I've just created the following docs for mri_mcsim
doug
mri_mcsim
--o top-output-dir
--base csdbase
--surface subjectname hemi
--nreps nrepetitions
--fwhm FWHM <FWHM2 FWHM3 ...>
--fwhm-max FWHMMax : sim with fwhm=1:FWHMMax (default 30)
--avgvtxarea : report cluster area based on average vtx area
--seed randomseed : default is to choose based on ToD
--label labelfile : default is ?h.cortex.label
--mask maskfile : instead of label
--no-label : do not use a label to mask
--no-save-mask : do not save mask to output (good for mult jobs)
--surfname surfname : default is white
--log LogFile
--done DoneFile : will create DoneFile when finished
--stop stopfile : default is ourdir/mri_mcsim.stop
--save savefile : default is ourdir/mri_mcsim.save
--save-iter : save output after each iteration
--sd SUBJECTS_DIR
--debug turn on debugging
--checkopts don't run anything, just check options and exit
--help print out information on how to use this program
--version print out version and exit
$Id: mri_mcsim.c,v 1.26 2014/04/11 15:31:56 greve Exp $
This program computes tables for performing multiple comparisons on
the surface based on a monte carlo simulation using white gaussian
noise smoothed on the surface. mri_mcsim was used to generate the
tables found in $FREESURFER_HOME/average/mult-comp-cor and used by
mri_glmfit-sim with the --cache option. The tables in mult-comp-cor
are for fsaverage and fsaverage_sym for a search space over the whole
cortex. mri_mcsim can be used to create new tables for a new subject
or for fsaverage/fsaverage_sum with a smaller search space.
Example 1: Create tables for a new subject for whole hemisphere
mri_mcsim --o /path/to/mult-comp-cor/newsubject/lh/cortex --base mc-z
--save-iter --surf newsubject lh --nreps 10000
This may take hours (or even days) to run; see below for
parallelizing. When running mri_glmfit-sim, add --cache-dir
/path/to/mult-comp-cor
Example 2: Create tables for superior temporal gyrus for fsaverage
First, create the label by running
mri_annotation2label --subject fsaverage --hemi lh --outdir labeldir
mri_mcsim --o /path/to/mult-comp-cor/fsaverage/lh/superiortemporal
--base mc-z
--save-iter --surf fsaverage lh --nreps 10000
--label labeldir/lh.superiortemporal.label
When running mri_glmfit, make sure to use --label
labeldir/lh.superiortemporal.label
When running mri_glmfit-sim, add --cache-dir /path/to/mult-comp-cor
--cache-label superiortemporal
Example 3: running simulations in parallel (two jobs, 5000 iterations
each for a total of 10000)
mri_mcsim --o /path/to/mult-comp-cor/fsaverage/lh/superiortemporal
--base mc-z.j001
--save-iter --surf fsaverage lh --nreps 5000
--label labeldir/lh.superiortemporal.label
mri_mcsim --o /path/to/mult-comp-cor/fsaverage/lh/superiortemporal
--base mc-z.j002
--save-iter --surf fsaverage lh --nreps 500
--label labeldir/lh.superiortemporal.label
When those jobs are done, merge the results into a single table with
mri_surfcluster
--csd
/path/to/mult-comp-cor/fsaverage/lh/superiortemporal/fwhm10/abs/th20/mc-z.j001.csd
--csd
/path/to/mult-comp-cor/fsaverage/lh/superiortemporal/fwhm10/abs/th20/mc-z.j002.csd
--csd-out
/path/to/mult-comp-cor/fsaverage/lh/superiortemporal/fwhm10/abs/th20/mc-z.csd
--csdpdf
/path/to/mult-comp-cor/fsaverage/lh/superiortemporal/fwhm10/abs/th20/mc-z.cdf
--csdpdf-only
Repeat the above command for each FWHM, sign (pos, neg, abs) and threshold
On 04/11/2014 11:08 AM, Douglas N Greve wrote:
> That should have been --nreps 10000 (not 100:)
>
> On 04/11/2014 11:04 AM, Douglas N Greve wrote:
>> On 04/11/2014 02:52 AM, Gautam Tammewar wrote:
>>> Hi all,
>>>
>>> I'm trying to run Monte Carlo simulations on cortical thickness data
>>> I got using mri_glmfit, but I'm running into some errors.
>>>
>>> 1) I used the command
>>>
>>> mri_glmfit-sim --glmdir
>>> lh.thickness.fwhm20.diagnosis_scanner.dods.glmdir/ --cache 2.0 pos
>>>
>>> The error I receive says "cannot find
>>> /usr/local/freesurfer_x86_64-5.1.0/average/mult-comp-cor/fsaverage/lh/cortex/fwhm40/pos/th20/mc-z.csd"
>> Even though you've smoothed by 20, the actual FWHM in the data is 40,
>> which is pretty high, and the tables do not go that high. This may
>> indicate that there is something wrong with your data
>>> I'm wondering why the script is looking in the directory "fwhm40"
>>> when the data is smoothed using FWHM 20. The directories available
>>> under fsaverage/lh/cortex go from fwhm01 to fwhm30.
>>>
>>> 2) Also, for another test I'm running, I created my own average
>>> subjects (instead of using fsaverage). Is there a way to generate the
>>> files that are created under /average/mult-comp-cor for this average
>>> subject so that I don't have to use the "--sim" flag to run simulations?
>>>
>> Yes, you can use mri_mcsim, something like
>>
>> mri_mcsim --o /path/to/mult-comp-cor/yoursubject/lh/cortex --base mc-z
>> --save-iter --surf yoursubject lh --nreps 100
>>
>> This may take several hours or days to run
>>
>> When running mri_glmfit-sim, add --cache-dir /path/to/mult-comp-cor
>>
>> doug
>>
>>> Thanks,
>>> GT
>>>
>>>
>>> _______________________________________________
>>> Freesurfer mailing list
>>> Freesurfer@nmr.mgh.harvard.edu
>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
--
Douglas N. Greve, Ph.D.
MGH-NMR Center
gr...@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is
addressed. If you believe this e-mail was sent to you in error and the e-mail
contains patient information, please contact the Partners Compliance HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you in error
but does not contain patient information, please contact the sender and properly
dispose of the e-mail.
