Knowledge check on steric effects

Member Site › Forums › Rosetta 3 › Rosetta 3 – General › Knowledge check on steric effects

Viewing 1 reply thread
  • Author
    Posts
    • October 16, 2014 at 4:13 pm #2036
      Anonymous

        Hi,

        I am doing packing on PDB ID 1UKU and looking at two solutions my code produces, one having much higher energy than the other. The only difference between them is that the conformation of one of the Tryptophan residues is different, the other 100 or so residues are unchanged.

        Visualizing the result in PyMol it looks like the higher energy configuration is interacting with a nearby Glutamic Acid residue. I’ve attached an image showing just these two residues (Res95 Trp and Res99 Glu). The image shows the two conformations for Trp, one which is interacting with Glu and has higher energy.

        Because my chemistry background is poor, I’m wondering if someone could briefly tell me specifically what type of interaction is happening here. And in particular, in Rosetta what energy term would be evaluating this interaction?

        Jason

      • October 16, 2014 at 5:33 pm #10451
        Anonymous

          PDBs (at least the PDBs which Rosetta produces) don’t have explicit connectivity information. Which atoms are bonded to which is implicit. So when PyMol reads in a PDB, it has to run through a few heuristics in order to determine which atoms are bonded. One of the major heuristic PyMol uses is atom-atom distances. If two atoms are within what PyMol considers to be bonding distance, it will draw a line representing a bond between them. It’s important to keep in mind that that doesn’t necessarily mean that there’s any sort of actual physical bonding-type interaction between the two atoms – they’re just close in space.

          Rosetta doesn’t use distance to determine atom bonding patterns – bonds between atoms are encoded in the chemical identity of the residues in the structure. So in this situation, while PyMol is drawing a “bond” between the atoms in the Trp and Glu, Rosetta wouldn’t see any such “bonding” interaction. The lines you see are *entirely* due to PyMol’s (inaccurate) distance heuristic.

          What you will see in Rosetta is calculated energies due to non-bonded effects. Specifically, most of the energy difference will likely be from the van der Waals (Lennard-Jones) repulsive interaction between the atoms which are too close (fa_rep). Other terms might also contribute, but fa_rep is the main term whose job it is to keep such close-approaches between non-bonded atoms from happening.

      • Author
        Posts
      Viewing 1 reply thread
      • You must be logged in to reply to this topic.