Is there any support in rosetta for writing out a PDB file with multiple configurations of the side-chains? I have a set of side-chain hypotheses at each residue along a fixed backbone and would like to visualize this as an ensemble. This is analogous to dumping a PDB file with a side-chain at every rotamer location in the Dunbrack rotamer set.
I think I know how to implement this from scratch, but if there is any built-in support along these lines that would be great.
I am not aware of anything in Rosetta that allows for multi-model output of PDBs.
There is something similar in the “CloudPDB” output format for the matcher, but that’s something of a special use case, and doesn’t have general support. (And isn’t something that’s even readily extendible even in it’s current form.)
I implemented my own version by modifying the dump_pdb_residue. The PDB format supports multiple conformatons but only gives one character to differentiate the conformations. I’m not sure if it is totally standard, but I extended this to support up to 36 possible configurations by using [A-Z] and [0-9]. PyMol seems to have no problem displaying this, but I haven’t tried other viewers.
I’ve attached the source as it is pretty general and might be useful for others. I’ve also attached an image of a small protein example that I loaded in PyMol showing several conformations for each side-chain.
To be totally correct one should probably modify this by appropriately setting the occupancy field. I am just setting it to 1.0 for all atoms which isn’t technically correct, but I’m just using this for visualization.