Back to FSGD Examples

Two Groups (One Factor/Two Levels), One Covariate

This models the input as two separate planes (DODS), one for each group. The two groups can be thought of as two levels of a single discrete factor. The covariate can be thought of as continuous factor (eg, Age)

FSGD File (g2v1.fsgd)

GroupDescriptorFile 1
Title OSGM
Class Group1
Class Group2
Variables Age
Input subject1 Group1 30
Input subject2 Group2 40

Nclasses = 2
Nvariables = 1

Regressors (DODS)

Nregressors = Nclasses*(Nvariables+1) = 2*(1+1) = 4
Regressor1: ones for subjects in Group 1, 0 otherwise. Codes intercept/mean for Group 1
Regressor2: ones for subjects in Group 2, 0 otherwise. Codes intercept/mean for Group 2
Regressor3: age for subjects in Group 1, 0 otherwise. Codes age slope for Group 1
Regressor4: age for subjects in Group 2, 0 otherwise. Codes age slope for Group 2

Contrasts

The number of columns in each contrast matrix must be the same as the number of regressors (Nregressors). If there is only one row in the contrast matrix, then the result will be a t-test and will have a sign. Reversing the signs in the contrast matrix will only change the sign of the output, not its magnitude. If there is more than one row, the result will be an F-test and will be unsigned.

Contrast 1 group.diff.mtx

Null Hypothesis: is there a difference between the group intercepts? Is there a difference between groups regressing out the effect of age?

1 -1 0 0

This is a t-test with Group1>Group2 being positive (red/yellow).

Contrast 2 group-x-age.mtx

Null Hypothesis: is there a difference between the group age slopes? Note: this is an interaction between group and age. Note: not possible to test with DOSS.

0 0 1 -1

This is a t-test with Group1>Group2 being positive (red/yellow).

Contrast 3 g1g2.intercept.mtx

Null Hypothesis: does mean of group intercepts differ from 0? Is there an average main effect regressing out age?

0.5 0.5 0 0

This is a t-test with (Group1+Group2)/2 > 0 being positive (red/yellow). If the mean is < 0, then it will be displayed in blue/cyan.

Contrast 4 g1g2.age.mtx

Null Hypothesis: does mean of group age slope differ from 0? Is there an average affect of age regressing out the effect of group?

0 0 0.5 0.5

This is a t-test with (Group1+Group2)/2 > 0 being positive (red/yellow). If the mean is < 0, then it will be displayed in blue/cyan.

mri_glmfit command

This is an example invocation of mri_glmfit. Depending upon your application, you may have other options as well.

mri_glmfit \
  --glmdir g2v1 \
  --y y.mgh \
  --fsgd g2v1.fsgd \
  --C group.diff.mtx \
  --C group-x-age.mtx \
  --C g1g2.intercept.mtx \
  --C g1g2.age.mtx

Fsgdf2G1V (last edited 2010-09-20 15:10:52 by MartinReuter)