Dear Shary,
At 05:10 PM 5/17/2005 -0400, (Mr.) Shahryar (Shary) Rafi-Tari wrote:
>Sorry to bother you again. I am really stuck now. I am comparing two
>groups each of which have 20 subjects (df=38) with SnPM. When the number
>of permutations is 100 (arbitrary), I can run suprathreshold cluster
>tests. However, when I permute 1000 times, the program gives me the
>following error message:
>
>
>Loading & conditioning SupraThreshold statistics...??? Error using ==> load
>Unable to read MAT-file
>C:\Files\NAAMA\Results\PermutationSNPMMethods\0PureADVsSeverWMD1000Perm10mmFWHMSupraThresholdClusterTest\SnPM_ST.mat
>
>File may be corrupt.
>
>Error in ==> snpm_pp at 752
> load(fullfile(CWD,'SnPM_ST'))
>
>??? Error while evaluating uicontrol Callback.
>
>I smoothed the variance with 10mm FWHM (10 10 10) in both cases. I am
>using the same set of images which have been processed in SPM2. I
>repeated this with the two permutations twice and the error message is
>persistent in the case of 1000 permutations.
>
>Why is this happening?
The first thing I suggest is to check the size of a file called
"SnPM_ST.mat" in your working directory. If the file is larger than a few
hundred megabytes, then the file may be too big to be used in SnPM. This
SnPM_ST file is a collection of voxel values and locations for high
t-scores (the tip of t-image iceberg) from all the permutations, to be used
in a cluster size test later on. More permutations you run, larger this
file becomes. Since you didn't have any problem with 100 permutations, I
suspect this SnPM_ST is growing too big to be handled when you run 1000
permutations. From my experience, using variance smoothing (or pseudo-t)
seems to accelerate the growth of SnPM_ST, especially when there aren't
large activation areas.
If a large SnPM_ST is the culprit, then a possible remedy is to change the
variable called "SnPMdefs.STprop" in snpm_defaults.m. The default is 0.10,
but try a smaller number (eg., 0.05, 0.025, and so on) until you no longer
see the error message.
Good luck!
-Satoru
Satoru Hayasaka ==============================================
Post-Doctoral Fellow, MR Unit, UCSF / VA Medical Center
Email: shayasak_at_itsa_dot_ucsf_dot_edu Phone:(415) 221-4810 x4237
Homepage: http://www.sph.umich.edu/~hayasaka
==============================================================
|