With 6 classes and 3 variables, you will have 24 regression coefficients and so 24 items in your contrast matrix (so good, so far). The order will be 1. MaleMaxT-Offset 2. MaleKahn-Offset 3. 4. 5. 6. 7. MaleMaxT-AgeSlope 8 9 10 11 12 13 MaleMaxT-HRSD_Reduction-Slope 14 MaleKahn-HRSD_Reduction-Slope 15 16 17 18 FemaleNone-HRSD_Reduction-Slope 19 MaleMaxT-SessionNumber-Slope 20 ...

So all the HRSD_Reduction-Slopevariables are in items 13-18, so you set them all to 1 and let the others be 0, which is what you have.

Bottom line: you have done it correctly.

On 10/13/2020 7:52 PM, Paul Dhami wrote:

External Email - Use Caution

Dear Freesurfer community,

I am having trouble with my contrast matrix, and would greatly appreciate any help with this.

I have 2 categorical variables, sex (2 levels) and training type (3 levels).

I also have three (continuous) variables: Age, percentage reduction in depression scores, and number of sessions.

I would like to assess the correlation between cortical thickness and % reduction in depression scores, while accounting for the effects of sex, training type, age, and number of sessions as nuisance variables.

Below is 1) my command line, 2) the FSGD fiile, and 3) the design matrix

mri_glmfit.bin --y MDD-lh-thickness.mgh --fsgd Reduction_Correlation.fsgd.txt --C ME_ReductionCorrelation_Contrast.txt --fwhm 20 --surf fsaverage lh --glmdir /Users/prabhjotdhami/Desktop/CorticalThickness/Data/myGLM

GroupDescriptorFile 1 Title MDD_correlation Class MaleMaxT Class MaleKahn Class MaleNone Class FemaleMaxT Class FemaleKahn Class FemaleNone Variables Age HRSD_Reduction SessionNumber Input yct-001 FemaleMaxT 24 62.5 20 Input yct-002 MaleKahn 18 70 20 Input yct-004 MaleMaxT 21 50 20 Input yct-005 MaleMaxT 24 25 20 Input yct-006 FemaleKahn 24 65.2173913 20 Input yct-012 FemaleKahn 17 71.42857143 20 Input yct-013 MaleMaxT 22 53.84615385 20 Input yct-014 FemaleMaxT 19 63.63636364 20 Input yct-015 FemaleMaxT 16 80.95238095 20 Input yct-016 FemaleKahn 24 65.2173913 20 Input yct-017 FemaleKahn 22 55 20 Input yct-018 FemaleMaxT 18 82.60869565 20 Input yct-019 MaleKahn 16 85 20 Input yct-020 FemaleMaxT 22 85 20 Input yct-021 MaleMaxT 24 81.81818182 20 Input yct-022 FemaleKahn 21 70.83333333 20 Input yct-023 FemaleMaxT 20 52 20 Input yct-024 MaleKahn 16 73.68421053 20 Input yct-025 MaleKahn 21 23.80952381 20 Input yct-026 FemaleMaxT 20 78.26086957 20 Input yct-027 FemaleMaxT 17 20 20 Input yct-030 FemaleMaxT 19 52 20 Input yd-p002 MaleNone 20 54.54545455 10 Input yd-p006 FemaleNone 23 30 10 Input yd-p008 MaleNone 22 44 10 Input yd-p010 FemaleNone 16 50 10 Input yd-p014 FemaleNone 22 3.846153846 10 Input yd-p015 MaleNone 19 10 10 Input yd-p022 FemaleNone 24 37.03703704 10 Input yd-p026 FemaleNone 21 47.61904762 10 Input yd-p027 MaleNone 23 31.57894737 10 Input yd-p032 FemaleNone 21 75 10 Input yd-p033 MaleNone 24 22.22222222 10 Input yd-p034 FemaleNone 22 25 10 Input yd-p035 MaleNone 24 30.43478261 10 Input yd-p047 MaleNone 16 78.26086957 10 Input yd-p048 MaleNone 17 20 10 Input yd-p051 FemaleNone 18 45 10 Input yd-p052 FemaleNone 20 47.82608696 10

0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 0

Any help would be greatly appreciated.

Thank you,

Paul

---

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