External Email - Use Caution
Hi Doug,
Thank you for your response. I tried the link which you sent yesterday. But got stuck while creating qdec.fsgd file. Below i have pasted the terminal output after loading the data.
1 Left-Lateral-Ventricle continuous 0 2 Left-Inf-Lat-Vent continuous 0 3 Left-Cerebellum-White-Matter continuous 0 4 Left-Cerebellum-Cortex continuous 0 5 Left-Thalamus-Proper continuous 0 6 Left-Caudate continuous 0 7 Left-Putamen continuous 0 8 Left-Pallidum continuous 0 9 3rd-Ventricle continuous 0 10 4th-Ventricle continuous 0 11 Brain-Stem continuous 0 12 Left-Hippocampus continuous 0 13 Left-Amygdala continuous 0 14 CSF continuous 0 15 Left-Accumbens-area continuous 0 16 Left-VentralDC continuous 0 17 Left-vessel continuous 0 18 Left-choroid-plexus continuous 0 19 Right-Lateral-Ventricle continuous 0 20 Right-Inf-Lat-Vent continuous 0 21 Right-Cerebellum-White-Matter continuous 0 22 Right-Cerebellum-Cortex continuous 0 23 Right-Thalamus-Proper continuous 0 24 Right-Caudate continuous 0 25 Right-Putamen continuous 0 26 Right-Pallidum continuous 0 27 Right-Hippocampus continuous 0 28 Right-Amygdala continuous 0 29 Right-Accumbens-area continuous 0 30 Right-VentralDC continuous 0 31 Right-vessel continuous 0 32 Right-choroid-plexus continuous 0 33 WM-hypointensities continuous 0 34 Optic-Chiasm continuous 0 35 CC_Posterior continuous 0 36 CC_Mid_Posterior continuous 0 37 CC_Central continuous 0 38 CC_Mid_Anterior continuous 0 39 CC_Anterior continuous 0 40 diagnosis discrete 2 1 Schizophrenia 2 Control 41 Age continuous 0 42 Birthplace discrete 3 1 Statutory 2 Census 3 Rural 43 Upbringingplace discrete 3 1 Statutory 2 Census 3 Rural 44 Livingplace discrete 3 1 Statutory 2 Census 3 Rural Continuous Factors: Mean: StdDev: ------------------- ----- ------- Left-Lateral-Ventricle 6784.459 3582.583 Left-Inf-Lat-Vent 363.189 163.344 Left-Cerebellum-White-Matter 13393.085 2875.553 Left-Cerebellum-Cortex 50654.337 7634.479 Left-Thalamus-Proper 7044.259 1254.886 Left-Caudate 3267.819 728.738 Left-Putamen 5221.959 863.333 Left-Pallidum 1961.781 357.087 3rd-Ventricle 1068.711 447.579 4th-Ventricle 1673.963 641.339 Brain-Stem 18975.526 3907.934 Left-Hippocampus 3852.407 684.810 Left-Amygdala 1625.463 378.770 CSF 983.626 305.507 Left-Accumbens-area 587.848 155.803 Left-VentralDC 3977.841 673.114 Left-vessel 28.378 21.035 Left-choroid-plexus 558.441 330.333 Right-Lateral-Ventricle 6224.989 3745.143 Right-Inf-Lat-Vent 440.070 168.625 Right-Cerebellum-White-Matter 13111.263 2744.138 Right-Cerebellum-Cortex 52495.385 7167.728 Right-Thalamus-Proper 6732.170 1748.668 Right-Caudate 3305.363 949.074 Right-Putamen 4908.819 1155.840 Right-Pallidum 1774.204 489.167 Right-Hippocampus 3766.470 1019.353 Right-Amygdala 1686.874 437.019 Right-Accumbens-area 539.763 151.780 Right-VentralDC 3818.130 666.970 Right-vessel 23.344 16.649 Right-choroid-plexus 620.948 414.269 WM-hypointensities 4434.148 10311.009 Optic-Chiasm 198.937 65.369 CC_Posterior 973.467 157.378 CC_Mid_Posterior 637.400 597.617 CC_Central 625.374 497.642 CC_Mid_Anterior 681.052 419.940 CC_Anterior 870.274 237.412 Age 31.963 7.019
Number of subjects: 27 Number of factors: 44 (4 discrete, 40 continuous) Number of classes: 54 Number of regressors: 2214 ============================================================ Data table loading completed successfully. SUBJECTS_DIR is '/usr/local/freesurfer/subjects' lh-Avg-Intercept-thickness ----------------------- Does the average thickness differ from zero? Nuisance factors: Age 1.000 1.000 0.000 0.000 0.000 0.000;
lh-Avg-thickness-Right-Amygdala-Cor ----------------------- Does the correlation between thickness and Right-Amygdala, accounting for diagnosis, differ from zero? Nuisance factors: Age 0.000 0.000 1.000 1.000 0.000 0.000;
lh-Diff-Schizophrenia-Control-Intercept-thickness ----------------------- Does the average thickness differ between Schizophrenia and Control? Nuisance factors: Age 1.000 -1.000 0.000 0.000 0.000 0.000;
lh-Diff-Schizophrenia-Control-Cor-thickness-Right-Amygdala ----------------------- Does the thickness--Right-Amygdala correlation differ between Schizophrenia and Control? Nuisance factors: Age 0.000 0.000 1.000 -1.000 0.000 0.000;
I highlighted the discrete factors in yellow color. I wanted to know how to go about this classes such as birthplace, living and upbringing place.
Also, let us know whether the below-pasted information is in the right format for qdec.fsgd file?? Here we have birth, living and upbringing places as main classes and statutory, census and rural as subclasses. So, my question is that the below-pasted classes are ok or not?? If not, just needed your assistance to create fsgd file along with contrasts files. Finally, can I use the qdec gui for this analysis or just the command-line?
*GroupDescriptorFile 1* *Title Untitled* *MeasurementName thickness* *Class BirthplaceStatutory* *Class BirthplaceCensus* *Class BirthplaceRural* *Class UpbringingplaceStatutory* *Class UpbringingplaceCensus* *Class UpbringingplaceRural* *Class LivingplaceStatutory* *Class LivingplaceCensus* *Class LivingplaceRural* *# Class schizophrenia* *# Class control* *Input 1 BirthplaceStatutory UpbringingplaceStatutory LivingplaceStatutory* *Input 3 BirthplaceCensus UpbringingplaceCensus LivingplaceCensus* *Input 4 BirthplaceRural UpbringingplaceRural LivingplaceRural* *Input 5 BirthplaceStatutory UpbringingplaceStatutory LivingplaceStatutory* *Input 6 BirthplaceRural UpbringingplaceCensus LivingplaceCensus* *Input 7 BirthplaceCensus UpbringingplaceCensus LivingplaceCensus* *Input 8 BirthplaceRural UpbringingplaceRural LivingplaceRural* *Input 9 BirthplaceRural UpbringingplaceRural LivingplaceCensus* *Input 10 BirthplaceStatutory UpbringingplaceStatutory LivingplaceStatutory* *Input 11 BirthplaceStatutory UpbringingplaceStatutory LivingplaceCensus* *Input 13 BirthplaceCensus UpbringingplaceCensus LivingplaceCensus* *Input 14 BirthplaceCensus UpbringingplaceCensus LivingplaceCensus* *Input 15 BirthplaceCensus UpbringingplaceCensus LivingplaceCensus* *Input 18 BirthplaceRural UpbringingplaceCensus LivingplaceCensus*
* Thank you*
*Best* *Vittal*
On Mon, Oct 15, 2018 at 7:45 PM vittal korann vittalkorann@gmail.com wrote:
Hi Doug,
Recently I started working on freesurfer for volume based analysis. For QDEC, I need to include more than 2 variables. For example if I include patients who hail from different places like statutory town, census town and rural. In present scenario, can I include all three as covariates. Here is the error that qdec showing while running GLM:
"ERROR: Subject 2 seems to have a mismatched number of discrete factors (expected 4, found 1)" "Reading source surface /usr/local/freesurfer/subjects/fsaverage/surf/lh.white ERROR: no contrasts specified." "Error in Analyze: command fai;ed mri_glmfit --y /usr/local/freesurfer/subjects/qdec/qdecAnalys/y.mgh --fsgd /usr/local/freesurfer/subjects/qdec/qdecAnalys/qdec.fsgd dods -glmdir /usr/local/freesurfer/subjects/qdec/qdecAnalys/ -surf fsaverage lh -label /usr/local/freesurfer/subjects/fsaverage/label/lh.aparc.label"
Could you help us out in this regard.
Would appreciate any help!!
Korann