Re: [Freesurfer] Cluster-Defining Threshold of p=0.025
Andy, please remember to post to the FS list, thanks. As for your
question, this is a bug in mri_glmfit-sim that hard codes the allowable
thresholds. I've attached a version that should work with 1.6.
On 09/14/2017 10:06 AM, Jahn, Andrew wrote:
Thanks again; I was able to create new simulation tables. However,
when I run mri_glmfit-sim, I get the following error:
foreach dir (results_Directory)
mri_glmfit-sim \
--glmdir {$dir} \
--mczsim 1.6 abs \
--cwp 0.05 \
--2spaces
end
ERROR: thresh = 1.6, must be 1.3, 2.0, 2.3, 3.0, 3.3, 4.0
I get this error even though I have the tables located in
$FREESURFER_HOME/average/mult-comp-cor/fsaverage/lh/cortex/fwhm10/abs.
Is there something I'm missing?
Best,
-Andy
--
Douglas N. Greve, Ph.D.
MGH-NMR Center
gr...@nmr.mgh.harvard.edu
Phone Number: 617-724-2358
Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2
www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
#!/bin/tcsh -f
set VERSION = '$Id: mri_glmfit-sim,v 1.66 2016/11/16 20:52:22 greve Exp $';
setenv LANG en_US.UTF-8
set inputargs = ($argv);
set glmdir = ()
set nulltype = ();
set nsim = ()
set thresh = ()
set csdbase = ()
set simsign = ()
set cwpvalthresh = .05
set Overwrite = 0;
set DoSim = 1;
set LF = ();
set AllDoneFile = ();
set Seed = ();
set fwhmOverride = ();
set fwhmAdd = 0;
set UseUniformPDF = 0;
set UniformPDFMin = ();
set UniformPDFMax = ();
set DiagCluster = 0;
set PBNodeType = ();
set UseCache = 0;
set PermForce = 0;
set PermResid = 0; # use residual when running permutation, ter Braak method
set UseGRF = 0;
set volsubject = fsaverage;
set subjectOverride = ();
set Bonferroni = 0;
set ReportCentroid = 0;
set annot = aparc
set IsPVR = 0;
set DoClusterMean = 1
set OutputAnnot = 1
set DoGRFOCNAnat = 0;
set y = ()
set nJobs = 1;
set DoBackground = 0;
set DoPBSubmit = 0;
set DoPoll = 0;
set tSleepSec = 10;
set CacheDir = $FREESURFER_HOME/average/mult-comp-cor
set CacheLabel = cortex;
set tmpdir = ();
set CleanUp = 1;
set PrintHelp = 0;
if($#argv == 0) goto usage_exit;
set n = `echo $argv | grep -e --help | wc -l`
if($n != 0) then
set PrintHelp = 1;
goto usage_exit;
endif
set n = `echo $argv | grep -e --version | wc -l`
if($n != 0) then
echo $VERSION
exit 0;
endif
source $FREESURFER_HOME/sources.csh
goto parse_args;
parse_args_return:
goto check_params;
check_params_return:
if(! -e $glmdir) then
echo "ERROR: cannot find $glmdir"
exit 1;
endif
set glmfitlog = $glmdir/mri_glmfit.log
if(! -e $glmfitlog) then
echo "ERROR: cannot find $glmfitlog"
exit 1;
endif
if($nulltype != perm) then
set fwhmfile = $glmdir/fwhm.dat
if(! -e $fwhmfile) then
echo "ERROR: cannot find $fwhmfile"
exit 1;
endif
# This may be overridden
set fwhm = `cat $fwhmfile`;
else
set fwhm = 0;
endif
set glmfitcwd = `cat $glmfitlog | awk '{if($1 == "cwd") print $2}'`
if(0 && ! -e $glmfitcwd) then
echo "ERROR: cannot find $glmfitcwd"
exit 1;
endif
set glmfitcwd = `pwd`;
set anattype = volume;
set subject = ();
set hemi = ();
set surf = "white";
set wls = ();
set glmfitcwd = `cat $glmfitlog | awk '{if($1 == "cwd") print $2}'`
# Go through the original mri_glmfit command-line
set glmfitcmd0 = `cat $glmfitlog | awk '{if($1 == "cmdline") print $0}'`
set glmfitcmd = ($glmfitcmd0);
echo $glmfitcmd
set gd2mtx = dods
set UseTable = 0;
set label = ();
while($#glmfitcmd)
set flag = $glmfitcmd[1]; shift glmfitcmd;
#echo $flag
switch($flag)
case "--surf"
case "--surface"
set subject = $glmfitcmd[1]; shift glmfitcmd;
set hemi= $glmfitcmd[1]; shift glmfitcmd;
set anattype = surface;
if($#glmfitcmd != 0) then
set c = `echo $glmfitcmd[1] | cut -c 1-2`;
if("$c" != "--") then
set surf = $glmfitcmd[1]; shift glmfitcmd;
endif
endif
breaksw
case "--fsgd"
shift glmfitcmd;
if($#glmfitcmd > 0) then
if($glmfitcmd[1] == doss) set gd2mtx = doss
endif
breaksw
case "--label"
set label = $glmfitcmd[1]; shift glmfitcmd;
breaksw
case "--perm-force"
# Probably never gets here
set PermForce = 1;
breaksw
case "--table"
case "--y"
if($#y == 0) set y = $glmfitcmd[1];
shift glmfitcmd;
if($DoClusterMean) then
if(! -e $y) then
set y = $glmfitcwd/$y
if(! -e $y) then
set y = $glmdir/`basename $y`
if(! -e $y) then
echo "ERROR: cannot find $y"
echo "If you do not have this file, run with --no-y as the first
option"
exit 1;
endif
endif
endif
endif
if("$flag" == "--table") set UseTable = 1;
breaksw
case "--wls"
set wls = $glmfitcwd/$glmfitcmd[1]; shift glmfitcmd;
if(! -e $wls) then
set wls = $glmdir/`basename $wls`
if(! -e $wls) then
echo "ERROR: cannot find $wls"
exit 1;
endif
Re: [Freesurfer] Cluster-Defining Threshold of p=0.025
There is actually an option called --thresh where you can give it a threshold to use. Somehow it did not make it onto the help docs On 09/13/2017 11:39 AM, Jahn, Andrew wrote: > Hi Doug, > > Thank you for the custom permutation script; it is very helpful. > > However, it is still not clear how to use mri_mcsim to create tables > with different thresholds. I understand how to use it to restrict the > search space and how to use it on different subjects, but I don't see > any option for specifying different thresholds. Is there a specific > option in mri_mcsim for doing this? > > > Thank you, > > -Andy > > > *From:* freesurfer-boun...@nmr.mgh.harvard.edu > <freesurfer-boun...@nmr.mgh.harvard.edu> on behalf of Douglas N Greve > <gr...@nmr.mgh.harvard.edu> > *Sent:* Thursday, September 7, 2017 5:55:30 PM > *To:* freesurfer@nmr.mgh.harvard.edu > *Subject:* Re: [Freesurfer] Cluster-Defining Threshold of p=0.025 > If you want to use that threshold, you will have to create tables for > that threshold using mri_mcsim. See > > https://surfer.nmr.mgh.harvard.edu/fswiki/BuildYourOwnMonteCarlo > > > Alternatively, I have a version of mri_glmfit-sim that will run > permutation with non-orthogonal designs. > > https://gate.nmr.mgh.harvard.edu/safelinks/greve/mri_glmfit-sim > > > This is the direction I'm pushing people in. To run it, use the normal > permutation options and add --perm-resid > > This uses the ter Braak method. Run mri_glmfit-sim --help to get more info > > > > > > On 09/07/2017 11:05 AM, Jahn, Andrew wrote: > > Hello, > > > > Is it possible to use cluster-defining thresholds such as p = 0.025 > > with mri_glmfit-sim? (This would roughly correspond to a negative log > > value of 1.6.) My design doesn't allow the use of permutations for > > simulating cluster sizes, so I have to use the mczsim option; however, > > from what I can tell it doesn't allow using any non-cached options. > > > > > > Thanks, > > > > -Andy > > > > > > > > ___ > > 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 > <http://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 > https://urldefense.proofpoint.com/v2/url?u=http-3A__www.partners.org_complianceline=DwIBAg=cjytLXgP8ixuoHflwc-poQ=h-S9eBUOU6jgx_fyX6Af8TeTt7dGNPjG7qjp2Qe-zMs=7dwQJDogv4w-2-MgP5exf4vmnDjsQooz0tQsUyZRkFk=8S0l8EjW9GU0a0bo8j1jLq0yijKG3KYh7lr-RArddLQ= > > . 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. > > > > ___ > 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.
Re: [Freesurfer] Cluster-Defining Threshold of p=0.025
Hi Doug, Thank you for the custom permutation script; it is very helpful. However, it is still not clear how to use mri_mcsim to create tables with different thresholds. I understand how to use it to restrict the search space and how to use it on different subjects, but I don't see any option for specifying different thresholds. Is there a specific option in mri_mcsim for doing this? Thank you, -Andy From: freesurfer-boun...@nmr.mgh.harvard.edu <freesurfer-boun...@nmr.mgh.harvard.edu> on behalf of Douglas N Greve <gr...@nmr.mgh.harvard.edu> Sent: Thursday, September 7, 2017 5:55:30 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Cluster-Defining Threshold of p=0.025 If you want to use that threshold, you will have to create tables for that threshold using mri_mcsim. See https://surfer.nmr.mgh.harvard.edu/fswiki/BuildYourOwnMonteCarlo Alternatively, I have a version of mri_glmfit-sim that will run permutation with non-orthogonal designs. https://gate.nmr.mgh.harvard.edu/safelinks/greve/mri_glmfit-sim This is the direction I'm pushing people in. To run it, use the normal permutation options and add --perm-resid This uses the ter Braak method. Run mri_glmfit-sim --help to get more info On 09/07/2017 11:05 AM, Jahn, Andrew wrote: > Hello, > > Is it possible to use cluster-defining thresholds such as p = 0.025 > with mri_glmfit-sim? (This would roughly correspond to a negative log > value of 1.6.) My design doesn't allow the use of permutations for > simulating cluster sizes, so I have to use the mczsim option; however, > from what I can tell it doesn't allow using any non-cached options. > > > Thanks, > > -Andy > > > > ___ > 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<http://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 https://urldefense.proofpoint.com/v2/url?u=http-3A__www.partners.org_complianceline=DwIBAg=cjytLXgP8ixuoHflwc-poQ=h-S9eBUOU6jgx_fyX6Af8TeTt7dGNPjG7qjp2Qe-zMs=7dwQJDogv4w-2-MgP5exf4vmnDjsQooz0tQsUyZRkFk=8S0l8EjW9GU0a0bo8j1jLq0yijKG3KYh7lr-RArddLQ= . 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. ___ 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.
Re: [Freesurfer] Cluster-Defining Threshold of p=0.025
If you want to use that threshold, you will have to create tables for that threshold using mri_mcsim. See https://surfer.nmr.mgh.harvard.edu/fswiki/BuildYourOwnMonteCarlo Alternatively, I have a version of mri_glmfit-sim that will run permutation with non-orthogonal designs. https://gate.nmr.mgh.harvard.edu/safelinks/greve/mri_glmfit-sim This is the direction I'm pushing people in. To run it, use the normal permutation options and add --perm-resid This uses the ter Braak method. Run mri_glmfit-sim --help to get more info On 09/07/2017 11:05 AM, Jahn, Andrew wrote: > Hello, > > Is it possible to use cluster-defining thresholds such as p = 0.025 > with mri_glmfit-sim? (This would roughly correspond to a negative log > value of 1.6.) My design doesn't allow the use of permutations for > simulating cluster sizes, so I have to use the mczsim option; however, > from what I can tell it doesn't allow using any non-cached options. > > > Thanks, > > -Andy > > > > ___ > 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.
