Hi everyone,
I am a new user of freesurfer. As a first step, I wrote a script (see below between ---) to run recon-all on the anatomical data. Since I found it very slow, and our server has multi-cores, I run this script manually several times, thus it appeared to be parallel.Then I have several questions:
1 From the screen output, I guess if a certain subject folder exists, freesurfer will refuse to work on this subject. Then I have assumption that if I run the same script multiple times would not overwrite results from previous session. Can I hold this believe? 2 After certain time, I found new folders with name such as [fsaverage], which I assumed as kind of average of subjects. Then I worry this was generated from one session (either the first one or the last one, depending on overwriting policy of this folder). For example, if session one worked on subject 1 4 7 ... and session two worked on subject 2 5 8... and session three worked on subject 3 6 9 ... ; is this [fsaverage] an average of (subject 1, 4, 7) or (subject 3, 6, 9)? Or I am so lucky that it is the average all processed subjects? 3 Is there a proper way to work in parallel than my stone-age way? 4 I searched the wiki and found some slides and pages which may help but not answering the above question. Would you please suggest an example "full" script of analysis? Thank you very much! ======= My environment: 64-bit Linux with OpenSUSE. 8-CPUs/32G mem Script: ------------- #!/bin/csh echo "beginning ..."
#environment setup source ~/1tests/FreeSurferEnv.csh
#Set environment and get the filelist set raw_dir = '~/2rawdata/' set fileNameList = `ls {$raw_dir}`
#Try all the subjects in a loop foreach file ($fileNameList) echo "Now processing file ($file): ..." echo "" set inName = ($raw_dir$file) recon-all -i $inName -s $file -autorecon-all end
echo "done!" ------------- =======