Hi Prapti, can you send the full terminal output? The ill-conditioned error is probably not related to the contrast. It is usually related to the design matrix, which makes it difficult to track down because you will have a different one at each vertex with --pvr. doug
On 6/13/13 1:14 PM, Gautam, Prapti wrote:
Dear FreeSurfers
I am trying to control for thickness in another surface using the pvr option in mri_glmfit. However, I'm confused as to how I can include the pvr option correctly in the contrast files. I have two groups, and would like to utilize DODS for this analyses as I think that my groups both have different intercepts and slopes.
This is my fsgd set up :
GroupDescriptorFile1 TitleMyTitle ClassControls CLASS FA VariablesAge InputPMS1Controls14.442163 InputPMS2Controls14.354552 .. InputPMS7FA15.726215 InputPMS28FA14.45859
This is the command I run: mri_glmfit --y lh.surface.mgh --fsgd SA_age.fsgd --C con.mtx --surf fsaverage lh --cortex --glmdir test.glmdir --pvr lh.thickness.psc.mgh --pvr lh.thickness2.psc.mgh
The contrast I would like to run is: 0 0 0 0 1 -1 so as to look at group interactions within thickness after controlling for another cortical surface. However, I can't get this to run as there is a "Matrix ill-conditioned" error.
The contrast that does run is 0 0 0 0 1 (after passing only one pvr file). But this option seems to be neither dods nor doss. Is there a correct way to do a DODS analyses looking for interactions using the pvr option?
Thank you,
Prapti
CONFIDENTIALITY NOTICE: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential or legally privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of this original message.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer