I am resending this message with a different subject title as my previous message had a blank subject line.
Hi Gwenaelle,
The subject ordering is alphabetical, and this same ordering is used in the design.mat file.
Here is the output with --debug. It fails to progress, although the processor is humming along at 100%.
Jay
[joel@beast stats]$ randomise -i GM_mod_merg_s3 -o test -m GM_mask -d design.mat -t design.con -n 5000 -T -V --debug
randomise options: -i GM_mod_merg_s3 -o test -m GM_mask -d design.mat -t design.con -n 5000 -T -V --debug
Loading Data: **********************************************************************
Data loaded
Confounds detected.
Subject | Design | group | label
1.000000 1.142857 1.000000 1.000000
2.000000 1.142857 1.000000 1.000000
3.000000 1.142857 1.000000 1.000000
4.000000 1.142857 1.000000 1.000000
5.000000 1.142857 1.000000 1.000000
6.000000 1.142857 1.000000 1.000000
7.000000 1.142857 1.000000 1.000000
8.000000 1.142857 1.000000 1.000000
9.000000 1.142857 1.000000 1.000000
10.000000 1.142857 1.000000 1.000000
11.000000 1.142857 1.000000 1.000000
12.000000 1.142857 1.000000 1.000000
13.000000 1.142857 1.000000 1.000000
14.000000 1.142857 1.000000 1.000000
15.000000 1.142857 1.000000 1.000000
16.000000 1.142857 1.000000 1.000000
17.000000 1.142857 1.000000 1.000000
18.000000 1.142857 1.000000 1.000000
19.000000 1.142857 1.000000 1.000000
20.000000 1.142857 1.000000 1.000000
21.000000 1.142857 1.000000 1.000000
22.000000 1.142857 1.000000 1.000000
23.000000 1.142857 1.000000 1.000000
24.000000 1.142857 1.000000 1.000000
25.000000 1.142857 1.000000 1.000000
26.000000 1.142857 1.000000 1.000000
27.000000 1.142857 1.000000 1.000000
28.000000 1.142857 1.000000 1.000000
29.000000 1.142857 1.000000 1.000000
30.000000 1.142857 1.000000 1.000000
31.000000 -0.857143 1.000000 2.000000
32.000000 -0.857143 1.000000 2.000000
33.000000 -0.857143 1.000000 2.000000
34.000000 -0.857143 1.000000 2.000000
35.000000 -0.857143 1.000000 2.000000
36.000000 -0.857143 1.000000 2.000000
37.000000 -0.857143 1.000000 2.000000
38.000000 -0.857143 1.000000 2.000000
39.000000 -0.857143 1.000000 2.000000
40.000000 -0.857143 1.000000 2.000000
41.000000 -0.857143 1.000000 2.000000
42.000000 -0.857143 1.000000 2.000000
43.000000 -0.857143 1.000000 2.000000
44.000000 -0.857143 1.000000 2.000000
45.000000 -0.857143 1.000000 2.000000
46.000000 -0.857143 1.000000 2.000000
47.000000 -0.857143 1.000000 2.000000
48.000000 -0.857143 1.000000 3.000000
49.000000 -0.857143 1.000000 3.000000
50.000000 -0.857143 1.000000 3.000000
51.000000 -0.857143 1.000000 3.000000
52.000000 -0.857143 1.000000 3.000000
53.000000 -0.857143 1.000000 3.000000
54.000000 -0.857143 1.000000 3.000000
55.000000 -0.857143 1.000000 4.000000
56.000000 -0.857143 1.000000 4.000000
57.000000 -0.857143 1.000000 4.000000
58.000000 -0.857143 1.000000 4.000000
59.000000 -0.857143 1.000000 4.000000
60.000000 -0.857143 1.000000 4.000000
61.000000 -0.857143 1.000000 4.000000
62.000000 -0.857143 1.000000 4.000000
63.000000 -0.857143 1.000000 4.000000
64.000000 -0.857143 1.000000 4.000000
65.000000 -0.857143 1.000000 4.000000
66.000000 -0.857143 1.000000 4.000000
67.000000 -0.857143 1.000000 4.000000
68.000000 -0.857143 1.000000 4.000000
69.000000 -0.857143 1.000000 4.000000
70.000000 -0.857143 1.000000 4.000000
1.204e+36 permutations required for exhaustive test of t-test 1
Input Design:
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
0.250000 -0.866025 0.000000 -0.000000
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.000000 0.816497
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
-0.250000 -0.288675 -0.707107 -0.408248
Input Contrast:
1.000000 0.000000 0.000000 0.000000
Contrast rank: 1
Dof: 66 original dof: 66
Doing 5000 random permutations
Starting permutation 1 (Unpermuted data)
statistic Maximum: -1.25225
Starting permutation 2
statistic Maximum: 1.14018
Hi Jay,
your command and design files seem fine.
Matt has the feeling that the tfce correction might make the whole thing stall. This might be for various reasons, one of which being a wrong subject ordering. Are you absolutely sure to have concatenated your subjects in GM_mod_merg in the same order as you assume they are in the design.mat? In other words, if you've run VBM with all the default settings, it will have concatenated the subjects in the alphabetical order, which then MUST correspond to the design.mat ordering.
In any case, even if you feel that the order is right, could you please re-run randomise with --debug and let us know the outcome?
Cheers,
Gwenaelle
--- En date de : Jeu 15.4.10, Jay Ives <[log in to unmask]> a écrit :
De: Jay Ives <[log in to unmask]>
Objet: [FSL]
À: [log in to unmask]
Date: Jeudi 15 avril 2010, 11h15
Hi Gwenaelle,
Here are the design files. The command I used was
randomise -i GM_mod_merg_s3 -m GM_mask -o fslvbm -d design.mat -t design.con -T -n 5000 -V
Thanks for the help.
J
----- Original Message -----
From: Gwenaëlle DOUAUD
To: [log in to unmask]
Sent: Tuesday, April 13, 2010 6:52 PM
Subject: Re: [FSL] FSLVBM GLM Setup
Hi Jay,
this should not take that long... There might be indeed something wrong in your design or the set up of your contrasts (though just showing them separately is not saying that much to us).
Could you please attach the design.mat and .con as well as the command you've typed to run randomise?
Cheers,
Gwenaelle
--- En date de : Mar 13.4.10, Jay Ives <[log in to unmask]> a écrit :
De: Jay Ives <[log in to unmask]>
Objet: Re: [FSL] FSLVBM GLM Setup
À: [log in to unmask]
Date: Mardi 13 avril 2010, 0h48
Hi Gwenaëlle (and Mark),
I did use your method 2, but thought I got it wrong because randomise stalled at iteration 2! I gave it overnight, but no progress.
With 70 subjects in 4 groups (of different sizes), is this expected to take a long time? It might have taken forever...
Thx
J
On Tue Apr 13 2:20 , Gwenaëlle DOUAUD sent:
Hi Mark and Jay,
there is no good answer to this question I'm afraid.
Say you've got two subjects in group A, 3 in B and 4 in C, then both approaches are valid:
Method I
A B
1 0
1 0
0 1
0 1
0 1
for the design.mat of the first A and B groups with
1 -1 (A-B)
-1 1 (B-A)
for the design.con
and then repeat for groups B and C, then groups A and C (which is what you did Mark).
Method II
A B C
1 0 0
1 0 0
0 1 0
0 1 0
0 1 0
0 0 1
0 0 1
0 0 1
0 0 1
for the design.mat of the 3 groups with
1 -1 0 (A-B)
-1 1 0 (B-A)
0 1 -1 (B-C)
0 -1 1 (C-B)
1 0 -1 (A-C)
-1 0 1 (C-A)
for the design.con (t-tests)
and
1 0 1 0 0 0
for the design.fts (F-test, as many columns as there are rows in your design.con, you just need to click in "F-tests" in the Glm gui and then click in front of the two relevant "Contrasts" you have already set up)
So with Method II, you can also ask the question of where are the changes *across the 3 groups* (F-test with the design.fts). You also get an increase in DoF but, as Tom Nichols said, if it happens that group C for instance has wildly smaller variance, you can get inflated significances (or reduced power if it has wildly larger variance).
So it depends on what your main question is, really.
Hope this helps,
Gwenaelle
--- En date de : Lun 12.4.10, Mark Walterfang <[log in to unmask]> a écrit :
> De: Mark Walterfang <[log in to unmask]>
> Objet: Re: [FSL] FSLVBM GLM Setup
> À: [log in to unmask]
> Date: Lundi 12 avril 2010, 14h18
> Hi all
>
> I'm in the same situation as Jay. I have three groups
> (illness1, illness2
> and controls), all matched to each other. I've run three
> two-way analyses,
> which is pretty laborious and I'm pretty sure it's not
> statistically ideal.
> What I can't work out is how to set up the design matrices
> & contrasts in
> the way Jay describes, as the online manual for Randomise
> doesn't really
> provide guidance here. Gwenaëlle, is this something you
> can advise on?
>
> Thanks in advance,
>
> Mark Walterfang
>
>
> On 10/4/10 12:29 PM, "Jay Ives" <[log in to unmask]>
> wrote:
>
> > I have 70 subjects in 4 groups and would like to test
> between individual
> > groups and combinations of the groups. Can someone
> please advise me how to set
> > up the design.mat and design.con files to do this?
> Thx
|