"Too many total number of fixed regressors" error in 3dTproject

AFNI Message Board
1

Hi, all

I am new to AFNI and I am trying to perform some resting-fMRI functional connectivity analysis. I used uber_subject.py to process the data first, however it generated an error “0m total number of fixed regressors (166) is too many for 235 retained time points!” during the 3dTproject program.

In uber_subject.py, “analysis initialization” is checked, “type” is set to “rest”, processing blocks is set to “tshift align tlrc volreg blur mask scale regress”; “stimulus timing files” is unchecked because I’m processing resting-state data, “expected options” are default except that the “volume registration base” is choosen to be the first one, “extra regress options” is unchecked, “extra align options is checked” and under it I also checked “align: use giant_move”. I also choose the tlrc template to be “MNI_avg152T1+tlrc”.

[size=large]This is the error part of the output file:[/size]
3dREMLfit -matrix X.xmat.1D -input pb04.par2.r01.blur+tlrc.HEAD
-fout -tout -Rbuck stats.par2_REML -Rvar stats.par2_REMLvar
-Rfitts fitts.par2_REML -Rerrts errts.par2_REML -verb

++ N.B.: 3dREMLfit command above written to file stats.REML_cmd
++ (b) Visualization/analysis of the matrix via ExamineXmat.R
++ (c) Synthesis of sub-model datasets using 3dSynthesize
++ ==========================================================
++ Wrote matrix values to file X.nocensor.xmat.1D
++ ----- Signal+Baseline matrix condition [X] (166x165): 2650.6 ++ OK ++
++ ----- Baseline-only matrix condition [X] (166x165): 2650.6 ++ OK ++
++ ----- stim_base-only matrix condition [X] (166x160): 1967.76 ++ OK ++
++ ----- polort-only matrix condition [X] (166x5): 1.14995 ++ VERY GOOD ++
++ 3dDeconvolve exits: -x1D_stop option was invoked
3dTproject -polort 0 -input pb04.par2.r01.blur+tlrc.HEAD -censor motion_par2_censor.1D -cenmode ZERO -ort X.nocensor.xmat.1D -prefix errts.par2.tproject
++ 3dTproject: AFNI version=AFNI_17.0.07 (Feb 7 2017) [64-bit]
++ Authored by: Cox the Algebraic (Linear)
++ input time points = 235 ; censored = 69 ; remaining = 166
++ Setting up regressors
++ 1 Blocks * 1 polynomials – 1 polort regressors

  • – 165 other fixed ort regressors
    [7m** ERROR:[0m total number of fixed regressors (166) is too many for 235 retained time points!
    [7m** FATAL ERROR:[0m Cannot continue after above errors :frowning: :frowning: :frowning: !!
    ** Program compile date = Feb 7 2017

Anyone has some ideas of what is going on?
I can’t put the entire output file here because it is too long, but I am happy to provide any further information if needed.

2

There is a slight error in 3dTproject – not in the algorithm (as far as I know) but in the error message you got.

It says “235 retained time points” – that is, the number left after censoring – but 235 is the TOTAL number of time points, and 166 is the number left after censoring. But the comparison that produces the error is (correctly) between the number of regressors being removed from the data and the number of un-censored time points. Which are equal here – which is bad. After the regression, there would be nothing left of the data. That is why the program stops.

I can’t see the rest of your command that gave this, but I guess that you are doing bandpassing. That is where most of the regressors come from (I guess). You only have 2 options: (1) don’t do bandpassing or at least leave a bigger pass band, or (2) don’t use this subject, which had a lot of censoring.