Dear experts,

I'm having some trouble running the generalized PPI toolbox, I'm hoping you might be familiar with the error I encounter.

I get the following output when I set the contrast to the omnibus F test:

Log File: sub-01_PPPI_4_29_2020.log
PPPI Version: 13.1.4-17-2014
Parameters used:
Processing subject: sub-01
VOI file          : /data/sub-01/amygd_left_sub-01.nii
Output file will be: sub-01_amygd_left_session#_cond_PPI_regressors.txt
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Valid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
Warning: Missing conditions!!! Invalid Contrast
Invalid Contrast
ESS image  1                    : (spm_resss)   ten ess_0001.img
spm{F} image  1                 :       ...written spmF_0001.img
Contrast          : Omnibus F-test for PPI Analyses
Analysis          : Psychophysiological Interactions
Extraction        : eigenvariate
Tasks             : ALL
Method            : Condition Specific
VOI has 32 voxels in 3x3x3 space
VOI has 32 voxels in 3x3x3 space. This is in the the same space as the input data and functional mask.
Index exceeds matrix dimensions.
Error in timeseries_extract (line 369)
Error in PPPI (line 461)
Error in gppi_20200327 (line 38)
PPPI(P);

Line 369 in timeseries_extract is:
if xY.Ic~=0
y=y-spm_FcUtil('Y0',SPM.xCon(xY.Ic),SPM.xX.xKXs,beta);
end

I don't get this error when I set the contrast to 0, but if I understood correctly, it's better to use the omnibus F test here to get rid of noise, right?

I hope you can shed some light on this error, thanks a lot!

Kind regards,
Rozemarijn

Did you work out it ? I encounter the same problem of timeserise extract recently. Can you tell me how to resolve this error ?

Hi,

This isn't an error. It's simply a warning.

The warning is generated because the code is trying to generate contrasts for the temporal (1st) and dispersion (2nd) derivatives. If your design matrix does not have these, it triggers the warning message that you are missing conditions.

If you notice, there is a valid contrast, followed by 2 invalid contrasts --> task, 1st derivative, 2nd derivative contrasts that are attempted to be generated. One could probably write the code better such that it doesn't trigger warnings when the derivatives are missing.

The contrast currently only takes in the task names.

Hope this helps.

-Donald

Thank you for your careful answer !