Hi Pablo, unfortunately, you've run into a bug in FSFAST in that some contrast matrices cannot be easily run. It uses some complicated (overly complicated?) logic behind the scenes to create the contrast matrices, and using a custom contrast matrix is not so easy. I'll write a patch to fix this, but it might take a week or two.
doug
Pablo Polosecki wrote:
Dear list:
I am analyzing functional data using FSFAST. I have 7 conditions in a blocked design and I would like to run an F-test using the following contrast matrix: [ -0.5000 0 0.5000 0 -0.5000 0.5000 0; 0 -1.0000 0 1.0000 0 0 0]
For this I am manually creating a matlab file with the contrast matrix in the “mystudy” folder. (Is there a way of having FSFAST create the matrix for me using mkanalysis-sess ?) I am attaching the matlab m-file with the matrix, which contains this matrix as the ContrastMtx_0 variable. In the matlab file, I set sumconds to 0. I am unsure as what to waht the WCond matrix should be in the matlab file. I am making it equal to the ContrastMtx_0, but this might be wrong.
Afterwards I run the analysis using selxavg3-sess, from matlab using (“rois” is the name of the analysis): unix([FS_HOME '/fsfast/bin/selxavg3-sess -analysis rois -sf sf -df df -float -overwrite -monly tmp_gen_rois.m']); tmp_gen_rois;
Within text output, I get the following: Computing contrast matrices …(some stuff regarding other contrast matrices)… 3 main_faces J= 3 eff = 287636151.3 vrf = 862908453.9536 vrfffx = 852078020.3613 r = 1.01
This last line says that for main_faces J=3, but my contrast matrix has 2 rows (I understand J is the number of lines in the matrix)! Could someone point out what I am doing wrong here?
Thank you very much
Pablo
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer