I'll explain what I do, in case someone has a solution to my problem.
I am trying to perform a docking protein-protein, where, a membrane protein arranged in a lipid membrane and balanced with water and ions, should be bound to a protein.
I need to know how much these structures interfere with docking.
Rebecca Alford, one of the membrane devs, says:
We do have an early version of a membrane protein-protein docking method. However, Rosetta membrane uses an implicit solvation model to emulate the hydrophobic lipid bilayer. So, this method won't account for the interference of structured waters at the interface. If you just want to dock proteins, use the MPDock method described here: http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004398
I do remember Patrick Barth's lab was working on modeling explicit waters that interfere with the docking of ligands into GPCRs. If this is what you are interested in, I would contact their lab to see if this protocol has made any progress and is available in the public release. Otherwise, I'm unsure Rosetta is the right tool for your task.
Thank you Lewis.
I know a little about the functioning of the implicit solvation, I even used Rebecca's protocol, but it did not solve my problem.
The main idea is how the whole environment influences the docking, ie, membrane, water and some metal ions. You know how much this environment interferes or assists in docking.
On the membrane, the problem is partially solved, I created some parameter files (https://www.rosettacommons.org/demos/latest/tutorials/prepare_ligand/prepare_ligand_tutorial), which at an early stage of the docking (docking prepack) solved the Problem, but an error appears in the docking.
If it's all right for you to contact Patrick Barth, thank you very much.
I'll forward this along to Patrick, but I do think you need to consider the limitations of the scorefunction. It's not set up for explicit treatment of all atoms.