[phenixbb] mr_rosetta problems

Terwilliger, Thomas C terwilliger at lanl.gov
Tue Aug 7 14:53:01 PDT 2012


Hi Ian,

I'm sorry for the trouble!  Can you try this:

cd /test_prerefine/WORK_2/WORK_1/

cat cat run_rebuild.sh

This should look something like:

cd "/Users/terwill/unix/misc/junk/test_prerefine/WORK_2/WORK_1"
"/Users/terwill/unix/rosetta/rosetta3.4/rosetta_source/bin/mr_protocols.default.macosgccrelease" @"rebuild.flags" > "rebuild.log" 2>&1

Then paste from this something like:

/Users/terwill/unix/rosetta/rosetta3.4/rosetta_source/bin/mr_protocols.default.macosgccrelease" @"rebuild.flags"

This should fail, but quickly and with the error message right there.  Then try editing "rebuild.flags" to specify the standard set of fragments files in some place like:

/Users/terwill/unix/PHENIX/phenix-dev-894/phenix/phenix/rosetta/dummy_fragments_file_3.gz
/Users/terwill/unix/PHENIX/phenix-dev-894/phenix/phenix/rosetta/dummy_fragments_file_9.gz

and that should work.  Then finally you can unzip all these fragment files and use those versions.  Then you can paste pieces of your fragments files into the standard ones and see what causes it to break.

I hope that helps!
All the best,
Tom T

________________________________
From: phenixbb-bounces at phenix-online.org [phenixbb-bounces at phenix-online.org] on behalf of Ian Slaymaker [iamslaymaker at gmail.com]
Sent: Tuesday, August 07, 2012 11:32 AM
To: PHENIX user mailing list
Subject: Re: [phenixbb] mr_rosetta problems

I ran the suggested test, and checked my fragment files. The fragment files seem to be spic and span, but the test still errored out with this and the rebuild.log file looked the same.
My fragment files look like this :

position:            1 neighbors:          200

 1je5 A   137 M L  -94.840  117.510 -178.319    2.237   10.852    31.857 3     0.000 P  1 F  1
 1je5 A   138 E L  -71.213  -39.327  177.116    2.237   10.852    31.857 3     0.000 P  1 F  1
 1je5 A   139 D L -112.228   93.181 -175.069    2.237   10.852    31.857 3     0.000 P  1 F  1
 1je5 A   140 V L  -70.333  120.424  178.205    2.237   10.852    31.857 3     0.000 P  1 F  1

Starting mr_rosetta_rebuild
Date: Tue Aug  7 10:15:26 2012
Directory: /test_prerefine

HOST = hpc-cmb
HOSTTYPE = x86_64-linux
USER = slaymake
PID = 30481
Checking rosetta paths:
  rosetta binary: /home/cmb-04/rr/slaymake/rosetta/rosetta3.4/rosetta_source/bin/mr_protocols.default.                                                   linuxgccrelease
  database_dir: /home/cmb-04/rr/slaymake/rosetta/rosetta3.4/rosetta_database
  script_dir: /home/cmb-04/rr/slaymake/rosetta/rosetta3.4/rosetta_source/src/apps/public/electron_dens                                                   ity
pilot_script_dir: /home/cmb-04/rr/slaymake/rosetta/rosetta3.4/rosetta_source/src/apps/pilot/frank/

Sequence rewritten to /test_prerefine/WORK_2/EDITED_seq.dat                                                    :
 > sequence information
AQLMDMRD

Changing to work directory: /test_prerefine/WORK_2


Rebuilding the model/test_prerefine/coords1.pdb
with rosetta
Running mr_rosetta_rebuild without map information

Generating 1 rebuilt models
================================================================================
Reading model from/test_prerefine/coords1.pdb
================================================================================
ALIGNMENT LINES:
## TARGET coord
# hhsearch
scores_from_program: 0 1.00
0 AQLMDMRD
0 -QLMDMRD
--

Alignment file matches PDB file
Start position of target sequence in sequence file:  0

Aligned model for comparison: /test_prerefine/WORK_2/WORK_1/aligned_model.pdb
Log for rebuilding: /test_prerefine/WORK_2/WORK_1/rebuild.log

******************* ERROR ENDING ***************
Failed to run REBUILD...see log file in /test_prerefine/WORK_2/WORK_1/rebuild.log

******************* ERROR ENDING ***************
hpc-cmb(231):


On Tue, Aug 7, 2012 at 10:05 AM, Andreas Förster <docandreas at gmail.com<mailto:docandreas at gmail.com>> wrote:
Hi Ian,

I had a similar issue once.  Check that your fragment files start with a line reading

 position:            1 neighbors:          200

or something like this.  If there's html code (download artifact), remove it.  Same at the end of the files.  Last line should be blank and before that, only fragments.



Andreas




On 07/08/2012 5:52, Terwilliger, Thomas C wrote:
Hi Ian,

I think this part of the error message is the best hint:

finished reading top 0 0mer fragments from
file/dataFiles/aat000_03_05.200_v1_3

I am guessing there is something odd about the fragment file all right.
You could try supplying this file in one of the tests and see if it
causes the test to fail:

phenix_regression.wizards.test_command_line_rosetta_quick test_prerefine

then:...

cd test_prerefine
phenix.mr_rosetta seq_file=seq.dat search_models=coords1.pdb
start=place_model stop=place_model data=fobs.mtz is_sub_process=True
already_placed=true refine_after_mr=False run_prerefine=True
number_of_prerefine_models=1 rosetta_fixed_seed=1571
fragment_files=/Users/terwill/unix/PHENIX/phenix-dev-894/phenix/phenix/rosetta/dummy_fragments_file_3.gz
fragment_files=/Users/terwill/unix/PHENIX/phenix-dev-894/phenix/phenix/rosetta/dummy_fragments_file_9.gz

Then, try above with your fragment files...

That might point out the problem!
All the best,
Tom T






*From:* phenixbb-bounces at phenix-online.org<mailto:phenixbb-bounces at phenix-online.org>

[phenixbb-bounces at phenix-online.org<mailto:phenixbb-bounces at phenix-online.org>] on behalf of Ian Slaymaker
[slaymake at usc.edu<mailto:slaymake at usc.edu>]
*Sent:* Tuesday, August 07, 2012 10:21 AM
*To:* phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>
*Subject:* [phenixbb] mr_rosetta problems


I am trying to get mr_rosetta working on our cluster and am running into
some problems I am not sure how to solve. The
phenix_regression.wizards.test_command_line_rosetta_quick_tests tests
went smoothly were completed.

However, my first test run with all the defaults (not submitted to
multiple processors) using my reflection file, HHR analysis file,
sequence file, and the 2 fragment files failed with the below errors. I
checked all the files, and it seems this is a problem with the fragment
files but I am not sure exactly what they are supposed to look like.

I thought it could be because our cluster doesnt have access to any
outside networks...is this a problem with rosetta not being able to
download pdb files? if so, is there a way to supply them internally?
The aligned model looks somewhat what I expected it to (same beta sheet
arrangement) but it seems to be incomplete.
Any help would be very much appreciated!



Splitting work into 100 jobs and running with 1 processors using sh
background=None in /MR_ROSETTA_1/GROUP_OF_ROSETTA_REBUILD_1/RUN_1/REBUILD_IN_SETS_1

Starting job 1...Log will be: /MR_ROSETTA_1/GROUP_OF_ROSETTA_REBUILD_1/RUN_1/REBUILD_IN_SETS_1/RUN_FILE_1.log

******************* ERROR ENDING ***************
mr_rosetta failed...

********************************************************************************
Sorry, a subprocess has failed...

END OF LOG FILE /MR_ROSETTA_1/GROUP_OF_ROSETTA_REBUILD_1/RUN_1/REBUILD_IN_SETS_1/RUN_FILE_1.log :

to match the actual PDB sequence:
(protein sequence)
New alignment line 4 for PDB:
(protein sequence)


Start position of target sequence in sequence file:  0

Aligned model for comparison: /MR_ROSETTA_1/GROUP_OF_ROSETTA_REBUILD_1/RUN_1/REBUILD_IN_SETS_1/RUN_1/WORK_1/aligned_model.pdb
Log for rebuilding: /MR_ROSETTA_1/GROUP_OF_ROSETTA_REBUILD_1/RUN_1/REBUILD_IN_SETS_1/RUN_1/WORK_1/rebuild.log

**************ERROR ENDING**********************
Failed to run REBUILD...see log file in /MR_ROSETTA_1/GROUP_OF_ROSETTA_REBUILD_1/RUN_1/REBUILD_IN_SETS_1/RUN_1/WORK_1/rebuild.log


The rebuild log had repeats of several errors that I am not sure how to address:

core.conformation.Conformation: [ WARNING ] missing heavyatom:  CD2 on residue PHE 3 (repeated several times)

core.fragments.ConstantLengthFragSet: line too short: Skipping line ' (repeated several times)


core.fragments.ConstantLengthFragSet: finished reading top 0 0mer
fragments from file/dataFiles/aat000_03_05.200_v1_3
protocols.loops.loops_main: Frag libraries debug none 1






_______________________________________________
phenixbb mailing list
phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>
http://phenix-online.org/mailman/listinfo/phenixbb


--
        Andreas Förster, Research Associate
        Paul Freemont & Xiaodong Zhang Labs
Department of Biochemistry, Imperial College London
            http://www.msf.bio.ic.ac.uk

_______________________________________________
phenixbb mailing list
phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>
http://phenix-online.org/mailman/listinfo/phenixbb

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20120807/d0e4ebfb/attachment.htm>


More information about the phenixbb mailing list