I suspect it has somenthing to do with the column that refers to age. Maybe the demean is wrong...because if I take off that column like this.
Ah - I've just noticed that you have *commas* instead of decimal points!!On 31 Jul 2012, at 16:01, sandra petris wrote:I get this:[fsl@localhost stats]$ fslinfo all_FA_skeletoniseddata_type FLOAT32dim1 182dim2 218dim3 182dim4 15datatype 16pixdim1 1.0000000000pixdim2 1.0000000000pixdim3 1.0000000000pixdim4 1.0000000000cal_max 0.0000cal_min 0.0000file_type NIFTI-1+2012/7/31 Stephen Smith <[log in to unmask]>
I think you may need spaces not tabs….I'm not sure.What do you get fromfslinfo all_FA_skeletonisedCheersOn 31 Jul 2012, at 15:55, sandra petris wrote:yes I have but I used tabs to separate the columns...
2012/7/31 Stephen Smith <[log in to unmask]>
Do you have spaces between the 0 and -1 entries in your design.con ? You need that.On 31 Jul 2012, at 15:51, sandra petris wrote:Hi Steve sorry to disturb you againnow my matrices are the following:/NumWaves 5/NumPoints 15/PPheights 1 1 1 1 1/Matrix1 1,0434 1,1634 1,9836 -4,73331 0,3511 0,6998 0,6660 -6,73331 0,3733 0,1620 1,3627 4,26671 -0,7298 1,1634 0,9047 -3,73331 -0,3459 -0,1429 0,6176 -1,73331 -0,0511 0,1898 -0,1555 -3,73331 -0,0954 -2,1662 0,0961 -4,73331 -0,7031 -1,0768 -0,6896 -5,73331 0,3342 0,4163 -0,7304 9,26671 0,8755 -0,1218 -0,4432 -2,73331 -1,3132 -1,0332 0,5826 6,26671 0,4781 -0,9344 0,2008 -0,73331 -0,0692 0,1785 -1,1205 18,26671 0,6799 1,1634 -1,0984 1,26671 -0,8272 0,3381 -2,1765 -4,7333/NumWaves 5/NumContrasts 6/PPheights 1 1 1 1 1/Matrix0 1 0 0 00-1 0 0 00 0 1 0 00 0-1 0 00 0 0 1 00 0 0-1 0I run the randomise -i all_FA_skeletonised -o tbss -m mean_FA_skeleton_mask -d design_nico.mat -t design_nico.con -n 10000 --T2 -V (without -D)and I got the following error message :(Loading Data:Data loadedERROR: Program failedAn exception has been thrownRuntime error:- detected by Newmat: process fails to convergeMatrixType = Rect # Rows = 15; # Cols = 4Trace: SVD; pinv.2012/7/31 Stephen Smith <[log in to unmask]>No - I meant you don't need BOTH - you need one or the other.Cheers.On 31 Jul 2012, at 15:36, sandra petris wrote:ah thanks a lot Steve! I'm gonna delete the -D option and the 1st colums, let'see what happensciao! Sandra2012/7/31 Stephen Smith <[log in to unmask]>
Hi - the answer is clearly in the error message :-)You don't need the -D option *and* the EV full of 1s.Steve.On 31 Jul 2012, at 12:34, sandra petris wrote:Hi Stephen,when I run randomise -i all_FA_skeletonised -o tbss -m mean_FA_skeleton_mask -d design_nico.mat -t design_nico.con -n 10000 -D --T2 -V I got the following error message:Loading Data:Data loadedWarning: You have demeaned your data, but at least one design column has non-zero meanERROR: Program failedAn exception has been thrownRuntime error:- detected by Newmat: process fails to convergeMatrixType = Rect # Rows = 15; # Cols = 4Trace: SVD; pinv.thanks in advance for any helpSandra2012/7/31 Stephen Smith <[log in to unmask]>
Hi - these look - what was the error message?On 30 Jul 2012, at 16:14, Sandra Petris wrote:Dear fsl experts,
I have a question regarding TBSS analysis. I would like to compute the correlation (or regression) between FA and behavioral scores.
I have one group of subjects (15 subjects) and behavioral scores (BS) taken from 3 tasks plus 1 regressors of no interest, age that I have demeaned (is it correct?)
I assume that the design matrix should look like this:
/NumWaves 5
/NumPoints 15
/PPheights 1 1 1 1 1
/Matrix
1 1,0434 1,1634 1,9836 -4,7333
1 0,3511 0,6998 0,6660 -6,7333
1 0,3733 0,1620 1,3627 4,2667
1 -0,7298 1,1634 0,9047 -3,7333
1 -0,3459 -0,1429 0,6176 -1,7333
1 -0,0511 0,1898 -0,1555 -3,7333
1 -0,0954 -2,1662 0,0961 -4,7333
1 -0,7031 -1,0768 -0,6896 -5,7333
1 0,3342 0,4163 -0,7304 9,2667
1 0,8755 -0,1218 -0,4432 -2,7333
1 -1,3132 -1,0332 0,5826 6,2667
1 0,4781 -0,9344 0,2008 -0,7333
1 -0,0692 0,1785 -1,1205 18,2667
1 0,6799 1,1634 -1,0984 1,2667
1 -0,8272 0,3381 -2,1765 -4,7333
Is the following contrast matrix correct?
/NumWaves 5
/NumContrasts 6
/PPheights 1 1 1 1 1
/Matrix
0 1 0 0 0
0 -1 0 0 0
0 0 1 0 0
0 0 -1 0 0
0 0 0 1 0
0 0 0 -1 0
When I run the randomize command I receive an error message.....are the NumWaves/NumConntrast an PPWheights correct?
Any help will be greatly appreciated
Best, Sandra
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------