Hello again,
I also tried the command line options. I get an error for ill-conditioned
matrix. Could it be because of possible correlations between covariates? For
example I am using Age and mean thickness, should I just use one of them?
Here is the error:
wrls-guest-166-198:subjects manishdalwani$ mri_glmfit --y
lh.thickness.CD.10.mlh.thickness.CD.10.mgh
lh.thickness.CD.10.mris_preproc.log
wrls-guest-166-198:subjects manishdalwani$ mri_glmfit --y
lh.thickness.CD.10.mgh --fsgd FSGD_CD.fsgd dods --C thickness_CD.mtx --surf
fsaverage lh --cortex --glmdir lh.thickness_CD.glmdir
gdfReadHeader: reading FSGD_CD.fsgd
INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done.
Continuous Variable Means (all subjects)
0 Age 16.6492 1.13124
1 Meanthickness 2.4772 0.0616129
2 IQ 98.08 8.24097
3 CD 6.44 2.77244
Class Means of each Continuous Variable
1 PATIENTS 16.6492 2.4772 98.0800 6.4400
INFO: gd2mtx_method is dods
Reading source surface /Applications/freesurfer/subjects/fsaverage/surf/lh.white
Number of vertices 163842
Number of faces 327680
Total area 65416.648438
AvgVtxArea 0.399267
AvgVtxDist 0.721953
StdVtxDist 0.195470
$Id: mri_glmfit.c,v 1.196.2.6 2011/05/05 20:54:25 greve Exp $
cwd /Applications/freesurfer/subjects
cmdline mri_glmfit --y lh.thickness.CD.10.mgh --fsgd FSGD_CD.fsgd dods --C
thickness_CD.mtx --surf fsaverage lh --cortex --glmdir lh.thickness_CD.glmdir
sysname Darwin
hostname wrls-guest-166-198.ucdenver.edu
machine i386
user manishdalwani
FixVertexAreaFlag = 1
UseMaskWithSmoothing 1
OneSampleGroupMean 0
y /Applications/freesurfer/subjects/lh.thickness.CD.10.mgh
logyflag 0
usedti 0
FSGD FSGD_CD.fsgd
labelmask /Applications/freesurfer/subjects/fsaverage/label/lh.cortex.label
maskinv 0
glmdir lh.thickness_CD.glmdir
IllCondOK 0
ReScaleX 1
DoFFx 0
Creating output directory lh.thickness_CD.glmdir
Loading y from /Applications/freesurfer/subjects/lh.thickness.CD.10.mgh
INFO: gd2mtx_method is dods
Saving design matrix to lh.thickness_CD.glmdir/Xg.dat
Normalized matrix condition is 22815.1
Design matrix ------------------
1.000 15.320 2.500 104.000 2.000;
1.000 18.250 2.440 113.000 9.000;
1.000 18.080 2.460 108.000 3.000;
1.000 14.600 2.560 96.000 11.000;
1.000 14.350 2.580 109.000 6.000;
1.000 17.620 2.530 90.000 6.000;
1.000 16.990 2.500 102.000 7.000;
1.000 17.720 2.450 96.000 6.000;
1.000 17.170 2.450 99.000 7.000;
1.000 17.100 2.360 92.000 1.000;
1.000 16.010 2.460 83.000 4.000;
1.000 17.040 2.430 95.000 4.000;
1.000 17.730 2.490 104.000 2.000;
1.000 15.280 2.480 100.000 8.000;
1.000 16.440 2.390 105.000 10.000;
1.000 16.700 2.580 95.000 8.000;
1.000 16.840 2.460 93.000 9.000;
1.000 17.710 2.380 115.000 5.000;
1.000 15.120 2.470 86.000 11.000;
1.000 16.780 2.520 99.000 7.000;
1.000 15.100 2.420 86.000 5.000;
1.000 16.060 2.510 88.000 6.000;
1.000 18.490 2.430 92.000 7.000;
1.000 16.700 2.610 102.000 11.000;
1.000 17.030 2.470 100.000 6.000;
--------------------------------
ERROR: matrix is ill-conditioned or badly scaled, condno = 22815.1
--------------------------------
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:
1. Your command line:
mri_glmfit --y lh.thickness.CD.10.mgh --fsgd FSGD_CD.fsgd dods --C
thickness_CD.mtx --surf fsaverage lh --cortex --glmdir lh.thickness_CD.glmdir
2. The FSGD file (if using one)
3. And the design matrix above
________________________________
From: Manish Dalwani <manishdalw...@yahoo.com>
To: Douglas Greve <gr...@nmr.mgh.harvard.edu>
Cc: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
Sent: Monday, June 24, 2013 1:01 PM
Subject: Re: [Freesurfer] Help with QDEC error
Hi Nick and Doug,
Can you suggest an alternative way if QDEC is unable to handle a single group?
I would like to run these analyses to address reviewers comments.
Best,
Manish
________________________________
From: Douglas Greve <gr...@nmr.mgh.harvard.edu>
To: Manish Dalwani <manishdalw...@yahoo.com>
Cc: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
Sent: Thursday, June 20, 2013 11:16 PM
Subject: Re: [Freesurfer] Help with QDEC error
I thought it would have worked with a single group. Maybe Nick knows
of the top of his head.
doug
On 6/20/13 12:36 PM, Manish Dalwani wrote:
Hi Doug and Freesurfers,
>
>
>I realized that I hadn't updated my group levels to one group. However, when I
>do that..the qdec complains about group.level having at minimum two levels.
>How can I run (using QDEC) single group regression?
>
>
>Thanks,
>Manish
>
>
>
>________________________________
> From: Manish Dalwani <manishdalw...@yahoo.com>
>To: Douglas Greve <gr...@nmr.mgh.harvard.edu>;
>"freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
>Sent: Thursday, June 20, 2013 1:26 PM
>Subject: Re: [Freesurfer] Help with QDEC error
>
>
>
>Hi Doug,
>
>
>I was out of town for a conference. Please see my attachments. I have the log
>file and screenshots of ill-matrix conditioning.
>Thanks for your help!
>Manish
>
>
>
>________________________________
> From: Douglas Greve <gr...@nmr.mgh.harvard.edu>
>To: freesurfer@nmr.mgh.harvard.edu
>Sent: Saturday, June 15, 2013 11:09 AM
>Subject: Re: [Freesurfer] Help with QDEC error
>
>
>
>
>Can you send the mri_glmfit.log file from
>/Applications/freesurfer/subjects/qdec/Untitled ?
>doug
>
>
>
>On 6/14/13 4:57 PM, Manish Dalwani wrote:
>
>Hello Freesurfers,
>>
>>
>>I am trying to run regression analyses within patients and one variable of
>>interest which adjusting for the nuisance variables age, IQ, thickness. When
>>i select the group and the CD (variable of interest) and run analyses (DODS),
>>I get the following error:
>>
>>
>>Qdec 1.4 (Qdec1.4)
>>
>>
>>Type: Error
>>Time: Fri Jun 14 14:49:37 2013
>>Description: Error in Analyze: command failed: mri_glmfit --y
>>/Applications/freesurfer/subjects/qdec/Untitled/y.mgh --fsgd
>>/Applications/freesurfer/subjects/qdec/Untitled/qdec.fsgd dods --glmdir
>>/Applications/freesurfer/subjects/qdec/Untitled --surf fsaverage lh --C
>>/Applications/freesurfer/subjects/qdec/Untitled/contrasts/lh-Avg-Intercept-thickness.mat
>> --C
>>/Applications/freesurfer/subjects/qdec/Untitled/contrasts/lh-Avg-thickness-CD-Cor.mat
>> --C
>>/Applications/freesurfer/subjects/qdec/Untitled/contrasts/lh-Diff-CONTROLS-PATIENTS-Intercept-thickness.mat
>> --C
/Applications/freesurfer/subjects/qdec/Untitled/contrasts/lh-Diff-CONTROLS-PATIENTS-Cor-thickness-CD.mat
>>
>>
>>Please advice!
>>
>>
>>Regards,
>>Manish Dalwani
>>Instructor
>>Dept. of Psychiatry
>>University of Colorado
>>
>>
>>_______________________________________________
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>
>_______________________________________________
>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.
>
>
>
>_______________________________________________
>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.
>
>
>_______________________________________________
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.
