When loading in coordinates for ligands, Rosetta ignores the ATOM/HETATM difference. It should work as-is with no extra effort on your part if you have the ligand as ATOM records in the pdb file.
Note, however, that the residue and atom names are very important to Rosetta, and need to match those in the corresponding .params file. If you've made the params file with a typical "excise ligand record, convert to molfile, molfile_to_params" process, often the easiest thing to do is cut-and-paste (e.g. with a text editor) the output ligand pdb file that molfile_to_params gives into the starting protein PDB. The resultant coordinates should be in the same place as the starting atom coordinates (something you can easily check with, e.g., PyMol).
The reason I ask this question is because I intend to specify resfile for the ligand as well as the protein. I understand that I can do several pre-designed ligand instead in order to accomplish the same job. My ligand is essentially something like: aa sequence "FGK". It's just that the ligand in ligand_docking app has got only one residue instead of three, therefore maybe I should use another app for this purpose?
Why do you want to specify a resfile for the ligand with the ligand_dock application? The ligand_dock application doesn't do any mutation, and the concept of docking implies full flexibility, so I don't know what you'd be specifying with the resfile.
The ligand_dock application is mainly intended for small molecule-protein docking, so if you're doing peptide docking, things might not work so well. Usually even if you're docking a single amino acid, you typically need to "fake it" as a small molecule (probably a better bet anyway, as Rosetta may or may not correctly handle the appropriate deprotonation/capping of the termini.)
Protonation and deprotonation is a good point. I have used flexPepDock app but it doesn't seem to fit my purpose for now.
I am running simulations trying to understand some small pocket interactions, making an initial pose of the peptide right inside the pocket. I don't need too much flexibility on the protein side too.
I am wondering if the resfile has chain ID spec, and if the ligand can be specified too as amino acids, it would be convenient. Or maybe I should simply specify the peptide as a part of the protein?
When loading in coordinates for ligands, Rosetta ignores the ATOM/HETATM difference. It should work as-is with no extra effort on your part if you have the ligand as ATOM records in the pdb file.
Note, however, that the residue and atom names are very important to Rosetta, and need to match those in the corresponding .params file. If you've made the params file with a typical "excise ligand record, convert to molfile, molfile_to_params" process, often the easiest thing to do is cut-and-paste (e.g. with a text editor) the output ligand pdb file that molfile_to_params gives into the starting protein PDB. The resultant coordinates should be in the same place as the starting atom coordinates (something you can easily check with, e.g., PyMol).
Thanks rmoretti~
The reason I ask this question is because I intend to specify resfile for the ligand as well as the protein. I understand that I can do several pre-designed ligand instead in order to accomplish the same job. My ligand is essentially something like: aa sequence "FGK". It's just that the ligand in ligand_docking app has got only one residue instead of three, therefore maybe I should use another app for this purpose?
Any suggestion?
Why do you want to specify a resfile for the ligand with the ligand_dock application? The ligand_dock application doesn't do any mutation, and the concept of docking implies full flexibility, so I don't know what you'd be specifying with the resfile.
The ligand_dock application is mainly intended for small molecule-protein docking, so if you're doing peptide docking, things might not work so well. Usually even if you're docking a single amino acid, you typically need to "fake it" as a small molecule (probably a better bet anyway, as Rosetta may or may not correctly handle the appropriate deprotonation/capping of the termini.)
There is a different, flexible peptide docking application (http://www.rosettacommons.org/manuals/archive/rosetta3.4_user_guide/d7/d...), although I've never used it. The other thing you might be able to do is put together something with RosettaScripts.
Hi rmoretti,
Protonation and deprotonation is a good point. I have used flexPepDock app but it doesn't seem to fit my purpose for now.
I am running simulations trying to understand some small pocket interactions, making an initial pose of the peptide right inside the pocket. I don't need too much flexibility on the protein side too.
Yes, ligand_dock app has an option to specify the resfile. It is described here:
http://www.ncbi.nlm.nih.gov/pubmed/22183535
I am wondering if the resfile has chain ID spec, and if the ligand can be specified too as amino acids, it would be convenient. Or maybe I should simply specify the peptide as a part of the protein?