help
help > RE: Question about the extract raw timeseries
Sep 5, 2021 08:09 PM | Alfonso Nieto-Castanon - Boston University
RE: Question about the extract raw timeseries
Dear YJ,
I imagine you are referring to the data/ROI_Subject#_Session#.mat files and not to the preprocessing/ROI_Subject#_Condition#.mat files? If that is the case, then the PCA components extracted from WM/CSF do not change with different conditions (they will change with different first-level covariates if you are using the default 'regress-out covariates' option in Setup.ROIs, so perhaps those covariates have changed in your project?). Alternatively, if you are referring to the preprocessing/ROI_Subject#_Condition#.mat files then those timeseries are not the raw data, but the data after preprocessing, so those will typically change with different condition definitions (as condition regressors are typically included as part of the confounding terms during the denoising step)
Hope this helps
Alfonso
Originally posted by ys1j13:
I imagine you are referring to the data/ROI_Subject#_Session#.mat files and not to the preprocessing/ROI_Subject#_Condition#.mat files? If that is the case, then the PCA components extracted from WM/CSF do not change with different conditions (they will change with different first-level covariates if you are using the default 'regress-out covariates' option in Setup.ROIs, so perhaps those covariates have changed in your project?). Alternatively, if you are referring to the preprocessing/ROI_Subject#_Condition#.mat files then those timeseries are not the raw data, but the data after preprocessing, so those will typically change with different condition definitions (as condition regressors are typically included as part of the confounding terms during the denoising step)
Hope this helps
Alfonso
Originally posted by ys1j13:
Dear Alfonso,
as I understand the variable "data" in the "data/ROI_Subject#_Condition#.mat" contains the extract raw timeseries, where data{1,1} for mean signal from grey matter; data{1,2} for PCs for White matte; data{1,3} for PCs for CSF.
However, I recently found that the timeseries in the variable "data" changes when I change the conditions in SETUP.
(particularly, the 2nd-16th PCs for the WM and the CSF are changed while the first PCs remain same).
Is there any reason why the raw timeseries is changed by the condition?
As I understood before, the effect of condition should regressed out when the denoising step.
Is the effect of condition also regressed out in the first ROI timeseries extraction?
Thank you.
Best regards,
YJ
as I understand the variable "data" in the "data/ROI_Subject#_Condition#.mat" contains the extract raw timeseries, where data{1,1} for mean signal from grey matter; data{1,2} for PCs for White matte; data{1,3} for PCs for CSF.
However, I recently found that the timeseries in the variable "data" changes when I change the conditions in SETUP.
(particularly, the 2nd-16th PCs for the WM and the CSF are changed while the first PCs remain same).
Is there any reason why the raw timeseries is changed by the condition?
As I understood before, the effect of condition should regressed out when the denoising step.
Is the effect of condition also regressed out in the first ROI timeseries extraction?
Thank you.
Best regards,
YJ
Threaded View
| Title | Author | Date |
|---|---|---|
| ys1j13 | Aug 25, 2021 | |
| Alfonso Nieto-Castanon | Sep 5, 2021 | |