fixbb with symmetry error

Member Site › Forums › Rosetta 3 › Rosetta 3 – Applications › fixbb with symmetry error

Viewing 3 reply threads
  • Author
    Posts
    • January 26, 2014 at 6:56 pm #1813
      Anonymous

        Dear rosetta_users,

        I need to make same mutation at a single position in all three chains of a homotrimer by using ALLAA (all_natural_aminoacids in the resfile).

        I have used the below command to run fixbb for the homotrimer with the symmetry flags, and the fixbb gave output without any error but the output had 30 chains of the monomer(outputfile=16.4MB, inputfile=845KB).

        ### ~/rosetta-3.4/rosetta_source/bin/fixbb.linuxgccrelease -database ~/rosetta-3.4/rosetta_database/ -s pdb.pdb -resfile resfile.resfile -symmetry:symmetry_definition pdb.symm

        I created pdb.symm by running the below command :
        ### perl ~/rosetta-3.4/rosetta_source/src/apps/public/symmetry/make_symmdef_file.pl -m NCS -p pdb.pdb -a A -i B C > pdb.symm

        So how do i change the commands/flags to make proper symmetrical mutations.

        Thanks in advance!!!

      • January 26, 2014 at 7:29 pm #9715
        Anonymous

          If the symmetry definition file is set up appropriately (and the protocol is properly symmetry aware), making a mutation for one chain should automatically make the mutations on all the symmetry partners.

          Getting 30 chains in the output indicates that the likely issue is with your symmetry file. Rosetta likely is using the symmetry file to build what it thinks are the missing symmetric partners, and is apparently making too many.

          I can’t see anything obviously wrong with your command lines, could you post your input pdb file and your symmetry definition file?

        • January 27, 2014 at 9:33 am #9716
          Anonymous

            Dear rmoretti,
            Please see the attached .symm file.
            Thanks

          • January 28, 2014 at 4:27 pm #9720
            Anonymous

              Dear rmoretti,
              I think i have solved the problem. The problem was with the commands i used to create .symm file.

              Initially i used
              ### perl ~/rosetta-3.4/rosetta_source/src/apps/public/symmetry/make_symmdef_file.pl -m NCS -p pdb.pdb -a A -i B C > pdb.symm – WRONG (which gave more than 30chains)

              But when i used
              ### perl ~/rosetta-3.4/rosetta_source/src/apps/public/symmetry/make_symmdef_file.pl -m NCS -p pdb.pdb -a A -i B > pdb.symm – WORKING
              (I used {-i B} instead of {-i B C}, the mutations were symmetrical in all three chains of the trimer).

              Can u just tell us how does option -a & -i needs to be used (for trimer the option i used is working, but is it perfect and also what options do we need to use for higher symmetries; sorry to say that the documentation was not helpful in this regard).

              Also if we want to include, say chain D (say a core protein in complex with a trimer) apart from the trimers chains A, B, C what options need to be used to make symmetrical mutations? I hope this answer will be useful for all users!!!

              Thanks for the great help.

            • January 27, 2014 at 4:07 pm #9719
              Anonymous

                It looks like your symmetry file is for a trimer of trimers. When you generated it, did you cut down the input PDB to just the single trimer that you wanted to model? If you have additional crystallographic partners in the file you passed to make_symmdef_file.pl, it will attempt to include those in the symmetry definition.

                Additionally, there is sometimes issues when you read in your PDB – often you’ll need to provide the monomer to the Rosetta run, and have Rosetta make the symmetric partners. What may be happening is that you’re providing your trimer (or trimer-of-trimers) to Rosetta, and then it’s symmetrizing it based on your trimer-of-trimer symmetry file. So instead of a trimer, you get 3×9 or 9×9 copies of the monomer in the output.

            • Author
              Posts
            Viewing 3 reply threads
            • You must be logged in to reply to this topic.