Dear Freesurfer Mailing List,

I wish to run a GLM in freesurfer with 3 categorical variables (haplotype-2levels, gender-2levels, group-2levels), and 1 continuous variables (age). Unfortunately, the MRI images I am using come from five different testing sites, so I also need to account for the potentially confounding effects of using different scanners. So my fsgd file like this:

GroupDescriptorFile 1
Title GLM_SiteGender
Class siteBrisbane-numRiskNoRisk-genderMale-groupCase
Class siteBrisbane-numRiskNoRisk-genderFemale-groupCase
Class siteBrisbane-numRiskNoRisk-genderMale-groupControl
Class siteBrisbane-numRiskNoRisk-genderFemale-groupControl
Class siteBrisbane-numRiskanyRisk-genderMale-groupCase
Class siteBrisbane-numRiskanyRisk-genderFemale-groupCase
Class siteBrisbane-numRiskanyRisk-genderMale-groupControl
Class siteBrisbane-numRiskanyRisk-genderFemale-groupControl
Class siteMelbourne-numRiskNoRisk-genderMale-groupCase
Class siteMelbourne-numRiskNoRisk-genderFemale-groupCase
Class siteMelbourne-numRiskNoRisk-genderMale-groupControl
Class siteMelbourne-numRiskNoRisk-genderFemale-groupControl
Class siteMelbourne-numRiskanyRisk-genderMale-groupCase
Class siteMelbourne-numRiskanyRisk-genderFemale-groupCase
Class siteMelbourne-numRiskanyRisk-genderMale-groupControl
Class siteMelbourne-numRiskanyRisk-genderFemale-groupControl
Class siteNewcastle-numRiskNoRisk-genderMale-groupCase
Class siteNewcastle-numRiskNoRisk-genderFemale-groupCase
Class siteNewcastle-numRiskNoRisk-genderMale-groupControl
Class siteNewcastle-numRiskNoRisk-genderFemale-groupControl
Class siteNewcastle-numRiskanyRisk-genderMale-groupCase
Class siteNewcastle-numRiskanyRisk-genderFemale-groupCase
Class siteNewcastle-numRiskanyRisk-genderMale-groupControl
Class siteNewcastle-numRiskanyRisk-genderFemale-groupControl
Class sitePerth-numRiskNoRisk-genderMale-groupCase
Class sitePerth-numRiskNoRisk-genderFemale-groupCase
Class sitePerth-numRiskNoRisk-genderMale-groupControl
Class sitePerth-numRiskNoRisk-genderFemale-groupControl
Class sitePerth-numRiskanyRisk-genderMale-groupCase
Class sitePerth-numRiskanyRisk-genderFemale-groupCase
Class sitePerth-numRiskanyRisk-genderMale-groupControl
Class sitePerth-numRiskanyRisk-genderFemale-groupControl
Class siteSydney-numRiskNoRisk-genderMale-groupCase
Class siteSydney-numRiskNoRisk-genderFemale-groupCase
Class siteSydney-numRiskNoRisk-genderMale-groupControl
Class siteSydney-numRiskNoRisk-genderFemale-groupControl
Class siteSydney-numRiskanyRisk-genderMale-groupCase
Class siteSydney-numRiskanyRisk-genderFemale-groupCase
Class siteSydney-numRiskanyRisk-genderMale-groupControl
Class siteSydney-numRiskanyRisk-genderFemale-groupControl
Variables age
Input 100105SA sitePerth-anyRiskAnyRisk-genderMale-groupCase 23
...

As the number of categorical variables really complicates the specification of contrast vectors, I was wondering if you could confirm whether these vectors are correct:

Main effect for haplotype (Risk vs No Risk):
0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Main effect for gender (Male vs Female):
0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0.5 -0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Main effect for group (Case vs Control):
0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Interaction between gender & group:
0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0.5 -0.5 -0.5 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Interaction between gender & haplotype:
0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Interaction between group & haplotype:
0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0.5 0.5 -0.5 -0.5 -0.5 -0.5 0.5 0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Interaction between gender, group, & haplotype:

0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0.5 -0.5 -0.5 0.5 -0.5 0.5 0.5 -0.5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0


Kind regards,

Bronwyn Overs
Research Assistant

Neuroscience Research Australia
Margarete Ainsworth Building
Barker Street Randwick Sydney NSW 2031 Australia
M 0411 308 769 T +61 2 9399 1883 F +61 2 9399 1265