Hey there,
The first thing to say is you shouldn't be running mixed
effects analyses on only 3 observations. You end up fitting 3 parameters
at each voxel, so you need _minimum_ 5 data points at each voxel to do a
mixed effects analysis. (Think about trying to make a statement about a
whole population when you have met only 3 .). I'm pretty sure this is
what's causing the errors.
We're not sure why this is taking so long. In our experience, simple
anayses like this might take an hour. It may be a knock on effect of the
error.
Sorry we can't be more helpful
Mark and Tim
>
> ---------- Forwarded message ----------
> Date: Mon, 26 Jan 2004 13:28:31 -0800
> From: Zrinka Bilusic <[log in to unmask]>
> Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
> To: [log in to unmask]
> Subject: [FSL] ndtri domain error
>
> Hello everybody,
> I am forwarding this email from Sam Harris here at UCLA, who has
> problems posting to the mailing list... Here are the details of the
> higher level analysis:
> >
> > At the lower level we have three 10 minute functional scans. Each of
> > which is an event-related paradigm with 3 EVs and 12 contrasts
> >
> > The higher level analysis was set up like this:
> >
> > Input 1 1(group) 1(EV1)
> > Input 2 1(group) 1(EV1)
> > Input 3 1(group) 1(EV1)
> >
>
> This analysis was started on Jan 22 at 13:09 (pacific time) and it is
> still running. As I am writing this, the analysis is at 40%, cope 6. So
> we are about half way through. First question - would you expect
> analysis like this to take that long (it is running on a new G5).
>
> The second question has to do with the "ndtri domain error" that seems
> to be happening at various points during the analysis, but has appeared
> consistently for all 6 copes. The analysis is still going, so this
> error is not quitting the analysis, which makes it even more
> troublesome. Below is a copy of the feat output where you can see the
> error. As you can see, for each cope, this error appears at different
> times during the analysis. Any ideas why this is happening?
>
> Thank you very much,
>
> zrinka
>
> >>>>>
> >>>>>
> >>>>> Starting higher-level stats...
> >>>>> Started FEAT at Thu Jan 22 13:09:25 PST 2004 on G5.local.
> >>>>>
> >>>>> /bin/cp
> >>>>> /Users/cohen/Harris/DATA/Belief_Pilot/Belief_Pilot_001/
> >>>>> BOLD_64_4.gfeat/design.fsf
> >>>>> design.fsf
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/feat_model design
> >>>>>
> >>>>> /bin/rm -f example_func.hdr example_func.img
> >>>>>
> >>>>> ln -s ../bg_image.hdr example_func.hdr
> >>>>>
> >>>>> ln -s ../bg_image.img example_func.img
> >>>>>
> >>>>> /bin/rm -f mask.hdr mask.img
> >>>>>
> >>>>> ln -s ../mask.hdr mask.hdr
> >>>>>
> >>>>> ln -s ../mask.img mask.img
> >>>>>
> >>>>> /bin/rm -f filtered_func_data.hdr filtered_func_data.img
> >>>>>
> >>>>> ln -s ../cope1.hdr filtered_func_data.hdr
> >>>>>
> >>>>> ln -s ../cope1.img filtered_func_data.img
> >>>>>
> >>>>> /bin/rm -f var_filtered_func_data.hdr var_filtered_func_data.img
> >>>>>
> >>>>> ln -s ../varcope1.hdr var_filtered_func_data.hdr
> >>>>>
> >>>>> ln -s ../varcope1.img var_filtered_func_data.img
> >>>>>
> >>>>> /bin/rm -f tdof_filtered_func_data.hdr tdof_filtered_func_data.img
> >>>>>
> >>>>> ln -s ../tdof_t1.hdr tdof_filtered_func_data.hdr
> >>>>>
> >>>>> ln -s ../tdof_t1.img tdof_filtered_func_data.img
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/flame --cope=filtered_func_data
> >>>>> --vc=var_filtered_func_data --mask=mask
> >>>>> --ld=stats --dm=design.mat --cs=design.grp --tc=design.con --ols
> >>>>> --nj=10000 --bi=500 --se=1 --fm
> >>>>> --zlt=1.55 --zut=1.95
> >>>>> Log directory is: stats
> >>>>> Setting up:
> >>>>> No f contrasts
> >>>>> nevs=1
> >>>>> ntpts=3
> >>>>> ngs=1
> >>>>> nvoxels=4093971
> >>>>> Running:
> >>>>> nmaskvoxels=168219
> >>>>> njumps = 10000
> >>>>> burnin = 500
> >>>>> sampleevery = 1
> >>>>> nsamples = 9500
> >>>>>
> >>>>> Metropolis Hasting Sampling
> >>>>> Number of voxels=168219
> >>>>> Percentage done:
> >>>>> 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
> >>>>> ndtri domain error
> >>>>> 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
> >>>>> 45 46 47 48 49 50 51 52 53 54
> >>>>> 55 56
> >>>>> ndtri domain error
> >>>>> 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
> >>>>> 79 80 81 82 83 84 85 86 87 88
> >>>>> 89 90 91
> >>>>> ndtri domain error
> >>>>> 92 93 94 95 96 97 98 99 100
> >>>>> Saving results
> >>>>>
> >>>>> Log directory was: stats
> >>>>>
> >>>>> echo 2 > stats/dof
> >>>>>
> >>>>> /bin/rm -f stats/zem* stats/zols* stats/mask*
> >>>>>
> >>>>> gzip -9 stats/mean* stats/fdof* stats/tstat* stats/fstat*
> >>>>> stats/fdof*: No such file or directory
> >>>>> stats/fstat*: No such file or directory
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/smoothest -d 2 -m mask -r stats/res4d >
> >>>>> stats/smoothness
> >>>>>
> >>>>> rm -f stats/res4d*
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/avwmaths stats/zstat1 -mas mask
> >>>>> thresh_zstat1
> >>>>>
> >>>>> echo 4093971 > thresh_zstat1.vol
> >>>>> zstat1: DLH=0.0158668 VOLUME=4093971 RESELS=320.061
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/cluster -i thresh_zstat1 -c stats/cope1 -t
> >>>>> 1.6 -p 0.05 -d 0.0158668
> >>>>> --volume=4093971 --othresh=thresh_zstat1 -o cluster_mask_zstat1
> >>>>> --connectivity=26 --mm
> >>>>> --olmax=lmax_zstat1_tal.txt > cluster_zstat1_tal.txt
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/cluster2html . cluster_zstat1 -tal
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/avwstats thresh_zstat1 -l 0.0001 -R
> >>>>> 0.000000 0.000000
> >>>>> Rendering using zmin=2.3 zmax=8
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/overlay 1 0 example_func -a thresh_zstat1
> >>>>> 2.3 8 rendered_thresh_zstat1
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/slicer rendered_thresh_zstat1 -S 2 750
> >>>>> rendered_thresh_zstat1.png
> >>>>>
> >>>>> /bin/cp /usr/local/bin/fsl/etc/luts/ramp.gif .ramp.gif
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/tsplot . -f filtered_func_data
> >>>>>
> >>>>> mkdir tsplot
> >>>>>
> >>>>> Finished FEAT at Thu Jan 22 21:23:08 PST 2004
> >>>>> To view the FEAT report point your web browser at
> >>>>> /Users/cohen/Harris/DATA/Belief_Pilot/Belief_Pilot_001/
> >>>>> BOLD_64_4.gfeat/cope1.feat/report.html
> >>>>>
> >>>>> Started FEAT at Thu Jan 22 21:23:10 PST 2004 on G5.local.
> >>>>>
> >>>>> /bin/cp
> >>>>> /Users/cohen/Harris/DATA/Belief_Pilot/Belief_Pilot_001/
> >>>>> BOLD_64_4.gfeat/cope1.feat/design.fsf
> >>>>> design.fsf
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/feat_model design
> >>>>>
> >>>>> /bin/rm -f example_func.hdr example_func.img
> >>>>>
> >>>>> ln -s ../bg_image.hdr example_func.hdr
> >>>>>
> >>>>> ln -s ../bg_image.img example_func.img
> >>>>>
> >>>>> /bin/rm -f mask.hdr mask.img
> >>>>>
> >>>>> ln -s ../mask.hdr mask.hdr
> >>>>>
> >>>>> ln -s ../mask.img mask.img
> >>>>>
> >>>>> /bin/rm -f filtered_func_data.hdr filtered_func_data.img
> >>>>>
> >>>>> ln -s ../cope2.hdr filtered_func_data.hdr
> >>>>>
> >>>>> ln -s ../cope2.img filtered_func_data.img
> >>>>>
> >>>>> /bin/rm -f var_filtered_func_data.hdr var_filtered_func_data.img
> >>>>>
> >>>>> ln -s ../varcope2.hdr var_filtered_func_data.hdr
> >>>>>
> >>>>> ln -s ../varcope2.img var_filtered_func_data.img
> >>>>>
> >>>>> /bin/rm -f tdof_filtered_func_data.hdr tdof_filtered_func_data.img
> >>>>>
> >>>>> ln -s ../tdof_t2.hdr tdof_filtered_func_data.hdr
> >>>>>
> >>>>> ln -s ../tdof_t2.img tdof_filtered_func_data.img
> >>>>>
> >>>>> /usr/local/bin/fsl/bin/flame --cope=filtered_func_data
> >>>>> --vc=var_filtered_func_data --mask=mask
> >>>>> --ld=stats --dm=design.mat --cs=design.grp --tc=design.con --ols
> >>>>> --nj=10000 --bi=500 --se=1 --fm
> >>>>> --zlt=1.55 --zut=1.95
> >>>>> Log directory is: stats
> >>>>> Setting up:
> >>>>> No f contrasts
> >>>>> nevs=1
> >>>>> ntpts=3
> >>>>> ngs=1
> >>>>> nvoxels=4093971
> >>>>> Running:
> >>>>> nmaskvoxels=649276
> >>>>> njumps = 10000
> >>>>> burnin = 500
> >>>>> sampleevery = 1
> >>>>> nsamples = 9500
> >>>>>
> >>>>> Metropolis Hasting Sampling
> >>>>> Number of voxels=649276
> >>>>> Percentage done:
> >>>>> 1 2 3 4 5 6
> >>>>> ndtri domain error
> >>>>> 7 8 9 10 11 12 13 14 15
> >>>>> ndtri domain error
> >>>>> 16 17 18 19 20 21 22 23 24
> >>>>> ndtri domain error
> >>>>> 25 26 27 28 29 30 31 32 33
> >>>>> ndtri domain error
> >>>>> 34 35 36 37 38 39 40 41 42
> >>>>> ndtri domain error
> >>>>> 43
> >>>>>
> >>>>>
>
>
--
-------------------------------------------------------------------------------
Tim Behrens
Centre for Functional MRI of the Brain
The John Radcliffe Hospital
Headley Way Oxford OX3 9DU
Oxford University
Work 01865 222782
Mobile 07980 884537
-------------------------------------------------------------------------------
|