Dear all, 

I am currently refining a rather well resolved protein at 0.97Å. However, even after many cycles of refinement the R-factors are stuck at around Rwork/Rfree 14.5/16.5. Considering number of aa, Rmerge (3.5%) and resolution this seems to be too high. I assume 10 – 25 macrocycles should be enough for effective weight refinement in phenix, however no improvement of the R-factors. 

A quick run with default settings in competitor software 'R' yields >2% better R-factors. 

I would appreciate suggestion how to improve my refinement protocol, values differing from default are listed below. 

Thank you very much in advance

Eike

    

refine {

    adp {

      individual {

        isotropic = element H

        anisotropic = not element H

      }

      group_adp_refinement_mode = one_adp_group_per_residue \

                                  *two_adp_groups_per_residue group_selection

    }

  }

  main {

    apply_overall_isotropic_scale_to_adp = False

    nqh_flips = False

    ordered_solvent = True

    place_ions = True

    number_of_macro_cycles = 10

    hydrogen_bonds = True

    use_convergence_test = True

    target = auto *ml mlhl ml_sad ls

    random_seed = 3326496

    wavelength = 0.885611

    nproc = 4

  }

  hydrogens {

    refine = individual *riding Auto

  }

  tls {

    find_automatically = False

    one_residue_one_group = False

  }

    ordered_solvent {

    mode = second_half filter_only every_macro_cycle \

           *every_macro_cycle_after_first

    h_bond_min_mac = 1

    h_bond_min_sol = 1

    h_bond_max = 6

    refine_occupancies = True

    new_solvent = isotropic *anisotropic

    b_iso_min = 0

  }

  peak_search {

    max_number_of_peaks = 1303

  }

  target_weights {

    optimize_xyz_weight = True

    optimize_adp_weight = True

    wxc_scale = 0.2

    wxu_scale = 0.2