Member Site › Forums › Rosetta 3 › Rosetta 3 – General › is it possible to generate models without hydrogen?
- This topic has 9 replies, 4 voices, and was last updated 12 years, 8 months ago by Anonymous.
-
AuthorPosts
-
-
March 2, 2012 at 8:24 pm #1190Anonymous
I am using the kinetic loop refinement protocol these days. Since my protein is large than 300aa and the loop is over 20 aa. So it is really time comsuming even I use 24 CPU to run the jobs. It only generate 1 model after one hour…. I am planning to generate 5000 loop models. Is it possible to generate the loop model with hydrogen? At least without hydrogen for the no loop region?
THX
-
March 3, 2012 at 3:59 pm #6738Anonymous
Rosetta builds hydrogens onto the pdb upon startup. Once you start using KIC to perturb the loop, the residues will change from all-atom to centroid in order to make the calculations faster. Since your loop is so large, I would recommend trying fragment-based loop modeling with CCD, and then maybe either KIC refinement, or just a general loop relax after the initial loop building stage. The fragment loop modeling will go quicker, and my hunch is that it may give you better results for your very long loop since your using known fragments instead of randomly sampling from the rama map and trying to close the loop. Also, you could try searching for a homologous 20aa loop and grafting it on your protein. I’m not sure how you would go about doing this, but maybe someone else on the forum does?
-
March 4, 2012 at 2:48 am #6740Anonymous
A) Loop modeling performs increasingly poorly on loops beyond 12 residues. I’ve never gotten good results on 20 residue loops.
Removing hydrogens is not possible. If you meant paring back the sidechains to centroid mode, that is possible. I guess you could run a ton of models in centroid-only, but then without doing fullatom refinement they’re hard to distinguish. Is that what you wanted…? I don’t think loop modeling supports skipping the FA step by default but it might be simple to hack into the code (loop modeling is kind of a mess so I’m not sure).
C) One hour is way too long for one model – what are your flags? Are you using MPI?
-
March 5, 2012 at 4:36 pm #6751Anonymous
yes, I compiled it with MPI option.
that’s strange, I don’t see the file you mentioned, here is what I get:
/soft/rosetta3.3/rosetta_source/build/src/release/linux/2.6/64/x86/gccbut there is no mpi directory:
AbinitioRelax.default.linuxgccrelease
AbinitioRelax.linuxgccrelease
AnchoredDesign.default.linuxgccrelease
AnchoredDesign.linuxgccrelease
AnchoredPDBCreator.default.linuxgccrelease
AnchoredPDBCreator.linuxgccrelease
AnchorFinder.default.linuxgccrelease
AnchorFinder.linuxgccrelease
antibody_mode.default.linuxgccrelease
antibody_mode.linuxgccrelease
apps
backrub.default.linuxgccrelease
backrub.linuxgccrelease
basic
benchmark.default.linuxgccrelease
benchmark.linuxgccrelease
BuildPeptide.default.linuxgccrelease
BuildPeptide.linuxgccrelease
ca_to_allatom.default.linuxgccrelease
ca_to_allatom.linuxgccrelease
cluster_alns.default.linuxgccrelease
cluster_alns.linuxgccrelease
cluster.default.linuxgccrelease
cluster.linuxgccrelease
combine_silent.default.linuxgccrelease
combine_silent.linuxgccrelease
core
CstfileToTheozymePDB.default.linuxgccrelease
CstfileToTheozymePDB.linuxgccrelease
ddg_benchmark.default.linuxgccrelease
ddg_benchmark.linuxgccrelease
ddg_monomer.default.linuxgccrelease
ddg_monomer.linuxgccrelease
design_contrast_and_statistic.default.linuxgccrelease
design_contrast_and_statistic.linuxgccrelease
docking_prepack_protocol.default.linuxgccrelease
docking_prepack_protocol.linuxgccrelease
docking_protocol.default.linuxgccrelease
docking_protocol.linuxgccrelease
enzyme_design.default.linuxgccrelease
enzyme_design.linuxgccrelease
extract_atomtree_diffs.default.linuxgccrelease
extract_atomtree_diffs.linuxgccrelease
extract_pdbs.default.linuxgccrelease
extract_pdbs.linuxgccrelease
fix_alignment_to_match_pdb.default.linuxgccrelease
fix_alignment_to_match_pdb.linuxgccrelease
fixbb.default.linuxgccrelease
fixbb.linuxgccrelease
FlexPepDocking.default.linuxgccrelease
FlexPepDocking.linuxgccrelease
FloppyTail.default.linuxgccrelease
FloppyTail.linuxgccrelease
fragment_picker.default.linuxgccrelease
fragment_picker.linuxgccrelease
full_length_model.default.linuxgccrelease
full_length_model.linuxgccrelease
gen_apo_grids.default.linuxgccrelease
gen_apo_grids.linuxgccrelease
gen_lig_grids.default.linuxgccrelease
gen_lig_grids.linuxgccrelease
idealize_jd2.default.linuxgccrelease
idealize_jd2.linuxgccrelease
InterfaceAnalyzer.default.linuxgccrelease
InterfaceAnalyzer.linuxgccrelease
libbasic.so
libcore.1.so
libcore.2.so
libcore.3.so
libcore.4.so
libcore.5.so
libnumeric.so
libObjexxFCL.so
libprotocols.so
libutility.so
ligand_dock.default.linuxgccrelease
ligand_dock.linuxgccrelease
ligand_rpkmin.default.linuxgccrelease
ligand_rpkmin.linuxgccrelease
loopmodel.default.linuxgccrelease
loopmodel.linuxgccrelease
loops_from_density.default.linuxgccrelease
loops_from_density.linuxgccrelease
match.default.linuxgccrelease
match.linuxgccrelease
membrane_abinitio2.default.linuxgccrelease
membrane_abinitio2.linuxgccrelease
minimize_with_cst.default.linuxgccrelease
minimize_with_cst.linuxgccrelease
minirosetta.default.linuxgccrelease
minirosetta_graphics.default.linuxgccrelease
minirosetta_graphics.linuxgccrelease
minirosetta.linuxgccrelease
mpi_msd.default.linuxgccrelease
mpi_msd.linuxgccrelease
mr_protocols.default.linuxgccrelease
mr_protocols.linuxgccrelease
numeric
ObjexxFCL
pepspec_anchor_dock.default.linuxgccrelease
pepspec_anchor_dock.linuxgccrelease
pepspec.default.linuxgccrelease
pepspec.linuxgccrelease
protocols
relax.default.linuxgccrelease
relax.linuxgccrelease
revert_design_to_native.default.linuxgccrelease
revert_design_to_native.linuxgccrelease
rna_database.default.linuxgccrelease
rna_database.linuxgccrelease
rna_denovo.default.linuxgccrelease
rna_denovo.linuxgccrelease
rna_design.default.linuxgccrelease
rna_design.linuxgccrelease
rna_extract.default.linuxgccrelease
rna_extract.linuxgccrelease
rosettaDNA.default.linuxgccrelease
rosettaDNA.linuxgccrelease
rosetta_scripts.default.linuxgccrelease
rosetta_scripts.linuxgccrelease
rotamer_recovery.default.linuxgccrelease
rotamer_recovery.linuxgccrelease
score_aln2.default.linuxgccrelease
score_aln2.linuxgccrelease
score_aln.default.linuxgccrelease
score_aln.linuxgccrelease
score.default.linuxgccrelease
score_jd2.default.linuxgccrelease
score_jd2.linuxgccrelease
score.linuxgccrelease
sequence_recovery.default.linuxgccrelease
sequence_recovery.linuxgccrelease
sequence_tolerance.default.linuxgccrelease
sequence_tolerance.linuxgccrelease
super_aln.default.linuxgccrelease
super_aln.linuxgccrelease
SymDock.default.linuxgccrelease
SymDock.linuxgccrelease
UBQ_E2_thioester.default.linuxgccrelease
UBQ_E2_thioester.linuxgccrelease
utility -
March 6, 2012 at 9:16 pm #6762Anonymous
thank you very much for kind reply.
it is a little bit strange when I using the command for compling:
scons bin mode=release extras=mpi -j24it said:
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelLoopMover.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelConstraintGenerator.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/ResidueVicinityRCG.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelAccumulator.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelEnzdesCstModule.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelData.os] Error 127
sh: mpiCC: command not found
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelWorkingSet.os] Error 127
scons: building terminated because of errors.in fact, I type: which mpiCC
/usr/lib64/mpi/gcc/openmpi/bin/mpiCCwhy Rosetta cannot find it? do you have any idea to fix?
thank you very much
-
March 6, 2012 at 10:03 pm #6763Anonymous
You’re getting this error because the path where mpiCC is installed is not being given to scons. The easiest way to fix this is to navigate to the rosetta/rosetta_source/tools/build directory and run the following command: ‘ln -s site.settings.killdevil site.settings’. This will cause anything in your PATH environment variable to be found by scons.
-
March 5, 2012 at 1:00 pm #6745Anonymous
thank you very much for your kind helps.
I do use MPI and here is my command:
mpiexec -np 24 loopmodel.linuxgccrelease -database /opt/rosetta3.3_bundles/rosetta_database @flags >log&
here is my flags:
-nstruct 1000 #number of models to build. 1000 recommended for production runs.
-in:file:native em.pdb
-loops:input_pdb em.pdb #starting pdb with loops to rebuild
-loops:loop_file input.loop #loop file-loops:relax fastrelax #does a minimization of the structure i
n the torsion space
-loops:extended #force phi-psi angles to be set to 180 degrees
independent of loop input file (recommended for production ru
ns)
-loops:remodel perturb_kic
-loops:refine refine_kic
-loops:optimize_only_kic_region_sidechains_after_move
-ex1 #Include extra chi1 rotamers
-ex2#-out:file:silent_struct_type binary #output file type
#-out:pdb
-loops:output_pdb aa_*pdb
-out:sf score.fsc#-out:file:silent loops.out #output file name
-out:file:fullatom #output file will be fullatom-run:jran 1578946
-run:constant_seed -
March 5, 2012 at 2:57 pm #6748Anonymous
Did you compile in MPI? In other words, if loopmodel.linuxgccrelease is a symlink, does it link to something that has “mpi” in its path somewhere? If you compile in MPI with extras=mpi, then compile without MPI, the executables in bin/ symlink to the newer executables, and you need to directly point at build/src/release/linux/3.0/64/x86/gcc/mpi/loopmodel.linuxgccrelease. Notice the mpi in the path.
-
March 5, 2012 at 9:18 pm #6753Anonymous
Rosetta doesn’t think you compiled MPI. Try recompiling with extras=mpi and see what it does….
-
March 7, 2012 at 6:20 am #6770Anonymous
thank you very much for your kind messages. but it still doesn’t work. do you have any else advices for this issue? here is the log file:
sh: mpiCC: command not found
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/apps/public/AbinitioRelax.o] Error 127
mpiCC -o build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelWorkingSet.os -c -isystem external/boost_1_46_1/boost/ -isystem external/boost_1_46_1/boost/ -O3 -ffast-math -funroll-loops -finline-functions -finline-limit=20000 -s -Wno-unused-variable -fPIC -DNDEBUG -DUSEMPI -Isrc -Iexternal/include -Isrc/platform/linux/64/gcc -Isrc/platform/linux/64 -Isrc/platform/linux -Iexternal/boost_1_46_1 -Iexternal/dbio -I/usr/local/include -I/usr/include src/protocols/forge/remodel/RemodelWorkingSet.cc
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelLoopMover.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelConstraintGenerator.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/ResidueVicinityRCG.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelAccumulator.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelEnzdesCstModule.os] Error 127
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelData.os] Error 127
sh: mpiCC: command not found
scons: *** [build/src/release/linux/2.6/64/x86/gcc/mpi/protocols/forge/remodel/RemodelWorkingSet.os] Error 127
scons: building terminated because of errors.
-
-
AuthorPosts
- You must be logged in to reply to this topic.