posting the response to the list: The problem is that femalezero and femaletwo only have 4 and 3 subjects, respectively. As you are trying to estimate 4 slopes + an intercept, you will need at least 5 subjects per group.
doug
On 08/16/2017 03:27 PM, Sullivan, Danielle Renee wrote:
Hi Doug,
I am just following up to this email. If you prefer I can repost on the FreeSurfer mail archive.
Thanks,
Danielle R. Sullivan (Miller), Ph.D.
National Center for PTSD
VA Boston Healthcare System Jamaica Plain
Boston University School of Medicine
OFFICE: (857) 364-4022
*From:* Sullivan, Danielle Renee *Sent:* Friday, August 11, 2017 1:33:03 PM *To:* Douglas N Greve *Subject:* Re: [Freesurfer] Matrix ill-conditioned error for dods thickness mri_glmfit analysis
I am still getting the same error. My apologies for the back and forth.
Here are updated files (now there are 5 classes and contrast mtx has been changed to reflect that).
Danielle R. Sullivan (Miller), Ph.D.
National Center for PTSD
VA Boston Healthcare System Jamaica Plain
Boston University School of Medicine
OFFICE: (857) 364-4022
*From:* Douglas N Greve greve@nmr.mgh.harvard.edu *Sent:* Friday, August 11, 2017 1:10:53 PM *To:* Sullivan, Danielle Renee *Subject:* Re: [Freesurfer] Matrix ill-conditioned error for dods thickness mri_glmfit analysis remove the definition too
On 08/11/2017 12:46 PM, Sullivan, Danielle Renee wrote:
Hi,
Thank you for the tips. I tried doing both and still run into the error (although now I get a warning that the class femaleone is defined but not used). Attached is my new FSGD (now all variables normalized via a Z-score), XG.mtx, Contrast, and Terminal output.
Thanks again for the help!
Danielle R. Sullivan (Miller), Ph.D.
National Center for PTSD
VA Boston Healthcare System Jamaica Plain
Boston University School of Medicine
OFFICE: (857) 364-4022
*From:* Douglas N Greve greve@nmr.mgh.harvard.edu *Sent:* Friday, August 11, 2017 11:12:21 AM *To:* Sullivan, Danielle Renee *Subject:* Re: [Freesurfer] Matrix ill-conditioned error for dods thickness mri_glmfit analysis there appear to be two problems. One is that you only have one instance of femaleone. This might be ok if you had no covariates, but you can't estimate a covariate slope from one data point. Second, the scale of your covariates is much different, so you may need to normalize them. I would try removing the lone femaleone and rerun
On 08/11/2017 10:35 AM, Sullivan, Danielle Renee wrote:
Hi Doug,
Thanks for your response and help.
Attached is the matrix and the terminal output. Please let me know if you need anything else.
Danielle R. Sullivan (Miller), Ph.D.
National Center for PTSD
VA Boston Healthcare System Jamaica Plain
Boston University School of Medicine
OFFICE: (857) 364-4022
can you send the design matrix Xg.mtx and the full terminal output? On 08/08/2017 09:13 AM, Sullivan, Danielle Renee wrote:
Hi FreeSurfer Experts,
I am trying to run a dods cortical thickness analysis examining the interaction of group on a continuous variable while controlling for other variables.
Specifically, I have 6 classes (gender by three groups [zero, one, two]). I am interested in the group interaction with a continuous variable while controlling for gender (among 4 other continuous variables). I set up my fsgd file (see attached) and contrast file
(36
dimensions- F test):
*0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 [1 -1
0 1
-1 0]/2*
*0 0 0 0 0 0 **0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 [1 -0 -1 1 0 -1]/2*
and ran this command:
*mri_glmfit --y
**lh.age.PC1_forFSGD.PC2_forFSGD.PC3_forFSGD.GGT_forFSGD.rs8042149_A_bysex_forfsgd.thickness.10.mgh
--fsgd
fsgd.age.PC1_forFSGD.PC2_forFSGD.PC3_forFSGD.GGT_forFSGD.rs8042149_A_bysex_forFSGD.152.thickness.08082017.txt
dods --C
/GLM_CONTRAST.MAT/Contrast.Male.Female.Age.PC1.PC2.PC3.GGT.RORAbySex.mtx
--surf fsaverage lh --cortex --glmdir
lh.age.PC1_forFSGD.PC2_forFSGD.PC3_forFSGD.GGT_forFSGD.rs8042149_A_bysex_forfsgd.thickness.10.glmdir
But keep getting this error:
*ERROR: matrix is ill-conditioned or badly scaled, condno = 1e+08*
*--------------------------------*
*Possible problem with experimental design:*
*Check for duplicate entries and/or lack of range of*
*continuous variables within a class.*
*If you seek help with this problem, make sure to send:*
- Your command line:*
mri_glmfit --y
lh.age.PC1_forFSGD.PC2_forFSGD.PC3_forFSGD.GGT_forFSGD.rs8042149_A_bysex_forfsgd.thickness.10.mgh
--fsgd
fsgd.age.PC1_forFSGD.PC2_forFSGD.PC3_forFSGD.GGT_forFSGD.rs8042149_A_bysex_forFSGD.152.thickness.08082017.txt
dods --C
/Volumes/VA_Imaging/Projects/salat/2389_TRT/dmiller/GGT_CRP/GLM_CONTRAST.MAT/Contrast.Male.Female.Age.PC1.PC2.PC3.GGT.RORAbySex.mtx
--surf fsaverage lh --cortex --glmdir
lh.age.PC1_forFSGD.PC2_forFSGD.PC3_forFSGD.GGT_forFSGD.rs8042149_A_bysex_forfsgd.thickness.10.glmdir
- The FSGD file (if using one)*
- And the design matrix above*
Any idea on what I am doing wrong?
Danielle R. Miller, Ph.D.
National Center for PTSD
VA Boston Healthcare System Jamaica Plain
Boston University School of Medicine
OFFICE: (857) 364-4022
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center gr...@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop:https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html
http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing:ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to
whom it is
addressed. If you believe this e-mail was sent to you in error and
the e-mail
contains patient information, please contact the Partners Compliance
HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to
you in error
but does not contain patient information, please contact the sender
and properly
dispose of the e-mail.
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html
http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html
http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/