Hi, all,
I have some questions about performing the FDR correction in Qdec.
I ran recon-all on all my subjects, then ran the surface group analysis following the Qdec tutorial. There are 3 factors in the qdec.table.dat file: age, gender and diagnosis. In the Design tab of Qdec, I selected the discrete factor 'diagnosis' for analysis. When the analysis was done, there is a description "Does the average thickness differ between Diseased and Normal?" in the Display tab. I think this should be the uncorrected map. Then I tried to run the FDR correction for multiple comparisons. I set the rate to be 0.05 and 0.01 respectively. However, I found no difference between the results using different rates. In the Terminal, the information are also the same:
"MRISfdr2vwth(): np = 163842, nv = 163842 Found 107008 of 163842 vertices above FDR threshold (of 1.48619)"
I wonder whether I did the analysis correctly. Are such results possible or reasonable?
Thank you in advance.
Best,
Xiaojing
That is strange. Can you change the FDR to something ridiculous like .5 and see if that changes things?
Xiaojing Long wrote:
Hi, all,
I have some questions about performing the FDR correction in Qdec.
I ran recon-all on all my subjects, then ran the surface group analysis following the Qdec tutorial. There are 3 factors in the qdec.table.dat file: age, gender and diagnosis. In the Design tab of Qdec, I selected the discrete factor 'diagnosis' for analysis. When the analysis was done, there is a description "Does the average thickness differ between Diseased and Normal?" in the Display tab. I think this should be the uncorrected map. Then I tried to run the FDR correction for multiple comparisons. I set the rate to be 0.05 and 0.01 respectively. However, I found no difference between the results using different rates. In the Terminal, the information are also the same:
"MRISfdr2vwth(): np = 163842, nv = 163842 Found 107008 of 163842 vertices above FDR threshold (of 1.48619)"
I wonder whether I did the analysis correctly. Are such results possible or reasonable?
Thank you in advance.
Best,
Xiaojing
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
I tried that. Still no change.
And I tried running Qdec on all of my datasets. The problem occurs on all of them.
Xiaojing
2010/7/9 Douglas N Greve greve@nmr.mgh.harvard.edu
That is strange. Can you change the FDR to something ridiculous like .5 and see if that changes things?
Xiaojing Long wrote:
Hi, all, I have some questions about performing the FDR correction in Qdec. I ran recon-all on all my subjects, then ran the surface group analysis following the Qdec tutorial. There are 3 factors in the qdec.table.dat file: age, gender and diagnosis. In the Design tab of Qdec, I selected the discrete factor 'diagnosis' for analysis. When the analysis was done, there is a description "Does the average thickness differ between Diseased and Normal?" in the Display tab. I think this should be the uncorrected map. Then I tried to run the FDR correction for multiple comparisons. I set the rate to be 0.05 and 0.01 respectively. However, I found no difference between the results using different rates. In the Terminal, the information are also the same: "MRISfdr2vwth(): np = 163842, nv = 163842 Found 107008 of 163842 vertices above FDR threshold (of 1.48619)" I wonder whether I did the analysis correctly. Are such results possible or reasonable? Thank you in advance. Best, Xiaojing
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 greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Now it works if I enter the FDR in the text box -> press Enter -> click on Set Using FDR
Thank you so much.
Best,
Xiaojing
2010/7/9 Douglas N Greve greve@nmr.mgh.harvard.edu
That is strange. Can you change the FDR to something ridiculous like .5 and see if that changes things?
Xiaojing Long wrote:
Hi, all, I have some questions about performing the FDR correction in Qdec. I ran recon-all on all my subjects, then ran the surface group analysis following the Qdec tutorial. There are 3 factors in the qdec.table.dat file: age, gender and diagnosis. In the Design tab of Qdec, I selected the discrete factor 'diagnosis' for analysis. When the analysis was done, there is a description "Does the average thickness differ between Diseased and Normal?" in the Display tab. I think this should be the uncorrected map. Then I tried to run the FDR correction for multiple comparisons. I set the rate to be 0.05 and 0.01 respectively. However, I found no difference between the results using different rates. In the Terminal, the information are also the same: "MRISfdr2vwth(): np = 163842, nv = 163842 Found 107008 of 163842 vertices above FDR threshold (of 1.48619)" I wonder whether I did the analysis correctly. Are such results possible or reasonable? Thank you in advance. Best, Xiaojing
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 greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
freesurfer@nmr.mgh.harvard.edu
-
Douglas N Greve -
Xiaojing Long