
++ [you can alter the -iresp TR via the -TR_times option]
++ -stim_times_AM2 1 using GLOBAL times
++ '-stim_times_AM2 1' average amplitude#1=-1.2
++ -stim_times_AM2 2 using GLOBAL times
++ '-stim_times_AM2 2' average amplitude#1=-0.6
++ -stim_times_AM2 3 using GLOBAL times
++ '-stim_times_AM2 3' average amplitude#1=-1.7
++ -stim_times 4 using GLOBAL times
*+ WARNING: '-stim_times 4' (GLOBAL) has 1 '*' fillers; do you want LOCAL times?
*+ WARNING: '-stim_times 4' (GLOBAL) has 1 times outside range 0 .. 1257 [PSFB syndrome]
+ dataset TR being used is 3 s -- unusable times follow
-1
*+ WARNING: !! '-stim_times 4' file 'onsettimes_other_S10.txt' has no good stimulus time values
------------------------------------------------------------
GLT matrix from 'SYM: +pleasant[0] -neutral[0]':
0 0 0 0 0 0 0 0 0 0 0 0 1 0 -1 0 0 0 0 0 0 0 0 0 0
------------------------------------------------------------
GLT matrix from 'SYM: =pleasant[0] -unpleasant[0]':
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 -1 0 0 0 0 0 0 0 0
------------------------------------------------------------
GLT matrix from 'SYM: +neutral[0] -unpleasant[0]':
0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 -1 0 0 0 0 0 0 0 0
------------------------------------------------------------
GLT matrix from 'SYM: +pleasant[1] -neutral[1]':
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 -1 0 0 0 0 0 0 0 0 0
------------------------------------------------------------
GLT matrix from 'SYM: +pleasant[1] -unpleasant[1]':
0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 -1 0 0 0 0 0 0 0
------------------------------------------------------------
GLT matrix from 'SYM: +neutral[1] -unpleasant[1]':
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 -1 0 0 0 0 0 0 0
++ Number of time points: 420 (no censoring)
+ Number of parameters: 25 [18 baseline ; 7 signal]
++ Memory required for output bricks = 306,708,480 bytes (about 307 million)
++ Wrote matrix values to file Xmat.x1D
++ ========= Things you can do with the matrix file =========
++ (a) Linear regression with ARMA(1,1) modeling of serial correlation:
3dREMLfit -matrix Xmat.x1D \
-input "psc.ssm.tsm.cl.dsp.hpf.reg.ts.1+orig psc.ssm.tsm.cl.dsp.hpf.reg.ts.2+orig" \
-fout -tout -rout -Rbuck decon_modulation_S10_REML -Rvar decon_modulation_S10_REMLvar \
-Rfitts fitts_modulation_S10_REML -verb
++ N.B.: 3dREMLfit command above written to file decon_modulation_S10.REML_cmd
++ (b) Visualization/analysis of the matrix via ExamineXmat.R
++ (c) Synthesis of sub-model datasets using 3dSynthesize
++ ==========================================================
*+ WARNING: -------------------------------------------------
*+ WARNING: Problems with the X matrix columns, listed below:
*+ WARNING: * Column 18 [other#0] is all zeros
*+ WARNING: -------------------------------------------------
++ ----- Signal+Baseline matrix condition [X] (420x25): 35.9063 ++ GOOD ++
*+ WARNING: !! in Signal+Baseline matrix:
* Largest singular value=4.92605
* 1 singular value is less than cutoff=4.92605e-07
* Implies strong collinearity in the matrix columns!
++ Signal+Baseline matrix singular values:
-6.68213e-28 0.00382082 0.00899561 0.0129972 0.023497
0.0383778 0.104277 0.254937 0.290219 0.574994
0.657288 0.794692 0.876287 0.991463 1.00464
1.01826 1.12197 1.1444 1.22125 1.3078
1.49453 1.8355 2.02681 2.26695 4.92605
++ ----- Signal-only matrix condition [X] (420x7): 1.07342 ++ VERY GOOD ++
*+ WARNING: !! in Signal-only matrix:
* Largest singular value=1.07924
* 1 singular value is less than cutoff=1.07924e-07
* Implies strong collinearity in the matrix columns!
++ Signal-only matrix singular values:
0 0.936663 0.977748 0.993082 1.00116
1.01211 1.07924
++ ----- Baseline-only matrix condition [X] (420x18): 34.9518 ++ GOOD ++
++ ----- stim_base-only matrix condition [X] (420x6): 11.0927 ++ VERY GOOD ++
++ ----- polort-only matrix condition [X] (420x12): 1.05251 ++ VERY GOOD ++
*+ WARNING: +++++ !! Matrix inverse average error = 0.0016 ** BEWARE **
++ Matrix setup time = 0.51 s
*+ WARNING: !! 3dDeconvolve -GOFORIT is set to 10: running despite 4 matrix warnings
*+ WARNING: !! See file 3dDeconvolve.err for all WARNING and ERROR messages !!
*+ WARNING: !! Please be sure you understand what you are doing !!
*+ WARNING: !! If in doubt, consult with someone or with the AFNI message board !!
++ current memory malloc-ated = 584,422,906 bytes (about 584 million)
++ Calculations starting; elapsed time=12.703
++ voxel loop:0123456789.0123456789.0123456789.0123456789.0123456789.
++ Calculations finished; elapsed time=80.845
++ Wrote bucket dataset into ./decon_modulation_S10+orig.BRIK
+ created 35 FDR curves in bucket header
Here just the 3dDeconvolve.err content:
*+ WARNING: '-stim_times 4' didn't read any good times from file 'onsettimes_other_S10.txt'
*+ WARNING: '-stim_times 4' (GLOBAL) has 1 '*' fillers; do you want LOCAL times?
*+ WARNING: '-stim_times 4' (GLOBAL) has 1 times outside range 0 .. 1257 [PSFB syndrome]
*+ WARNING: !! '-stim_times 4' file 'onsettimes_other_S10.txt' has no good stimulus time values
*+ WARNING: -------------------------------------------------
*+ WARNING: Problems with the X matrix columns, listed below:
*+ WARNING: * Column 18 [other#0] is all zeros
*+ WARNING: -------------------------------------------------
*+ WARNING: !! in Signal+Baseline matrix:
* Largest singular value=4.92605
* 1 singular value is less than cutoff=4.92605e-07
* Implies strong collinearity in the matrix columns!
*+ WARNING: !! in Signal-only matrix:
* Largest singular value=1.07924
* 1 singular value is less than cutoff=1.07924e-07
* Implies strong collinearity in the matrix columns!
*+ WARNING: +++++ !! Matrix inverse average error = 0.0016 ** BEWARE **
*+ WARNING: !! 3dDeconvolve -GOFORIT is set to 10: running despite 4 matrix warnings
*+ WARNING: !! See file 3dDeconvolve.err for all WARNING and ERROR messages !!
*+ WARNING: !! Please be sure you understand what you are doing !!
*+ WARNING: !! If in doubt, consult with someone or with the AFNI message board !!
Would you think it better to keep this way of deconvolution for this subject or simply do another model without the "other" condition?
Best,
Jana
