External Email - Use Caution

Thanks so much for the quick response!

Mimicking the behavior of a local run was my goal. Right now, I run all three "portions" of trac-all in one script (using the python code shown below):

 # call each step in tracula    trac_all_prep_call = ["trac-all",
"-prep", "-c",
f"{config_file}",
"-log", f"{prep_log_file}"]
    trac_all_bed_call = ["trac-all",
"-bedp", "-c",
f"{config_file}",
"-log", f"{bed_log_file}"]
    trac_all_path_call = ["trac-all",
"-path", "-c" ,
f"{config_file}",
"-log", f"{path_log_file}"]
    for trac_call 
in [trac_all_prep_call, trac_all_bed_call, trac_all_path_call]:
        o= subprocess.run(trac_call)

So my goal when suggesting that change was to have the trac-all -bedp command work when the onpbs variable is False. However, it sounds like you are saying that in order to run bedpost not on a cluster I need to do so "outside" of trac-all. So, should I replace the trac_all_bed_call above with something like:

bedpost -n $nstick -model 1 /path/to/subjects/dmri/directory

Would that work? Sorry if I am misunderstanding. Thanks so much for your help!

Thanks,

Amy

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Yendiki, Anastasia <AYENDIKI@mgh.harvard.edu>
Sent: Wednesday, March 23, 2022 9:10 AM
To: Freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] bedpostx error inside trac-all

* External Email - Caution *

Hi Amy - This will have the same effect as running it locally. All the the *_mgh version does is submit each slice as a separate job on a certain type of cluster. If you run trac-all with --jobs <filename> , it will save a text file with the commands that would've been submitted as jobs, instead of actually trying to submit them. Then you can submit them to a cluster yourself, using the syntax that's specific to your cluster.

Anastasia.

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Mirro, Amy <amirro@wustl.edu>
Sent: Wednesday, March 23, 2022 10:04 AM
To: Freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] bedpostx error inside trac-all

External Email - Use Caution

Hello,

I am trying to run trac-all -bedp and I am running into the following error:

/opt/freesurfer/bin/bedpostx_mgh: 128: /opt/freesurfer/bin/bedpostx_mgh: Syntax error: "(" unexpected

Others have encountered this error previously, and you had advised running bedpostx locally (as oppsoed to bedpostx_mgh). I was wondering, however, if there is a way to still run bedpostx inside trac-all.

I was looking at the source code for trac-all in your github (branch 7.2), and I noticed that line 611 calls bedpostx_mgh despite being within the "if (! $onpbs)" conditional. Would changing line 611 from:

set cmd = (bedpostx_mgh -n $nstick -model 1 $dtdir/dmri)

set cmd = (bedpostx -n $nstick -model 1 $dtdir/dmri)

fix the problem?

If it's helpful:

I am running on centos7 and my build stamp is: freesurfer-linux-centos7_x86_64-7.2.0-20210720-aa8f76b

Thanks,

Amy

The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.