hello:
I am using the following protocol which someone used in the forum to build loop based on crystal structure. I am going to fix the left part of the region since they are from crystal structure and I find that all the regions are rebuild by Rosetta. I am just wondering, where is the problem?
thank you very much
/opt/rosetta3.4/rosetta_source/bin/loopmodel.default.linuxgccrelease \
-database /opt/rosetta3.4/rosetta_database \
-nstruct $nstruct \
-loops:input_pdb ../crystal.pdb \
-loops:loop_file ../input.loop \
-loops:remodel perturb_kic \
-loops:refine refine_kic \
-loops.max_kic_build_attempts 100 \
-loops:fix_natsc true \
-loops:neighbor_dist 0.0 \
-loops:relax fastrelax \
-loops:extended \
-ex1 \
-ex2 \
-out:path:pdb ./ \
-out:suffix $i \
-out:file:fullatom \
-out:pdb \
Double check your loop file....
What does "left part" mean...?
I checked many times, and I don't know where the problem it is.....
Is the backbone moving or just the sidechains? What do the occupancies of the moving region look like in the PDB (are they nonzero)? Post the loop file (just paste it into the text box).
both backbone and side chain moves. I trying to include the hydrogen as the input file, but it always claim loop definition out of range. So, I removed all hydrogen as initial structure. Here is the loop file:
LOOP 193 208 200 0 1
If those values are meant to be PDB numberings, you need to include the PDB chain; for example
LOOP 193A 208A 200A 0 1
Rosetta interprets wthem as 1-indexed residue IDs (whole pose numbered starting from 1) normally.
thanks for kind comments. In fact, I remubered the residues from 1.....
You have fastrelax turned on - try turning that off. I don't know if fastrelax limits itself to just the defined loop.
IC, it works now. However, I find my disfulde bridge was borken.....
THX