Hi Everyone,

I'm trying to run resting-state analyses and have used information here: [afni.nimh.nih.gov] to get some basic ideas I understand that the website was not specifically for resting state data so I was hoping to get some feedback about how to run the analyses. In particular, I want to make sure I'm extracting signal from the nuisance variables correctly.

I've run basic preprocessing on my data (including bandpass filter) and the data have been warped to template space. I have CSF and white matter masks generated using other programs. Then, I have the following:
---
3dDeconvolve -input Rest_bp+tlrc \
-mask t1h+tlrc.nii \
-GOFORIT 4 \
-polort 1 -num_stimts 8 \
-stim_file 1 Median_B2_CSF.1D -stim_label 1 Median_CSF \
-stim_file 2 Median_WM.1D -stim_label 2 Median_WM \
-stim_file 3 Rest_mt.1D[1] -stim_label 3 roll -stim_base 3 \
-stim_file 4 Rest_mt.1D[2] -stim_label 4 pitch -stim_base 4 \
-stim_file 5 Rest_mt.1D[3] -stim_label 5 yaw -stim_base 5 \
-stim_file 6 Rest_mt.1D[4] -stim_label 6 dS -stim_base 6 \
-stim_file 7 Rest_mt.1D[5] -stim_label 7 dL -stim_base 7 \
-stim_file 8 Rest_mt.1D[6] -stim_label 8 dP -stim_base 8 \
-full_first \
-tout -rout -fout -bucket Rest_bp_Decon+tlrc -fitts full_fitts_Decon+tlrc -errts errts_Decon+tlrc

3dcalc -float -a Rest_bp+tlrc -b errts_Decon+tlrc -expr 'a-b' -prefix Rest_bp_clean_Decon+tlrc
---

Is it correct to use the errts data as reflecting the variance not accounted for by white matter, CSF, and motion to reflect the "true" signal? If so, then I assume the next 3dcalc step (which subtracts errts from the Rest_bp+tlrc) is also ok, right?

Thanks for your help.

Jatin