You are here

RemodelMover and Input Pose

1 post / 0 new
RemodelMover and Input Pose
#1

I have been trying to create an automated protocol for modelling a covalently linked peptide. Using PeptideStubMover I have created the covalent link and the first three residues. For the rest of this 12 aa long peptide, I wanted to use the remodel mover to have control over the secondary structure and amino acid composition. But it appears that RemovelMover cannot use the updated pose. I receive a segmentation fault. bp.txt does contain the new residue that is added by PeptideStubMover. If I run two movers discretely (run peptidestubmover, save pdb, run remodelmover) there are no errors. 

<PeptideStubMover name="linker" reset="false" update_pdb_numbering="true">
	<Append resname="GLY" repeat="1" jump="false"  anchor_rsd="%%target_res%%" connecting_atom="%%linker_atom%%" anchor_atom="%%target_atom%%" />
	<Append resname="GLY" repeat="2" jump="false"  anchor_rsd="14" />
</PeptideStubMover>

<RemodelMover name="link_tag" blueprint="bp.txt" quick_and_dirty="1"/>

<PROTOCOLS>
      <Add mover="linker"/>
      <Add mover="link_tag"/>
</PROTOCOLS>


I thought I had a brilliant idea of writing and rereading the pdb file after the `PeptideStubMover`.  But it appears that `tmp_linked.pdb` is required and read at the start of the protocol, although DumpPdb updates it midrun, SavePoseMover does not read the updated version. If I supply the correct pdb file for RemodelMover in the `tmp_linked.pdb` from the start the run continuous.
 

<PeptideStubMover name="linker" reset="false" update_pdb_numbering="true">
	<Append resname="GLY" repeat="1" jump="false"  anchor_rsd="%%target_res%%" connecting_atom="%%linker_atom%%" anchor_atom="%%target_atom%%" />
	<Append resname="GLY" repeat="2" jump="false"  anchor_rsd="1368" />
</PeptideStubMover>

<DumpPdb name="save_linked" fname="tmp_linked.pdb"/>
<SavePoseMover name="load_linked" restore_pose="1" reference_name="linked_pose" pdb_file="tmp_linked.pdb"/>


<RemodelMover name="link_tag" blueprint="bp.txt" quick_and_dirty="1"/>

<PROTOCOLS>
      <Add mover="linker_site"/>
      <Add mover="save_linked"/>
      <Add mover="load_linked"/>
      <Add mover="link_tag"/>
</PROTOCOLS>


Now after this long story (and probably confusing), my question is how I can update the pose after PeptideStubMover such that RemodelMover gets the correct pose? Is there a way to read the pdb that is just written by DumpPdb?

I have read on the manual that RemodelMover has it's own IO, I am guessing it's not possible unless I modify its code. Running these two Movers discretely works too. I am just trying to push the limits of automating the protocols with fewer scripts and learning what I can and cannot do.

Post Situation: 
Tue, 2020-12-29 09:41
ozyo