The contrast is not specified in the fsgd. The fsgd just declares that you have two classes. The contrast is specified in your .mat file. The +1 -1 means that you will subtract ASD from Normals.
doug
Robert Levy wrote:
Hello,
I ran a group contrast in an ROI, correcting for multiple comparisons using the cluster size threshold method. It seems to have worked, however the one element that I don't understand as well is the basic between group contrast using an fsgd file, so I'm wondering if my commands really do what I want them to do. Is the below a correct usage to accomplish that? (apologies again for average7, it is because the data is relatively old.) The output looks very similar to the results of the contrast using intergroup_avg so that is promising. The main thing I'm uncertain about is how the fsgd file specifes that the two classes are to be contrasted.
Thanks, Rob
# perform the actual contrast mri_glmfit \ --really-use-average7 \ --y rh.ASvfix_8s_BwGroupError5.fsfast.mgh \ --fsgd fsgd_and_contrast_files/BwGroupError5.fsgd \ --glmdir rh.ASvfix_8s_BwGroupError5.fsfast.glmdir \ --pca \ --surf average7 rh \ --C fsgd_and_contrast_files/BwGroupError5.mat \ --label average7/label/dACC_rACC_final-rh.label \ --fwhm 13.488420
Contents of matfile: [levy@vantz fsgd_and_contrast_files]$ cat BwGroupError5.mat +1 -1
Contents of fsgd file: [levy@vantz fsgd_and_contrast_files]$ cat BwGroupError5.fsgd GroupDescriptorFile 1 Title HCvASD Class Normals Class ASD
Input mano001 Normals Input mano002 Normals Input mano004 Normals Input mano005 Normals Input mano007 Normals Input mano008 Normals Input mano009 Normals Input mano010 Normals Input mano011 Normals Input mano012 Normals Input mano013 Normals Input mano014 Normals Input mano016 Normals Input mano025 Normals Input manoA02 ASD Input manoA03 ASD Input manoA04 ASD Input manoA05 ASD Input manoA07 ASD Input manoA08 ASD Input manoA09 ASD Input manoA10 ASD Input manoA11 ASD Input manoA13 ASD
And just for the sake of completenes, here is a sample of what I ran after that:
# monte carlo simulation data mri_glmfit --really-use-average7 \ --y rh.ASvfix_8s_BwGroupError5.fsfast.mgh \ --fsgd fsgd_and_contrast_files/BwGroupError5.fsgd \ --glmdir rh.ASvfix_8s_BwGroupError5.fsfast.glmdir \ --surf average7 rh \ --C fsgd_and_contrast_files/BwGroupError5.mat \ --sim mc-full 1000 2 BwGroupError5_ASvfix_8s_rhMCSIM01 \ --fwhm 13.488420 \ --label average7/label/dACC_rACC_final-rh.label ( x10 )
# apply simulation data to show significant clusters mri_surfcluster \ --src rh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.mgh \ --csd BwGroupError5_ASvfix_8s_rhMCSIM01-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM02-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM03-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM04-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM05-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM06-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM07-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM08-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM09-BwGroupError5.csd \ --csd BwGroupError5_ASvfix_8s_rhMCSIM10-BwGroupError5.csd \ --sum rh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM.sum \ --ocp rh.ASvfix_8s_BwGroupError5.fsfast.glmdir/BwGroupError5/sig.clusterMCSIM.mgh \ --thmax infinity