Hello,
I am using randomise with TFCE to examine the effects of diagnosis (HC, MCI, AD), sex (F, M), and APOE-genotype (carrier, non) on the FA skeleton in 450 subjects. As suggested by the GLM Wiki page, I set up my design and contrast matrices using a factor effects approach. I also included age as a covariate, which was centered across the mean of all subjects. I've included my matrices below. Do they look correct?
All main effects and interactions in the model were significant. To run post hoc comparisons for the 2- and 3-way interactions which approach would you suggest? For example, if I was running posthoc comparisons for the sex*APOE interaction, would you suggest:
1. Extract the average FA for all subjects from the all_FA_skeleton based on significant clusters from the sex*APOE4 fmap and run follow-up statistics in R, or
2. Run voxelwise t-tests within each level of the factors (e.g., (1) within females, examine voxelwise differences between e4 and none4 carriers; (2) within males, examine voxelwise differences between e4 and none4 carriers; etc)?
Thanks,
Jenna
------ DESIGN MATRICES ------
#### Description
column 1 = mean
column 2 = sex (1 M; -1 F)
column 3 = APOE (1 non-carrier; -1 carrier)
column 4 = DX1 (1 HC, -1 AD)
column 5 = DX2 (1 MCI, -1 AD)
column 6 = age
column 7= sex*e4 (col2*col3)
column 8 = sex*dx1
column 9 = sex*dx2
column 10 = apoe*dx1
column 11 = apoe*dx2
column 12 = sex*apoe*dx1
column 13 = sex*apoe*dx2
### File contents
/NumWaves 13
/NumPoints 450
/Matrix
1 -1 1 0 1 9.414004376 -1 0 -1 0 1 0 -1
1 1 1 0 1 11.41400438 1 0 1 0 1 0 1
1 1 1 0 1 6.014004376 1 0 1 0 1 0 1
1 1 1 0 1 8.714004376 1 0 1 0 1 0 1
1 1 1 -1 -1 8.514004376 1 -1 -1 -1 -1 -1 -1
1 -1 1 -1 -1 13.61400438 -1 1 1 -1 -1 1 1
1 1 1 1 0 8.314004376 1 1 0 1 0 1 0
------ CONTRAST MATRICES ------
#### Description
contrast 1 = sex
contrast 2 = APOE
contrast 3 = DX1
contrast 4 = DX2
contrast 5 = sex*apoe
contrast 6 = sex*dx1
contrast 7= sex*dx2
contrast 8 = apoe*dx1
contrast 9 = apoe*dx2
contrast 10 = sex*apoe*dx1
contrast 11 = sex*apoe*dx2
### File contents
/NumWaves 13
/NumPoints 11
/Matrix
0 1 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 0 0 0
0 0 0 0 0 0 1 0 0 0 0 0 0
0 0 0 0 0 0 0 1 0 0 0 0 0
0 0 0 0 0 0 0 0 1 0 0 0 0
0 0 0 0 0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 0 0 0 0 1
------ FTEST MATRICES ------
#### Description
contrast 1 = sex
contrast 2 = APOE
contrast 3 = DX
contrast 4 = sex*apoe
contrast 5 = sex*dx
contrast 6= apoe*dx
contrast 7 = 3-way interaction
### File contents
/NumWaves 11
/NumPoints 7
/Matrix
1 0 0 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0 0 0
0 0 1 1 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 0
0 0 0 0 0 1 1 0 0 0 0
0 0 0 0 0 0 0 1 1 0 0
0 0 0 0 0 0 0 0 0 1 1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|