You are here

Segmentation Fault during prepacking

5 posts / 0 new
Last post
Segmentation Fault during prepacking
#1

Dear all,

   When I do prepack of a antibody-antigen complex with the following command,

mpirun -np 6 docking_prepack_protocol.mpi.linuxgccrelease\
  -in:file:s nb0-ag-start.pdb\
  -ex1\
  -ex2\
  -partners H_A\
  -ensemble1 nb_ensemble.list\
  -ensemble2 ag_ensemble.list\
  -docking:dock_rtmin\
  -out:path:all prepacking
 

I always get a "segmentation Fault" like the following: 

[END_BACKTRACE]

[FILE]: SIGSEGV
[LINE]: 11
[START_MESSAGE]
Segmentation Fault

[END_MESSAGE]
[END_CRASH_REPORT]

   I cannot find anything wrong with my input files or commands and am totally stuck here. I wonder whether it is related to my computer or the inputs. Could you help me check this and find out the real problem? Please download the inputs files from the following link:

  https://drive.google.com/file/d/1SPq_9d2SyzIJl5kYW8LM-B1JOvrwQ6dx/view?usp=sharing

Thank you very much!

 

Category: 
Post Situation: 
Fri, 2020-06-12 11:24
Sunyp_IM

Hello,

Presumably you got past this stage because I see a question about ranking docked models. I requested access to the drive to see your inputs.

It's difficult to say exactly what went wrong with seg faults, but if I were to hazard a guess, it would be that the members of your ensemble did not have the same number of residues as the protein in your input structure. Or maybe that some residues had missing backbone atoms, so Rosetta discarded those residues while reading in the ensemble. When Rosetta was trying to thread in the backbone from the ensemble, it seg faulted.

Best,

Shourya

 

Mon, 2020-06-22 13:28
ssrb

Hello Shourya,

You are correct. It is an input problem. Thank you!

 

Wed, 2020-07-01 06:18
Sunyp_IM

Hi all,

Apologies if this should be a new thread, but I'm having a similar issue performing docking prepack for a MPDock run. Is there any way to systematically work through where the seg fault could be coming from? I've looked over the crash report and logs in general, and nothing seems immediately obvious. Is there a way to check the number of backbone atoms? I've added OXTs using pdbfixer and thought that would fix the problem, but the command still results in a seg fault. Are hydrogens also included in the backbone calculations? That I'm missing some of those is the only other possibility I can think of.

Thanks,

Jeff

Mon, 2020-07-06 13:52
jzahratk

Hi, jeff,

As for my case, the seg fault is caused by the indifference between the number of atoms of the antigen in the start structure and that in the ensemble. So my suggestion is to carefully check your input. There are numerous applications that can help you count the number of backbone or any tpye of atoms, you can google it.

cheers.

Thu, 2020-07-09 18:40
Sunyp_IM