Hi all,
I am trying to solve a protein-ligand complex structure. The resolution of the data is 2.8Ĺ. The molecular replacement solution is found in space group P21 21 21.  The current best model:

                                start         final
  ------------------------------------------------
  R-work:               0.2836        0.2484
  R-free:                0.3236        0.3066
  RMS(angles):     0.60            0.63
  RMS(bonds):      0.002          0.002

==============MolProbity validation ==============

                 Ramachandran outliers:   2.4% (Goal: < 0.2%)
                  Ramachandran favored:  89.2% (Goal: > 98%)
                      Rotamer outliers:   4.7% (Goal: 1%)
                       C-beta outliers:   0    (Goal: 0)
                            Clashscore:  10.89
                           Overall score:   2.64

Four ligands are placed for every dimer of the protein. And we are not sure whether we put the ligands right, since first it's thought that only two ligands bind every dimer of the protein.

I created a SA omit map (composite). The density for the most of the N-terminal DNA binding domain is very bad, and the density for C-terminal ligand binding domain is much better. And, I could see density for all 4 ligands. Does this necessarily mean that the model for the N-terminal DNA binding domain is not right or should I try other types of omit map (simple, refined or iterative building)? Also, I am wondering what's the difference between different types of omit map (simple, refined, iterative building) and whether it's the most reliable method to tell whether the structure model is right.

So, I am not sure whether the structure model I have is correct. It's suspected that the molecular solution is in the wrong space group or the data is twinned and only two ligands bind every dimer of the protein......But, I am not sure what I should check or what to do to improve the model. Thank you so much!

Best,
Wei