How to do ab initio with fixed part of protein

You want to look at rosetta_demos/homology_modeling_with_end_extension demo. Note that the demo is written for Rosetta 3.3 rather than 3.4.

This is it! I’ve skipped the first part because I had already a model. I forgot, however to build the full model. If I remember it wasn’t necessary for Rosetta-3.3.

This is it! I’ve skipped the first part because I had already a model. I forgot, however to build the full model. If I remember it wasn’t necessary for Rosetta-3.3. Thanks for helping.

Will the extension script (hm_with_extension) work with Rosetta52wk-mpi?
I was trying to extend 23 N-terminal residues and got the following error. Any help is appreciated.

#########################################################################################
protocols.relax.FastRelax: (3) ================== Using default script ==================
protocols.evaluation.ChiWellRmsdEvaluatorCreator: (3) Evaluation Creator active …
protocols.jd2.MPIFileBufJobDistributor: (1) TIME_STAMP: Tue Feb 4 17:49:40 2014 assign job 1 batch: 00000
protocols.general_abinitio: (3) AbrelaxMover: S_0001
protocols.general_abinitio: (3) TIME_STAMP: Tue Feb 4 17:49:40 2014 AbrelaxMover: start…NO_BATCH S_0001
loops: (3) Extended: (-0,+2) LOOP 1 25 0 0 0
loops: (3) Extended: (-2,+0) LOOP 309 510 0 0 0
protocols.topo_broker: (3) Enlarged loops:
protocols.topo_broker: (3) LOOP begin end cut skip_rate extended
protocols.topo_broker: (3) LOOP 1 25 0 0 0
protocols.topo_broker: (3) LOOP 309 510 0 0 0
protocols.topo_broker: (3)
protocols.topo_broker: (9) Edge Jump Jump #
0001–0310

protocols.topo_broker: (4) Edge Jump Jump #
0001–0310

protocols.topo_broker: ( Edge Jump Jump #
0001–0310

protocols.topo_broker: (5) Edge Jump Jump #
0001–0310

protocols.topo_broker: (7) Edge Jump Jump #
0001–0310

protocols.topo_broker: (2) Edge Jump Jump #
0001–0310

ERROR: lastChunk <= centroid_input_pose_.total_residue()
ERROR:: Exit from: src/protocols/topology_broker/RigidChunkClaimer.cc line: 508

ERROR: lastChunk <= centroid_input_pose_.total_residue()
ERROR:: Exit from: src/protocols/topology_broker/RigidChunkClaimer.cc line: 508

ERROR:

lastChunk <= centroid_input_pose_.total_residue()
ERROR:: Exit from: src/protocols/topology_broker/RigidChunkClaimer.cc line: 508
ERROR: lastChunk <= centroid_input_pose_.total_residue()
ERROR:: Exit from: src/protocols/topology_broker/RigidChunkClaimer.cc line: 508
ERROR: lastChunk <= centroid_input_pose_.total_residue()
ERROR:: Exit from: src/protocols/topology_broker/RigidChunkClaimer.cc line: 508
protocols.jd2.JobDistributor: (

[ERROR] Exception caught by JobDistributor for job S_0006

[ERROR] EXCN_utility_exit has been thrown from: src/protocols/topology_broker/RigidChunkClaimer.cc line: 508
ERROR: lastChunk <= centroid_input_pose_.total_residue()

The demo says:

“Outline
We will break down homology_modeling_with_end_extension into 2 parts:
Part I describes a protocol for homology modeling of the protein (the region of the protein that has a homolog in the pdb).
Part II describes a protocol for ab initio modeling of the n-terminal region while keeping the previously modeled homologous region in Part 1 rigid.”

It seems to me that Part II is exactly what you want.

“Error opening RBSeg file” suggests that the file is missing, not misformatted. Check that your path to the 2B.region file in your options file is correct, that the filename is correct, and that Rosetta has permission to access the file.

The error “Error opening RBSeg file” occurs in many places in Rosetta, unfortunately. However, in all of them, it is code similar to this:

protocols/loops/LoopsFileIO.cc
162- std::ifstream infile( filename.c_str() );
163-
164- if (!infile.good()) {
165: utility_exit_with_message( “[ERROR] Error opening RBSeg file ‘” + filename + “‘” );
166- }

This means that the error occurs when the file is OPENED, not when the file is READ. In other words, it cannot be a formatting error, it is an error in opening the file. That usually means the file isn’t present, the file name is wrong, or the file path is unavailable to Rosetta.

I see 2B.region in your tpb file – is that what the file is named in the same directory? Maybe it needs to be ./2B.region or similar? As far as I can tell the formatting is correct. Try deleting the line and rewriting it, maybe there’s a nonprinting character causing a problem?

My guess is that you’ve put the location of the 2B.region file to be the relative location to the tpb file, but Rosetta is interpreting it as relative to the running directory. That is, if the 2b.tpb and 2B.region files are in the /home/eunwook/runningdirectory/rosetta_inputs/ directory, and the 2b.tpb file simply lists it as “2B.region”, when you run Rosetta in the directory /home/eunwook/runningdirectory/, Rosetta will be expecting it at /home/eunwook/runningdirectory/2B.region.

Try using the absolute path to 2B.region in the tpb file, rather than a relative path.

I’m not entirely sure which script you’re referring to – where did you find it?

At any rate, I’m guessing that the step equivalent to “Generate a full length model if the n-terminal region was omitted” from the Rosetta/demos/public/homology_modeling_with_end_extension/ protocol was skipped.

If that’s the case you’ll have an input protein that’s 23 amino acids shorter than it’s supposed to be. That means that your amino acid designations for the rigid chunk won’t match up with the input protein, and you’ll get errors similar to what you’re seeing.