[Neurodebian-users] Ubuntu 10.10, csh or tcsh and FSL 4.1
Deryk S. Beal, Ph.D.
dsbeal at bu.edu
Thu Jun 9 04:15:24 UTC 2011
Hi Michael
Thank you for the quick reply and the advice on an adequate shell for both
FS and FSL.
The commands I had difficulty with stem from the use of FS5.1.0 and the
trac-all - prep command that calls a number of FSL routines. The
configuration file and the routines that it calls seem to be written for
csh/tcsh.
I have encountered errors, and still do even with fslmachtype.sh and the csh
apparently working correctly, that imply that my FSL distro is not sourced
properly. Here is an example of the output from trac-all -prep -c when it
faults before even processing one subject (thank you in advance for any
advice):
trac-all -prep -c
/media/5424CE1A24CDFECC/UofT/data/Tracula/tracula_config.txt
INFO: SUBJECTS_DIR is /usr/local/freesurfer/subjects
INFO: Diffusion root is /usr/local/freesurfer/subjects
Actual FREESURFER_HOME /usr/local/freesurfer
flirt: error while loading shared libraries: libnewimage.so: cannot open
shared object file: No such file or directory
trac-preproc -c /usr/local/freesurfer/subjects/C001/scripts/dmrirc.local
-log /usr/local/freesurfer/subjects/C001/scripts/trac-all.log -cmd
/usr/local/freesurfer/subjects/C001/scripts/trac-all.cmd
#-------------------------------------
/usr/local/freesurfer/bin/trac-preproc
#-------------------------------------
#@# Image corrections Wed Jun 8 16:42:47 EDT 2011
mri_convert /media/5424CE1A24CDFECC/UofT/data/Tracula/C001_merged.nii.gz
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig.nii.gz
mri_convert /media/5424CE1A24CDFECC/UofT/data/Tracula/C001_merged.nii.gz
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig.nii.gz
$Id: mri_convert.c,v 1.179.2.2 2011/05/16 20:53:47 greve Exp $
reading from /media/5424CE1A24CDFECC/UofT/data/Tracula/C001_merged.nii.gz...
TR=17000.00, TE=0.00, TI=0.00, flip angle=0.00
i_ras = (-1, 0, 0)
j_ras = (0, 1, 0)
k_ras = (0, 0, 1)
writing to /usr/local/freesurfer/subjects/C001/dmri/dwi_orig.nii.gz...
cp /media/5424CE1A24CDFECC/UofT/data/Tracula/bvecs.txt
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig.mghdti.bvecs
cp /media/5424CE1A24CDFECC/UofT/data/Tracula/bvals.txt
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig.mghdti.bvals
flip4fsl /usr/local/freesurfer/subjects/C001/dmri/dwi_orig.nii.gz
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig_flip.nii.gz
INFO: input image orientation is LAS
INFO: input image determinant is -10
fslswapdim /usr/local/freesurfer/subjects/C001/dmri/dwi_orig.nii.gz x y z
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig_flip.nii.gz
fslorient -forceradiological
/usr/local/freesurfer/subjects/C001/dmri/dwi_orig_flip.nii.gz
fslorient: error while loading shared libraries: libnewimage.so: cannot open
shared object file: No such file or directory
INFO: found /usr/local/freesurfer/subjects/C001/dmri/dwi_orig.mghdti.bvals,
converting to FSL format
INFO: found /usr/local/freesurfer/subjects/C001/dmri/dwi_orig.mghdti.bvecs,
converting to FSL format
mv -f /usr/local/freesurfer/subjects/C001/dmri/dwi_orig_flip.mghdti.bvecs
/usr/local/freesurfer/subjects/C001/dmri/bvecs
mv -f /usr/local/freesurfer/subjects/C001/dmri/dwi_orig_flip.mghdti.bvals
/usr/local/freesurfer/subjects/C001/dmri/bvals
eddy_correct /usr/local/freesurfer/subjects/C001/dmri/dwi_orig_flip.nii.gz
/usr/local/freesurfer/subjects/C001/dmri/dwi.nii.gz 0
Input does not exist or is not in a supported format
mv -f /usr/local/freesurfer/subjects/C001/dmri/bvecs
/usr/local/freesurfer/subjects/C001/dmri/bvecs.norot
xfmrot /usr/local/freesurfer/subjects/C001/dmri/dwi.ecclog
/usr/local/freesurfer/subjects/C001/dmri/bvecs.norot
/usr/local/freesurfer/subjects/C001/dmri/bvecs
grep: /usr/local/freesurfer/subjects/C001/dmri/dwi.ecclog: No such file or
directory
wc: /usr/local/freesurfer/subjects/C001/dmri/dwi.ecclog: No such file or
directory
ERROR: Transform file should be an eddy_correct log file or a .mat file
Linux dsb-P35-DS3L 2.6.35-28-generic #50-Ubuntu SMP Fri Mar 18 18:42:20 UTC
2011 x86_64 GNU/Linux
trac-preproc exited with ERRORS at Wed Jun 8 16:43:10 EDT 2011
Here are the contents of the FS subject folder after it faults. It is as
thought the dwi.ecclog was never written:
bvals bvecs.norot dwi_orig.mghdti.bvals dwi_orig.mghdti.bvecs
dwi_orig.nii.gz xfms
Here is my CONFIG FILE:
# In case you want to analyze only Huey and Louie
# Default: Run analysis on all subjects
#
set runlist = (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22)
# Input diffusion DICOMs (file names relative to dcmroot)
# If original DICOMs don't exist, these can be in other image format
# but then bvecfile, bvalfile, and nb0 must be specified (see below)
#
set dcmroot = /media/5424CE1A24CDFECC/UofT/data/Tracula
set dcmlist = (C001_merged.nii.gz \ C002_merged.nii.gz \ C003_merged.nii.gz
C004_merged.nii.gz \ C005_merged.nii.gz \ C006_merged.nii.gz \
C007_merged.nii.gz \ C008_merged.nii.gz \ C009_merged.nii.gz \
C010_merged.nii.gz \ C011_merged.nii.gz \ S001_merged.nii.gz \
S002_merged.nii.gz \ S003_merged.nii.gz \ S004_merged.nii.gz \
S005_merged.nii.gz \ S006_merged.nii.gz \ S007_merged.nii.gz \
S008_merged.nii.gz \ S009_merged.nii.gz \ S010_merged.nii.gz \
S011_merged.nii.gz)
# Diffusion gradient table
# Must be specified if inputs are not MGH DICOMs
# Three-column format, one row for each volume in the diffusion data set
# Default: Read from DICOM header
#
set bvecfile = /media/5424CE1A24CDFECC/UofT/data/Tracula/bvecs.txt
# Diffusion b-value table
# Must be specified if inputs are not MGH DICOMs
# Single-column format, one value for each volume in the diffusion data set
# Default: Read from DICOM header
#
set bvalfile = /media/5424CE1A24CDFECC/UofT/data/Tracula/bvals.txt
# Number of low-b images
# Must be specified if inputs are not DICOM
# Default: Read from DICOM header
#
set nb0 = 4
# Perform registration-based B0-inhomogeneity compensation?
# Default: 0 (no)
#
set dob0 = 0
# Input B0 field map magnitude DICOMs (file names relative to dcmroot)
# Only used if dob0 = 1
# Default: None
#
# set b0mlist = (huey/fmag/XXX-1.dcm dewey/fmag/XXX-1.dcm
louie/fmag/XXX-1.dcm)
# Input B0 field map phase DICOMs (file names relative to dcmroot)
# Only used if dob0 = 1
# Default: None
#
# set b0plist = (huey/fphas/XXX-1.dcm dewey/fphas/XXX-1.dcm
louie/fphas/XXX-1.dcm)
# Echo spacing for field mapping sequence (from sequence printout)
# Only used if dob0 = 1
# Default: None
#
# set echospacing = 0.7
# Perform registration-based eddy-current compensation?
# Default: 1 (yes)
#
set doeddy = 1
# Rotate diffusion gradient vectors to match eddy-current compensation?
# Only used if doeddy = 1
# Default: 1 (yes)
#
set dorotbvecs = 1
# Fractional intensity threshold for BET mask extraction from low-b images
# This mask is used only if usemaskanat = 0
# Default: 0.3
#
set thrbet = 0.3
# Perform diffusion-to-T1 registration by flirt?
# Default: 1 (yes)
#
set doregflt = 0
# Perform diffusion-to-T1 registration by bbregister?
# Default: 0 (no)
#
set doregbbr = 1
# MNI template (the only option for inter-subject registration in this
version)
# Default: $FSLDIR/data/standard/MNI152_T1_1mm_brain.nii.gz
#
set mnitemp = /usr/share/fsl/4.1/data/standard/MNI152_T1_1mm_brain.nii.gz
# Use brain mask extracted from T1 image instead of low-b diffusion image?
# Has no effect if there is no T1 data
# Default: 1 (yes)
#
set usemaskanat = 1
# Paths to reconstruct
# Default: All paths in the atlas
#
set pathlist = ( lh.cst_AS rh.cst_AS \
lh.unc_AS rh.unc_AS \
lh.ilf_AS rh.ilf_AS \
fmajor_PP fminor_PP \
lh.atr_PP rh.atr_PP \
lh.ccg_PP rh.ccg_PP \
lh.cab_PP rh.cab_PP \
lh.slfp_PP rh.slfp_PP \
lh.slft_PP rh.slft_PP )
# Number of path control points
# Default: 5
#
set ncpts = 5
# List of training subjects
# This text file lists the locations of training subject directories
# Default: $FREESURFER_HOME/trctrain/trainlist.txt
#
set trainfile = $FREESURFER_HOME/trctrain/trainlist.txt
# Use long (more descriptive) directory hierarchy for saving path
distributions?
# By default, paths distributions are saved directly under
$subjectname/dpath
# Default: 0 (no)
#
set dopathsubdirs = 0
# Number of MCMC burn-in iterations
# (Path samples drawn initially by MCMC algorithm and discarded)
# Default: 200
#
set nburnin = 200
# Number of MCMC iterations
# (Path samples drawn by MCMC algorithm and used to estimate path
distribution)
# Default: 5000
#
set nsample = 5000
# Frequency with which MCMC path samples are retained for path distribution
# Default: 5 (keep every 5th sample)
#
set nkeep = 5
Deryk S. Beal, Ph.D., CCC-SLP, S-LP(C), Reg. CASLPO
Speech-Language Pathologist
C.I.H.R. Post Doctoral Research Fellow
Department of Cognitive and Neural Systems Speech Laboratory
Boston University, 677 Beacon Street, Boston, MA 02215
dsbeal at bu.edu
http://blogs.bu.edu/dsbeal/about/
*************************************************************************
This email may contain confidential and/or privileged information for the
sole use of the intended recipient. Any review or distribution by others is
strictly prohibited. If you have received this email in error, please
contact the sender and delete all copies. Opinions, conclusions or other
information expressed or contained in this email are not given or endorsed
by the sender unless otherwise affirmed independently by the sender.
On Wed, Jun 8, 2011 at 10:22 PM, Michael Hanke <mih at debian.org> wrote:
> Hi,
>
> On Wed, Jun 08, 2011 at 04:42:03PM -0400, Deryk S. Beal, Ph.D. wrote:
> > Why is fslmachtype.sh not distributed with the Neurodebian version? One
> must
> > be able to run fsl in tcsh or csh for certain Freesurfer commands.
>
> I wasn't aware that this is the case. Would you mind sharing what
> commands are affected?
>
> In any case fslmachtype.sh is not shipped, because it is not necessary.
> You can safely remove this line from fsl.csh. The Debian package doesn't
> provide out-of-the-box csh support, because (not just) in Debian csh is
> considered harmful:
>
> http://lintian.debian.org/tags/csh-considered-harmful.html
>
> However, if there is a problem with Freesurfer we need to work it out --
> it is just that we haven't seen problems so far and happily use FSL with
> ksh/bash/dash/zsh/...
>
> Michael
>
> --
> Michael Hanke
> http://mih.voxindeserto.de
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/neurodebian-users/attachments/20110609/69c7aaaa/attachment-0002.html>
More information about the Neurodebian-users
mailing list