Hello Nik, Here is the solution to the mystery: The tutorial as it currently runs expects the user to start with the right hemisphere (stated near bottom of p. 2). The file that you need for stage 1 of the left hemisphere is created during Stage 3 for the right hemisphere (see below). This is not meant to be perverse! Rather, the intent is to capitalize on the effort put into revising the template landmarks for the right hemisphere. By mirror-flipping a good set of right hemisphere landmarks, you are likely to be much closer to the mark when you start on the left hemisphere.
The relevant operations in the stage 3 script are extracted below. If you do not wish to do (or repeat) the full set of right hemisphere steps, you can probably use this to kludge your way into generating the left hemisphere template border. Basically you need to take the right hemisphere medial wall borders and mirror-flip them to be approximately aligned for the left hem. There are other ways to do this in Caret on an ad hoc basis. I hope this helps. We should amend the tutorial to make this more clear. If you learn something that could help in improving the tutorial, please let us know. David Stage-3.FS-to-F99.sh if [ $HEM_FLAG = "R" ] ; then ### Prepare flipped right hem landmarks (medial wall segments for stage 1 and full landmarks for stage 2) as templates for left hem # Mirror-flip right hem morphed sphere MATRIX="-1.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 1.0" caret_command -surface-apply-transformation-matrix $CASE.R.SPHERE_smooth.$NUMBER_NODES.coord $CASE.R.CLOSED.$NUMBER_NODES.topo $CASE.R.SPHERE_smooth.X-Flip.$NUMBER_NODES.coord -matrix $MATRIX # unproject right hem medial wall borders from mirror-reflected initial sphere in same subject. caret_command -surface-border-unprojection $CASE.R.SPHERE_smooth.X-Flip.$NUMBER_NODES.coord $CASE.R.CLOSED.$NUMBER_NODES.topo $CASE.R.MedialWallSegments.borderproj $CASE.Flipped-R_For-LEFT.TEMPLATE.MedialWallSegments.SPHERE.border # unproject right hem medial wall borders from mirror-reflected initial sphere in same subject. caret_command -surface-border-unprojection $CASE.R.SPHERE_smooth.X-Flip.$NUMBER_NODES.coord $CASE.R.CLOSED.$NUMBER_NODES.topo $CASE.R.LANDMARKS.$NUMBER_NODES.borderproj $CASE.Flipped-R_For-LEFT.TEMPLATE.LANDMARKS.SPHERE.border fi On May 22, 2012, at 1:08 AM, Nikhil Sharma wrote: > Has anyone found the elusive > $CASE.Flipped-R_For-LEFT.TEMPLATE.MedialWallSegments.SPHERE.border file? > > Best Wishes > > Nik > > > On May 9, 2012, at 7:46 PM, Donna Dierker wrote: > >> I'm putting this back on the list, because I don't know the answer to either >> question. >> >> I searched for both MedialWallSegments and flipped (the latter with species >> set to macaque), and I didn't find it either way. Maybe David or someone >> else knows? >> >> The second issue is perplexing. Look at the script output carefully, and >> pinpoint the line where it is complaining. Before that line, try adding >> this line: >> >> caret_command -file-convert -format-convert BINARY *coord >> >> I had an issue a while back where certain Linux versions had a compiler >> issue that was causing coord files to be read incorrectly. Switching >> versions helped. Which version are you runing, and what is your OS? >> >> >> On May 8, 2012, at 12:02 PM, Leon wrote: >> >>> Hi, Donna >>> >>> Thank you for your tip on searching files on the website. It is very >>> helpful. >>> >>> I have a few related questions regarding the tutorial and would appreciate >>> your help on it: >>> >>> 1) The file "Macaque.FS.R.TEMPLATE.MedialWallSegments.SPHERE.border" is >>> found in the folder, but when processing left hemisphere according to the >>> Stage-1.FS-to-F99 bash, it requires a file called " >>> $CASE.Flipped-R_For-LEFT.TEMPLATE.MedialWallSegments.SPHERE.border", which >>> I do not find on SumsDB. Do you have any idea where I should look for it or >>> generate it? >>> >>> 2) I noticed that in the script, several coordinate files (WM, PIAL, >>> MIDTICKNESS) are sharing the same topology files. During the running, I >>> constantly get the error " MyMonkey.R.Midthickness.62227.coord: Topology >>> File MyMonkey.R.CLOSED.62227.topo is not for use with coordinate file >>> MyMonkey.R.Midthickness.62227.coord. Topo file has tiles with node numbers >>> exceeding >>> the number of coordinates in the coordinate file. >>> >>> I have checked the original rh.white, rh.pial and rh.inflated and they look >>> fine in FreeSurfer. Do you know how I should solve this issue of mismatched >>> topology and coordinate files? >>> >>> Thank you! >>> Leon >>> >>> From: Donna Dierker <[email protected]> >>> To: Leon <[email protected]>; "Caret, SureFit, and SuMS software users" >>> <[email protected]> >>> Sent: Tuesday, May 8, 2012 12:36 PM >>> Subject: Re: [caret-users] A question about FS-to-F99 tutorial >>> >>> I went to http://sumsdb.wustl.edu and selected Browse/Search: Basic Archive >>> Search. >>> >>> I entered TEMPLATE.MedialWallSegments in the filename box and got this >>> search result: >>> >>> Macaque.FS.R.TEMPLATE.MedialWallSegments.SPHERE.border >>> >>> Selecting "Show Parent" led me here: >>> >>> FS-to-F99_Aug10.zip >>> http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=8284563 >>> >>> I used a similar strategy to find Chip.LPI.nii: >>> >>> FS_MACAQUE.zip >>> http://sumsdb.wustl.edu/sums/archivelist.do?archive_id=8284039 >>> >>> Usually tutorials specify the archive_id of the datasets you need at the >>> beginning. >>> >>> >>> On May 7, 2012, at 8:48 PM, Leon wrote: >>> >>>> Hi, Caret experts >>>> As I am going through the FS-to-F99 tutorial, I notice that the downloaded >>>> package does not contain the following files that are needed for running >>>> the State-1.FS-to-F99.sh bash: >>>> >>>> 1) $CASE.Flipped-R_For-LEFT.TEMPLATE.MedialWallSegments.SPHERE.border >>>> >>>> Also, if I want to run the tutorial files, I could not find the volume >>>> file Chip.LPI.nii neither. >>>> >>>> Could someone help on locating these files? >>>> >>>> Thank you very much in advance! >>>> >>>> Leon >>>> _______________________________________________ >>>> caret-users mailing list >>>> [email protected] >>>> http://brainvis.wustl.edu/mailman/listinfo/caret-users >>> >>> >>> >> >> >> _______________________________________________ >> caret-users mailing list >> [email protected] >> http://brainvis.wustl.edu/mailman/listinfo/caret-users > > > _______________________________________________ > caret-users mailing list > [email protected] > http://brainvis.wustl.edu/mailman/listinfo/caret-users
_______________________________________________ caret-users mailing list [email protected] http://brainvis.wustl.edu/mailman/listinfo/caret-users
