AFNI Message Board

Dear AFNI users-

We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:

https://discuss.afni.nimh.nih.gov

Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.

The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.

Sincerely, AFNI HQ

History of AFNI updates  

|
November 18, 2009 05:13PM
Dear All,

I am attempting to extract the true HRF from a data set for some ROIs (or is it RsOI?) and am having difficulty.
I included the -iresp flag in my deconvolution( appended), but what I believe this outputted was the prespecified Rmodel mutiplied by the deconvolved magnitude: the HRF is identical for every condition and I did pre-specify the HRF using stim_times & BLOCK(4,1).
I believe you can use a variable shape HRF using -stimfile with -min/maxlag but also came across this:
[afni.nimh.nih.gov]
"Use 3dDeconvolve's -stim_times option with TENT(0,17.5,8) as the basis function, for example. "
Unfortunately, the link there is broken so I'm not completely clear on what it does.

So, is this the right way to get what I want (i.e. replace BLOCK with TENT and use the -iresp flag)?

Even though I used BLOCK, my stimulus is such that I want the HRF starting 4 secs after the times specified in stim_times, i.e. the end of the stimulus. How should I go about accounting for that in TENT?

Is there any downside to using a larger time domain, e.g. if my TR is 2, why not use TENT(0, 20, 10)?

Also, given that I'm interested in the combo of certain conditions (see the gltsym commands) is there a way to do that with -iresp instead of having to only refer to stim_lables?

Finally, I wanted to confirm that when I apply 3dmaskave to the -iresp output files that each row corresponds to the ROI magnitude at each time point, and that each row reflects one TR.



3dDeconvolve \
-input test_reg_blur6mm_scaled_detrended_resample3+orig \
-polort 16 -nfirst 0 \
-num_stimts 12 \
-stim_times 1 ../../1D/test-dpe.1D 'BLOCK(4,1)' -stim_label 1 dpe \
-stim_times 2 ../../1D/test-dpi.1D 'BLOCK(4,1)' -stim_label 2 dpi \
-stim_times 3 ../../1D/test-de.1D 'BLOCK(4,1)' -stim_label 3 de \
-stim_times 4 ../../1D/test-di.1D 'BLOCK(4,1)' -stim_label 4 di \
-stim_times 5 ../../1D/test-pe.1D 'BLOCK(4,1)' -stim_label 5 pe \
-stim_times 6 ../../1D/test-pi.1D 'BLOCK(4,1)' -stim_label 6 pi \
-stim_file 7 'motion[1]' -stim_label 7 roll \
-stim_file 8 'motion[2]' -stim_label 8 pitch \
-stim_file 9 'motion[3]' -stim_label 9 yaw \
-stim_file 10 'motion[4]' -stim_label 10 IS \
-stim_file 11 'motion[5]' -stim_label 11 RL \
-stim_file 12 'motion[6]' -stim_label 12 AP \
-gltsym 'SYM: +dpe -dpi' -glt_label 1 dpe_dpi \
-gltsym 'SYM: +dpe +de +pe -dpi -di -pi' -glt_label 2 e_i \
-gltsym 'SYM: +pe -di' -glt_label 3 pe_pi \
-gltsym 'SYM: +dpe -di' -glt_label 4 dpe_di \
-gltsym 'SYM: +dpe +de +pe' -glt_label 5 e \
-gltsym 'SYM: +dpi +di +pi' -glt_label 6 i \
-iresp 1 HRF-dpe-Test.irf \
-iresp 2 HRF-dpi-Test.irf \
-iresp 3 HRF-de-Test.irf \
-iresp 4 HRF-di-Test.irf \
-iresp 5 HRF-pe-Test.irf \
-iresp 6 HRF-pi-Test.irf \
-bucket MORL-Test-HRF
Subject Author Posted

Extracting times series data for IR from data

Marc Ettlinger November 18, 2009 05:13PM

Re: Extracting times series data for IR from data

ziad November 18, 2009 11:14PM

Re: Extracting times series data for IR from data

Marc Ettlinger November 19, 2009 10:45AM

Re: Extracting times series data for IR from data

ziad November 19, 2009 10:51AM

Re: Extracting times series data for IR from data

Marc Ettlinger November 20, 2009 05:20PM

Re: Extracting times series data for IR from data

David Perlman November 20, 2009 05:45PM

Re: Extracting times series data for IR from data

Marc Ettlinger November 23, 2009 10:13AM