History of AFNI updates  

|
April 16, 2009 01:23PM
We are having a problem running 3dDeconvolve in our functional connectivity analysis. I have 5 question, that I have labeled below. I apologize in advance for the length (and possible incoherence).

To start, originally I ran 3dDeconvolve using a “local times” duration modulated regressor format AND my regressor files were .txt files (I see in examples that .1d files are used, (1)should I have performed a preprocessing step to convert my txt files to .1d files? (2)What would this step look like?). (3)Should I use the waver function on the local times duration modulated regressor files in the .txt format? (4)Does this work? [I’ve included an example of my dmBLOCK regressor files and my 3dDecon script, in case I have missed something]

I ask these question because when we create the functional connectivity regressor files, they are in a “global times” format, and are comprised of 1/0/-1. Thus, when we run 3dDecon for the second time with the inclusion of the functional connectivity regressors we get a fatal error (I assume it’s because I specify –local_times and obviously the functional connectivity regressors are not in a local time format).

From what I understand (which is very little), we have to perform the waver function on this regressor file, which converts it into a .1D file, and which is still in a global time format (a single column). (5)My idea is to use the make_stim_times.py to convert the functional connectivity regressor .txt file into a “local times” format, THEN perform the waver. But this gets back to the question of whether waver will work on local time regressors. I tried using the make_stim_times.py on the .1D regressor file once it had been “waver-ed” but this created a file of “*” (So I know that’s not a solution) .

I hope this post is at least somewhat coherent. Help would be great! Thanks.

3dDeconvolve -input pb04.1086*.BRIK \
-num_stimts 16 \
-polort 3 \
-local_times \
-stim_times_AM1 1 Att.Neg.txt 'dmBLOCK' -stim_label 1 'atneg’ \
-stim_times_AM1 2 Att.Pos.txt 'dmBLOCK' -stim_label 2 'atpos' \
-fout –tout

Sample Regressor file:
1:1 3:1.5 6.8:0.7
*
2:1.2 7:0.9

Subject Author Posted

Functional Connectivity Decon with local times

Steve Greening April 16, 2009 01:23PM

Re: Functional Connectivity Decon with local times

Gang Chen April 16, 2009 06:34PM

Re: Functional Connectivity Decon with local times

Steve Greening April 17, 2009 10:57AM

Re: Functional Connectivity Decon with local times

Gang Chen April 17, 2009 01:42PM

Re: Functional Connectivity Decon with local times

Steve Greening April 17, 2009 02:53PM