Unable to run Rosetta LoopModeling application

[Anonymous]

I am a new Rosetta user trying to run the LoopModel application demo.
but I get the following print out with the command line:’loopmodel.linuxgccrelease @flags -database’

core.init: command: loopmodel.linuxgccrelease @flags -database
core.init: ‘RNG device’ seed mode, using ‘/dev/urandom’, seed=-683210121 seed_offset=0 real_seed=-683210121
core.init.random: RandomGenerator:init: Normal mode, seed=-683210121 RG_type=mt19937
core.init: ROSETTA3_DB not defined
protocols::loopbuild: ==== Loop protocol: =================================================
protocols::loopbuild: remodel no
protocols::loopbuild: intermedrelax no
protocols::loopbuild: refine no
protocols::loopbuild: relax no
Warning: Unable to locate database file chemical/residue_type_sets/fa_standard

ERROR: Unable to open file: ./chemical/residue_type_sets/fa_standard/residue_types.txt

ERROR:: Exit from: src/core/chemical/ResidueTypeSet.cc line: 88
loopmodel.linuxgccrelease: src/utility/exit.cc:119: void utility::exit(const std::string&, int, const std::string&, int): Assertion `false’ failed.
Aborted

My digging around tells me I may need to specify a database path, however I
am unsure of what the path does for the program. 1) What must I do to get the
demo to run? 2)Why is the database path need?
I would really appreciate anyone’s help!
-Brandon

[Anonymous]

1) Well, yes, you need to give it the database path. When you got Rosetta, you either got it as a tarball that included a rosetta_database subfolder, or you had a link to the academic downloads site which has just the database as its own tarball. Find where you put that and pass the path to the database flag. If you put it at /home/myself/rosetta_database, then “-database /home/myself/rosetta_database” will do it. Make sure you have write access to wherever you put it; for some operations it generates binary versions of text files as a speed optimization; this requires you to be able to write the new version of the file.

2) You need a database (path) because we avoided hard-coding energy function and chemical data into Rosetta. Rosetta’s code doesn’t know what alanine is, but it knows to read the database automatically, where it will find alanine when it needs it. Rosetta’s code doesn’t have the entire Dunbrack rotamer library coded in, it just reads those values from a database file if you need to to repacking.

[Anonymous]

This solves my issue, thank you!
-Brandon

[Author]

Posts