You are here

St13runtime_error in Rosetta ligand docking

3 posts / 0 new
Last post
St13runtime_error in Rosetta ligand docking


Hi all

I'm trying to dock the ligand into receptor with rosetta ligand docking with following command line

mpirun -np 20 rosetta_scripts.mpi.linuxgccrelease @options

where options file looks like this

-in:file:s crystal_complex.pdb
-in:file:extra_res_fa LIG.params
-out:path:all dock_res
-nstruct 2000
-packing:no_optH false
-packing:flip_HNQ true
-packing:ignore_ligand_chi true
-parser:protocol /dmc-hit-tools/toolbox/neoscan_lib/dock.xml
-mistakes:restore_pre_talaris_2013_behavior true
-ignore_zero_occupancy false
-qsar:max_grid_cache_size 1
-qsar:grid_dir tmp

It works well, but sometimes unknown error occurs and when I look at the ROSETTA_CRASH.log file in result folder,

I could see error contents like below

[FILE]: St13runtime_error
[LINE]: 0
value type is 6 not 2


I could not figure out what cause this error and the thing that I really can not understand is when I run this command again with same input in same condition

it runs well without any error!

Any kind of comment or advice would be welcomed about this error




Post Situation: 
Thu, 2023-02-09 18:18

That error seems to be coming from the JSON file reader.

If you're not using a JSON file with the StartFrom mover, the most likely cause is the cached grids (which are stored in JSON format). It may be that the grid caching mechanism isn't quite compatible with MPI runs (e.g. the different processes may be overwriting the grid cache directory.)

I would recommend trying it without the `-qsar:grid_dir` option, and see if that solves the issue. (It will complain about the option saving time, but it doesn't save *that* much time these days, not so much that it's worth dealing with crashes related to it.)

Fri, 2023-02-10 08:23

Thanks a lot

when I run  without '-qsar:grid_dir' option it runs without any error

Sun, 2023-02-19 18:39