Dear FreeSurfer team,
We are investigating if there is a difference in cortical thicknesses between persons with a condition and controls, regressing out the effect of site, sex, age, but have some concerns on how to specify our contrasts. We don’t think that the age-slopes will be different between the patients and the controls but would like to check that formally using DODS before trying DOSS. Currently, we have only one female control from two of the sites and are left with 10 instead of 12 classes. Would be very grateful if you have time to check that we’ve got this right.
Also, when we want to check for interactions between class and age to see if we can use DOSS instead, is the attached “interactionmatrix” correct? I’ve looked at the tutorials and this examplehttps://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg49602.html, but tried to adapt it to our situation. If we do find an interaction, would that cluster then show up on the cluster summary in the glmdir-folder?
Thank you and sorry for all the questions, just wanting to make sure we are doing this correctly.
Kind regards,
Carl Trolle Research Fellow Spaulding Neuroimaging Lab
FreeSurfer version: freesurfer-linux-centos7_x86_64-7.1.1-20200723-8b40551 Platform: CentOS Stream 8
Yes, that all looks correct. Good job on a complicated design!
On 9/19/2022 12:32 PM, Trolle, Carl wrote:
Dear FreeSurfer team,
We are investigating if there is a difference in cortical thicknesses between persons with a condition and controls, regressing out the effect of site, sex, age, but have some concerns on how to specify our contrasts. We don’t think that the age-slopes will be different between the patients and the controls but would like to check that formally using DODS before trying DOSS. Currently, we have only one female control from two of the sites and are left with 10 instead of 12 classes. Would be very grateful if you have time to check that we’ve got this right.
Also, when we want to check for interactions between class and age to see if we can use DOSS instead, is the attached “interactionmatrix” correct? I’ve looked at the tutorials and this example https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg49602.html, but tried to adapt it to our situation. If we do find an interaction, would that cluster then show up on the cluster summary in the glmdir-folder?
Thank you and sorry for all the questions, just wanting to make sure we are doing this correctly.
Kind regards,
Carl Trolle
Research Fellow
Spaulding Neuroimaging Lab
FreeSurfer version: freesurfer-linux-centos7_x86_64-7.1.1-20200723-8b40551
Platform: CentOS Stream 8
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Douglas N. Greve -
Trolle, Carl