Repeating sequence in Rosetta de novo protein folding

Member Site › Forums › Rosetta 3 › Rosetta 3 – General › Repeating sequence in Rosetta de novo protein folding

Viewing 2 reply threads
  • Author
    Posts
    • February 8, 2018 at 4:52 am #2847
      Anonymous

        Hi,

        I am trying to do a de novo protein folding from a protein sequence with repeating Gly and Arg only.

        So the sequence looks like GRGRGRGRGR….. I put the sequence onto Robetta Fragment Server and it gave me this error message:

        =======================================================================================

        Target:    GR25

        Status:    Error

        Date Submitted:    02/07/18 01:54:32 AM

        Started:    02/07/18 04:29:00 AM

        Ended:    02/07/18 05:07:22 AM

        Expires:    02/14/18 05:07:22 AM   (6 days 08:41:40)

        Sequence:    

        GRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGRGR

        Length:    50

        Exclude Homologues:    no

        Error

        sam target99 failed! 

        =========================================================================================

        This might be a dumb question but I was wondering if there are some restrictions on the usage of de novo protein folding in Rosetta when the sequence is repeating…?

        Thanks for your help! 

         

        Orion

         

      • February 8, 2018 at 5:16 pm #14033
        Anonymous

          Fragment prediction is based on finding chunks of PDBs with similar sequences and/or similar predicted secondary structure to the request.  I would guess that nothing looks like this that’s in the PDB, so the system errors out.  I guess try BLASTing this sequence against what’s been crystallized to see if there are any hits.

        • February 9, 2018 at 5:18 am #14034
          Anonymous

            Do you have any reason to believe that your sequence will be folded/structured in vivo?

            If you perform disorder prediction by submitting that sequence to DISOPRED3 at bioinf.cs.ucl.ac.uk/psipred/ then you will see that it is predicted to be “disordered state, protein binding”  with a confidence near 1.0.

            Warning: If you ask Rosetta to fold a sequence, it will give you the best folded structure it can find. But if that sequence prefers to be unstructured, then your prediction may have little value.

             

        • Author
          Posts
        Viewing 2 reply threads
        • You must be logged in to reply to this topic.