Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

...

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

...

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

...

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

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

I tried this command:

mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

and got this error: ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mri_glmfit.log

then I copied the .log file into the contrast directory C-pre-vs-post and tried the same command again and got this error:

ERROR: could not determine file for /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mask

Ususally I would run the MC on the lh-Diff-1-3-Intercept-long.area-spc folder but with the rmanova I get these two folders: 40kids-lh.thickness10-rmanova and C-pre-vs-post

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 18:57:36 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

what do you mean that it did not seem to work?

On 03/30/2017 11:25 AM, Clara Kühn wrote:

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

...

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

...

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

...

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

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

ah, ok. I used the proper glmdir now with this command: mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

And this is the terminal output:

cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

WARNING: unrecognized mri_glmfit cmd option doss

log file is /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//cache.mri_glmfit-sim.log

cd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2 /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64/bin/mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

$Id: mri_glmfit-sim,v 1.36.2.5 2012/10/01 22:31:37 greve Exp $ Fri Mar 31 09:17:53 CEST 2017 Linux etsch 4.4.0-66-generic #87-Ubuntu SMP Fri Mar 3 15:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux ckuehn setenv SUBJECTS_DIR /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/ FREESURFER_HOME /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64

Original mri_glmfit command line: cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

DoSim = 0 UseCache = 1 DoPoll = 0 DoPBSubmit = 0 DoBackground = 0 DiagCluster = 0 gd2mtx = doss fwhm = -nan ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor////cortex/fwhmNaN/abs/th30/mc-z.csd

For some reason it doesn't recognize that I smoothed the data with --fwhm 10 during mris_surf2surf. And the other thing is, that it doesn't find the correct path for the mc-z.csd. It should be in /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/81kids_template/lh/cortex/fwhm10/abs/mc-z.csd

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 19:13:43 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

you are not giving it the proper glmdir dir. the glmdir is

/data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova

On 03/30/2017 01:12 PM, Clara Kühn wrote:

I tried this command:

mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

and got this error: ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mri_glmfit.log

then I copied the .log file into the contrast directory C-pre-vs-post and tried the same command again and got this error:

ERROR: could not determine file for /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mask

Ususally I would run the MC on the lh-Diff-1-3-Intercept-long.area-spc folder but with the rmanova I get these two folders: 40kids-lh.thickness10-rmanova and C-pre-vs-post

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 18:57:36 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

what do you mean that it did not seem to work?

On 03/30/2017 11:25 AM, Clara Kühn wrote:

...

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

...

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

...

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

...

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

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

Hi Doug I am not conciously using a mask but I attached the glmfit.log At the end it actually states "ResidualFWHM -nan"

How can I fix that?

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 10. April 2017 21:41:40 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

It estimates the FWHM from the residuals of the analysis (it does not use whatever value you smoothed it with). For some reason this estimate is not-a-number (NaN). Are you using a mask? Can you send the terminal output from running mri_glmfit?

On 03/31/2017 03:29 AM, Clara Kühn wrote:

ah, ok. I used the proper glmdir now with this command: mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

And this is the terminal output:

cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

WARNING: unrecognized mri_glmfit cmd option doss

log file is /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//cache.mri_glmfit-sim.log

cd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2 /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64/bin/mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

$Id: mri_glmfit-sim,v 1.36.2.5 2012/10/01 22:31:37 greve Exp $ Fri Mar 31 09:17:53 CEST 2017 Linux etsch 4.4.0-66-generic #87-Ubuntu SMP Fri Mar 3 15:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux ckuehn setenv SUBJECTS_DIR /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/ FREESURFER_HOME /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64

Original mri_glmfit command line: cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

DoSim = 0 UseCache = 1 DoPoll = 0 DoPBSubmit = 0 DoBackground = 0 DiagCluster = 0 gd2mtx = doss fwhm = -nan ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor////cortex/fwhmNaN/abs/th30/mc-z.csd

For some reason it doesn't recognize that I smoothed the data with --fwhm 10 during mris_surf2surf. And the other thing is, that it doesn't find the correct path for the mc-z.csd. It should be in /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/81kids_template/lh/cortex/fwhm10/abs/mc-z.csd

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 19:13:43 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

you are not giving it the proper glmdir dir. the glmdir is

/data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova

On 03/30/2017 01:12 PM, Clara Kühn wrote:

...

I tried this command:

mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

and got this error: ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mri_glmfit.log

then I copied the .log file into the contrast directory C-pre-vs-post and tried the same command again and got this error:

ERROR: could not determine file for /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mask

Ususally I would run the MC on the lh-Diff-1-3-Intercept-long.area-spc folder but with the rmanova I get these two folders: 40kids-lh.thickness10-rmanova and C-pre-vs-post

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 18:57:36 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

what do you mean that it did not seem to work?

On 03/30/2017 11:25 AM, Clara Kühn wrote:

...

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

...

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

...

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

...

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

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

Oh, yes, sorry: command line: mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

output: gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /data/p_npsy001_neurotrain/new_syntax_training/Pilot2-NachtestungPilot0/logfiles cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname etsch machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//mask.mgh search space = 97361.000000 DOF = 39 Starting fit and test Fit completed in 0.551233 minutes Computing spatial AR1 in volume. fMRIspatialAR1(): hit 0 voxels WARNING: no voxels in AR1 computation Residual: ar1mn = (-nan,-nan,-nan) fwhm = (-nan,-nan,-nan) -nan Writing results C-pre-vs-post maxvox sig=3.9739 F=18.6032 at index 4623 0 0 seed=1492398301 mri_glmfit done

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 11. April 2017 15:58:27 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

sorry, that is the log file. What I want is the terminal output, ie, when you run it, all the stuff that is printed to the screen.

On 4/11/17 5:07 AM, Clara Kühn wrote:

Hi Doug I am not conciously using a mask but I attached the glmfit.log At the end it actually states "ResidualFWHM -nan"

How can I fix that?

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 10. April 2017 21:41:40 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

It estimates the FWHM from the residuals of the analysis (it does not use whatever value you smoothed it with). For some reason this estimate is not-a-number (NaN). Are you using a mask? Can you send the terminal output from running mri_glmfit?

On 03/31/2017 03:29 AM, Clara Kühn wrote:

ah, ok. I used the proper glmdir now with this command: mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

And this is the terminal output:

cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

WARNING: unrecognized mri_glmfit cmd option doss

log file is /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//cache.mri_glmfit-sim.log

cd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2 /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64/bin/mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

$Id: mri_glmfit-sim,v 1.36.2.5 2012/10/01 22:31:37 greve Exp $ Fri Mar 31 09:17:53 CEST 2017 Linux etsch 4.4.0-66-generic #87-Ubuntu SMP Fri Mar 3 15:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux ckuehn setenv SUBJECTS_DIR /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/ FREESURFER_HOME /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64

Original mri_glmfit command line: cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

DoSim = 0 UseCache = 1 DoPoll = 0 DoPBSubmit = 0 DoBackground = 0 DiagCluster = 0 gd2mtx = doss fwhm = -nan ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor////cortex/fwhmNaN/abs/th30/mc-z.csd

For some reason it doesn't recognize that I smoothed the data with --fwhm 10 during mris_surf2surf. And the other thing is, that it doesn't find the correct path for the mc-z.csd. It should be in /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/81kids_template/lh/cortex/fwhm10/abs/mc-z.csd

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 19:13:43 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

you are not giving it the proper glmdir dir. the glmdir is

/data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova

On 03/30/2017 01:12 PM, Clara Kühn wrote:

I tried this command:

mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

and got this error: ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mri_glmfit.log

then I copied the .log file into the contrast directory C-pre-vs-post and tried the same command again and got this error:

ERROR: could not determine file for /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mask

Ususally I would run the MC on the lh-Diff-1-3-Intercept-long.area-spc folder but with the rmanova I get these two folders: 40kids-lh.thickness10-rmanova and C-pre-vs-post

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 18:57:36 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

what do you mean that it did not seem to work?

On 03/30/2017 11:25 AM, Clara Kühn wrote:

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

_______________________________________________ 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

_______________________________________________ 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.

Hi Doug,

attached you find the rmanova directory after I ran the mri_glmfit command. I hope this can shed some light on the problem...

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 11. April 2017 18:29:52 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Still a bit of a mystery. Can you tar up and send me the rmanova folder?

On 04/11/2017 11:18 AM, Clara Kühn wrote:

Oh, yes, sorry: command line: mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

output: gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /data/p_npsy001_neurotrain/new_syntax_training/Pilot2-NachtestungPilot0/logfiles cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname etsch machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//mask.mgh search space = 97361.000000 DOF = 39 Starting fit and test Fit completed in 0.551233 minutes Computing spatial AR1 in volume. fMRIspatialAR1(): hit 0 voxels WARNING: no voxels in AR1 computation Residual: ar1mn = (-nan,-nan,-nan) fwhm = (-nan,-nan,-nan) -nan Writing results C-pre-vs-post maxvox sig=3.9739 F=18.6032 at index 4623 0 0 seed=1492398301 mri_glmfit done

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 11. April 2017 15:58:27 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

sorry, that is the log file. What I want is the terminal output, ie, when you run it, all the stuff that is printed to the screen.

On 4/11/17 5:07 AM, Clara Kühn wrote:

Hi Doug I am not conciously using a mask but I attached the glmfit.log At the end it actually states "ResidualFWHM -nan"

How can I fix that?

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 10. April 2017 21:41:40 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

It estimates the FWHM from the residuals of the analysis (it does not use whatever value you smoothed it with). For some reason this estimate is not-a-number (NaN). Are you using a mask? Can you send the terminal output from running mri_glmfit?

On 03/31/2017 03:29 AM, Clara Kühn wrote:

ah, ok. I used the proper glmdir now with this command: mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

And this is the terminal output:

cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

WARNING: unrecognized mri_glmfit cmd option doss

log file is /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//cache.mri_glmfit-sim.log

cd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2 /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64/bin/mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

$Id: mri_glmfit-sim,v 1.36.2.5 2012/10/01 22:31:37 greve Exp $ Fri Mar 31 09:17:53 CEST 2017 Linux etsch 4.4.0-66-generic #87-Ubuntu SMP Fri Mar 3 15:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux ckuehn setenv SUBJECTS_DIR /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/ FREESURFER_HOME /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64

Original mri_glmfit command line: cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

DoSim = 0 UseCache = 1 DoPoll = 0 DoPBSubmit = 0 DoBackground = 0 DiagCluster = 0 gd2mtx = doss fwhm = -nan ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor////cortex/fwhmNaN/abs/th30/mc-z.csd

For some reason it doesn't recognize that I smoothed the data with --fwhm 10 during mris_surf2surf. And the other thing is, that it doesn't find the correct path for the mc-z.csd. It should be in /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/81kids_template/lh/cortex/fwhm10/abs/mc-z.csd

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 19:13:43 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

you are not giving it the proper glmdir dir. the glmdir is

/data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova

On 03/30/2017 01:12 PM, Clara Kühn wrote:

I tried this command:

mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

and got this error: ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mri_glmfit.log

then I copied the .log file into the contrast directory C-pre-vs-post and tried the same command again and got this error:

ERROR: could not determine file for /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mask

Ususally I would run the MC on the lh-Diff-1-3-Intercept-long.area-spc folder but with the rmanova I get these two folders: 40kids-lh.thickness10-rmanova and C-pre-vs-post

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 18:57:36 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

what do you mean that it did not seem to work?

On 03/30/2017 11:25 AM, Clara Kühn wrote:

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

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

---

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

I added --surf mytemplate lh to the command and now it works.

Thank you so much!!

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 11. April 2017 20:13:21 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

oh, I think I see the problem now. The input is surface-based data, but you did not include a --surf subject hemi option, so it thinks that it is a volume.

On 04/11/2017 12:58 PM, Clara Kühn wrote:

Hi Doug,

attached you find the rmanova directory after I ran the mri_glmfit command. I hope this can shed some light on the problem...

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 11. April 2017 18:29:52 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Still a bit of a mystery. Can you tar up and send me the rmanova folder?

On 04/11/2017 11:18 AM, Clara Kühn wrote:

...

Oh, yes, sorry: command line: mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

output: gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /data/p_npsy001_neurotrain/new_syntax_training/Pilot2-NachtestungPilot0/logfiles cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname etsch machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//mask.mgh search space = 97361.000000 DOF = 39 Starting fit and test Fit completed in 0.551233 minutes Computing spatial AR1 in volume. fMRIspatialAR1(): hit 0 voxels WARNING: no voxels in AR1 computation Residual: ar1mn = (-nan,-nan,-nan) fwhm = (-nan,-nan,-nan) -nan Writing results C-pre-vs-post maxvox sig=3.9739 F=18.6032 at index 4623 0 0 seed=1492398301 mri_glmfit done

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 11. April 2017 15:58:27 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

sorry, that is the log file. What I want is the terminal output, ie, when you run it, all the stuff that is printed to the screen.

On 4/11/17 5:07 AM, Clara Kühn wrote:

Hi Doug I am not conciously using a mask but I attached the glmfit.log At the end it actually states "ResidualFWHM -nan"

How can I fix that?

Cheers, Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 10. April 2017 21:41:40 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

It estimates the FWHM from the residuals of the analysis (it does not use whatever value you smoothed it with). For some reason this estimate is not-a-number (NaN). Are you using a mask? Can you send the terminal output from running mri_glmfit?

On 03/31/2017 03:29 AM, Clara Kühn wrote:

ah, ok. I used the proper glmdir now with this command: mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

And this is the terminal output:

cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

WARNING: unrecognized mri_glmfit cmd option doss

log file is /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova//cache.mri_glmfit-sim.log

cd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2 /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64/bin/mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/qdec/rmanova/ --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

$Id: mri_glmfit-sim,v 1.36.2.5 2012/10/01 22:31:37 greve Exp $ Fri Mar 31 09:17:53 CEST 2017 Linux etsch 4.4.0-66-generic #87-Ubuntu SMP Fri Mar 3 15:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux ckuehn setenv SUBJECTS_DIR /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/ FREESURFER_HOME /afs/cbs.mpg.de/software/freesurfer/5.3.0/ubuntu-xenial-amd64

Original mri_glmfit command line: cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2017.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx

DoSim = 0 UseCache = 1 DoPoll = 0 DoPBSubmit = 0 DoBackground = 0 DiagCluster = 0 gd2mtx = doss fwhm = -nan ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor////cortex/fwhmNaN/abs/th30/mc-z.csd

For some reason it doesn't recognize that I smoothed the data with --fwhm 10 during mris_surf2surf. And the other thing is, that it doesn't find the correct path for the mc-z.csd. It should be in /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/81kids_template/lh/cortex/fwhm10/abs/mc-z.csd

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 19:13:43 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

you are not giving it the proper glmdir dir. the glmdir is

/data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova

On 03/30/2017 01:12 PM, Clara Kühn wrote:

I tried this command:

mri_glmfit-sim --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post --cache-dir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//average/mult-comp-cor/ --cache 3.0 abs --cwp 0.05 --2spaces

and got this error: ERROR: cannot find /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mri_glmfit.log

then I copied the .log file into the contrast directory C-pre-vs-post and tried the same command again and got this error:

ERROR: could not determine file for /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post/mask

Ususally I would run the MC on the lh-Diff-1-3-Intercept-long.area-spc folder but with the rmanova I get these two folders: 40kids-lh.thickness10-rmanova and C-pre-vs-post

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Donnerstag, 30. März 2017 18:57:36 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

what do you mean that it did not seem to work?

On 03/30/2017 11:25 AM, Clara Kühn wrote:

Thank you! I completely didn't see that. It worked.

I was wondering how to proceed next. The tutorial ends with running the glmfit command. I've tried correcting for multiple comparisons with Monte Carlo but that doesn't seem to work. How can I interpret the files that are the output from the repeated measures ANOVA?

Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Mittwoch, 29. März 2017 18:15:57 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Since there are 40 subjects, you would only have 40+1 columns, not 80+1

On 03/29/2017 03:42 AM, Clara Kühn wrote:

Hi Doug,

I've also attached the fsgd file for you. Yes, I have 40 subjects and a contrast column. I adapted this from this tutorial: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova where the example has 3 time points and therefore 2 contrast columns. Since I want to look at only 2 I only have one contrast column which is also my 81st column in the contrast matrix.

Did I adapt it in a wrong way for my 2 timepoints? Cheers Clara

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Dienstag, 28. März 2017 18:37:58 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

I'm not sure what's in your FSGD file, but you design matrix has 41 columns and your contrast matrix has 80. I'm guessing that you have 40 subjects and that you created the contrast matrix based on DODS and not DOSS (see the web page on the rmanova about creating contrasts)

On 03/28/2017 03:50 AM, Clara Kühn wrote:

Sorry, I had a typo in the fsgd file. This is the correct command line and terminal output:

mri_glmfit --glmdir $SUBJECTS_DIR/qdec/rmanova/ --y $SUBJECTS_DIR/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd $SUBJECTS_DIR/qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C $SUBJECTS_DIR/qdec/rmanova/C-pre-vs-post.mtx

gdfReadHeader: reading /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd INFO: DeMeanFlag keyword not found, DeMeaning will NOT be done. Continuous Variable Means (all subjects) 0 pre-vs-post 0 1 Class Means of each Continuous Variable 1 Subject1 0.0000 2 Subject2 0.0000 3 Subject3 0.0000 4 Subject4 0.0000 5 Subject5 0.0000 6 Subject6 0.0000 7 Subject7 0.0000 8 Subject8 0.0000 9 Subject9 0.0000 10 Subject10 0.0000 11 Subject11 0.0000 12 Subject12 0.0000 13 Subject13 0.0000 14 Subject14 0.0000 15 Subject15 0.0000 16 Subject16 0.0000 17 Subject17 0.0000 18 Subject18 0.0000 19 Subject19 0.0000 20 Subject20 0.0000 21 Subject21 0.0000 22 Subject22 0.0000 23 Subject23 0.0000 24 Subject24 0.0000 25 Subject25 0.0000 26 Subject26 0.0000 27 Subject27 0.0000 28 Subject28 0.0000 29 Subject29 0.0000 30 Subject30 0.0000 31 Subject31 0.0000 32 Subject32 0.0000 33 Subject33 0.0000 34 Subject34 0.0000 35 Subject35 0.0000 36 Subject36 0.0000 37 Subject37 0.0000 38 Subject38 0.0000 39 Subject39 0.0000 40 Subject40 0.0000 INFO: gd2mtx_method is doss

$Id: mri_glmfit.c,v 1.196.2.8 2012/11/01 18:51:41 greve Exp $ cwd /home/raid2/ckuehn cmdline mri_glmfit --glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ --y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh --fsgd /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd doss --C /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx sysname Linux hostname parana machine x86_64 user ckuehn FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 0 y /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh logyflag 0 usedti 0 FSGD /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/2014.03.27-40kids-rmanova.fsgd glmdir /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/ Loading y from /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2/81kids_template/GLMprep/40kids-lh.thickness-rmanova.sm10.mgh INFO: gd2mtx_method is doss Saving design matrix to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//Xg.dat Normalized matrix condition is 1 Matrix condition is 40 Pruning voxels by thr: 0.000000 Found 97361 voxels in mask Saving mask to /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova//mask.mgh search space = 97361.000000 ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

----- Ursprüngliche Mail ----- Von: "Douglas N Greve" greve@nmr.mgh.harvard.edu An: freesurfer@nmr.mgh.harvard.edu Gesendet: Montag, 27. März 2017 19:49:04 Betreff: Re: [Freesurfer] glmfit mismatch between X and C with same no of columns

Need command line and terminal output

On 03/27/2017 09:08 AM, Clara Kühn wrote:

Dear Freesurfer experts,

I'm trying to run a repeated measures anova as described here: https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova I have 40 subjects with 2 time points each, so it's quite a large matrix. When running the mri_glmfit command I get the following error:

ERROR: dimension mismatch between X and contrast /data/p_npsy001_neurotrain/kids_data/sem_con-tp1and2//qdec/rmanova/C-pre-vs-post.mtx X has 41 cols, C has 81 cols

I'm not sure what to do as the matrix and the contrast (see attachments) actually have the same amount of columns...

Any help would be much appreciated, thank you! Clara

---

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

---

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

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer