weird prolines with hydroxyls from fixbb.linuxgccrelease

Member Site › Forums › Rosetta 3 › Rosetta 3 – General › weird prolines with hydroxyls from fixbb.linuxgccrelease

Viewing 2 reply threads
  • Author
    Posts
    • August 25, 2010 at 10:01 pm #659
      Anonymous

        Hello, the fixbb executable from 3.1.0 is placing hydroxyl groups on my proline rings when I try to repack them (using NATAA in resfile). fixbb from 3.0.0 has the same problem when run with the 3.1.0 database, but a pre-3 mini release does not. Anybody seen that before? Any quick fixes?

        If somebody is interested in pursuing this (and promises to get back to me), I could provide a tar.gz directory with an example in it.

        thanks,

        -Louis

      • August 26, 2010 at 1:21 pm #4587
        Anonymous

          Placing hydroxyprolines is weird. I assume from your tone that you are not deliberately doing anything with hydroxyprolines anywhere in your trajectory. I assume these are the standard hydroxyprolines, like in collagen? http://en.wikipedia.org/wiki/Hydroxyproline

          I don’t have a copy of the 3.1 database to check but you probably have something screwy in the params files. You may want to delete your copy and refresh it, (or just use the 3.0 database with 3.1; it’s equivalent for fixbb). You can also look through the files (database)/chemical/residue_type_sets/fa_standard/patches.txt and residue_types.txt and see where the hydroxyproline is and just remove it from the allowed set. You can also try deleting any large files of the pattern (database)/Dunbrack*.lib.bin from your database, maybe they got corrupted somehow (or you moved them between systems)

        • August 27, 2010 at 11:13 pm #4588
          Anonymous

            Thanks for the answer Steven. I agree that it’s probably related to the library. I’m not trying to do anything weird. Sounds like erasing hydroxylprolines from the database would be the best bet. I’ll let you know how it goes when I try it. By the way, a late-July copy of the trunk seems to do the same thing, so you might want to be on the look-out.

            thanks,

            -Louis

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