You are here

symmetric design in the presence of a nonsymmetric ligand

7 posts / 0 new
Last post
symmetric design in the presence of a nonsymmetric ligand
#1

Hello,

I am trying to design a C3 symmetrical protein with a ligand bound at the C3 interface. What is the best way around this in RosettaScripts? I was thinking to first symmetrize the pose with SetupForSymmetry and then add the ligand with AddChain mover. But it seams, SetupForSymmetry is not compatible with extra_res. As soon as the script arrives at SetupForSymmetry, it throws the following error:

"The residue INV_VRT could not be generated.  Has a suitable params file been loaded?  (Note that custom params files not in the Rosetta database can be loaded with the -extra_res or -extra_res_fa command-line flags."

I have loaded the the params files with these flags "-extra_res_cen XXX.cen.params" and "-extra_res_fa XXX.fa.params" . 

Any suggestions would be greatly appreciated,

Thanks!

Post Situation: 
Fri, 2016-09-30 15:17
msardejani

The issue you're having is with the  INV_VRT type, rather than the ligand type you're doing. (I presume you'd get the same error if you didn't use the ligand.)

There's two possiblities. The first is that you're using a database version that's different from the executable version. Don't. Use the version of the database which comes with the version of Rosetta you're using. You may want to check your environment variables to see if you have the envrionment variable set to an older version. -- At any rate, explicitly using the `-database` option on the commandline with the appropriate directory should work.

The other possibility is you have a version of Rosetta that has an issue with symmetry and centroid mode. Look at the database. You should have INV_VRT.params file in both database/chemical/residue_type_sets/centroid/residue_types/rosetta_specific/ and database/chemical/residue_type_sets/fa_standard/residue_types/rosetta_specific/ 

If you have only the latter, you can either get an updated version of Rosetta, or you can pass the database/chemical/residue_type_sets/fa_standard/residue_types/rosetta_specific/INV_VRT.params file with the -extra_res_cen option.

Fri, 2016-09-30 15:31
rmoretti

Thanks for the reply, really appreciate it.

I am using 2016.32.58837 with the database that comes with it. And INV_VRT.params is in both locations that you mentioned.  I also tried passing the INV_VRT.params with -extra_res_cen option but that didn't help either.

In the begining of the log file there are these lines:

core.chemical.ResidueTypeSet: (0) For ResidueTypeSet centroid there is no shadow_list.txt file to list known PDB ids.
core.chemical.ResidueTypeSet: (0)     This will turn off PDB component loading for ResidueTypeSet centroid
core.chemical.ResidueTypeSet: (0)     Expected file: ~/rosetta/rosetta_src_2016.32.58837_bundle/main/database/chemical/residue_type_sets/centroid_pre_talaris2013/shadow_list.txt
core.chemical.ResidueTypeSet: (0) Finished initializing centroid residue type set.  Created 33 residue types

 

Is this normal?

Fri, 2016-09-30 18:26
msardejani

Sorry, I had passed -restore_pre_talaris_2013_behavior so and apparently this was causing the problem. 

 

 

Sat, 2016-10-01 00:17
msardejani

Is that possible at all to have a nonsymmetric ligand in a symmetirc pose?

I symmetrized the pose with SetupForSymmetry and then added the ligand with AddChain mover. The problem now is, because the pose is C3 symmetrical, three copies of ligand get added. 

Sat, 2016-10-01 01:18
msardejani

No, or at least with out major (untested) contorsions. One reason is that once you put a non-symmetric ligand into the pose, you break the symmetry of the structure.

The approach you'll want to take with this is to enforce symmetry without having "true" symmtry. For design this would normally take the form of ResidueTypeLinkingConstraints, usually added with the FavorSymmetricSequence mover in RosettaScripts (https://www.rosettacommons.org/docs/latest/scripting_documentation/RosettaScripts/Movers/movers_pages/FavorSymmetricSequenceMover) .

This will enforce symmetry by adding a penalty to designs which are asymmetric. Usually what you would want to do is run multiple rounds of design, increasing the weight on the res_type_linking_constraint each time. At very high levels, you should only get symmetric sequences. (Though you'll still have structural differences between the multiple subunits.)

Sat, 2016-10-01 11:21
rmoretti

Thanks very much! I shall try this.

Sat, 2016-10-01 12:55
msardejani