Reply To: Modeling phosphate ion binding site in protein.

[Anonymous]

I you have a good idea as to the general area where the phosphate should dock (e.g. from mutational studies or homology considerations), it might be helpful to limit the docking runs to that region. You can specify the areas to sample with the -docking:ligand:start_from and -docking:uniform_trans flags.

For a really thorough study, you may want to locate crystal structures of protein-phosphate (or small-molecule phosphates) complexes with known affinities. You can then run these controls through the protocol to benchmark how well Rosetta does in predicting phosphate position, and also to get a more phosphate-specific calibration curve (and potentially error bars) for the REU to Kd conversion.